A person-centred AAC intervention to support interpersonal

interaction in persons with dementia: An exploratory study

by

Auriel Adele May

A thesis submitted in partial fulfilment of the requirements for the degree

Philosophiae Doctor
in
Augmentative and Alternative Communication

in the Centre for Augmentative and Alternative Communication

at the

FACULTY OF HUMANITIES

UNIVERSITY OF PRETORIA

SUPERVISOR: Professor Shakila Dada

CO-SUPERVISOR: Professor Janice Murray

December 2020

© University of Pretoria
© University of Pretoria
 ACKNOWLEDGEMENTS

– I can do all things through Christ who gives me strength –

(Philippians 4:13)

Against the backdrop of a global pandemic, this study encountered many uncertainties. My
resilience and tenacity to persevere was anchored in my faith, and the support received from
integral individuals and institutions. I wish to express my immeasurable gratitude to:

• Professor Shakila Dada: As a fledgling researcher, you recognised my potential. Thank

you for taking me under your wing and for generously sharing your knowledge with me.
I deeply appreciate all the opportunities that have developed my critical thinking skills. I
am honoured to have been mentored by an academic scholar of your calibre and will
always strive to emulate your outstanding work ethic and research excellence.

• Professor Janice Murray: I sincerely appreciate your brilliant co-supervision and astute
attention to detail. Your expert guidance has shaped this study in invaluable ways.

• Bevan, my husband: This dream could not have been realised without you by my side.
Your unwavering support gave me the courage to pursue my goals with conviction.
Thank you for experiencing with me every step of this journey; for consoling me
through the tears and celebrating the triumphs. Thank you for graciously listening to all
my research ideas. You are my safe place and I love you dearly.

• My parents: Thank you for your steadfast love and prayers, and for instilling in me the
value of hard work and the importance of education. Dane and Trisha, you both have
inspired me to dream bigger. Taitum, my darling niece, you are my greatest blessing and
the motivation for everything I do. You all mean the world to me.

• Mrs Pushpa Pillay: Thank you for the wholesome meals you prepared at pivotal points
of this study. Your acts of kindness are sincerely appreciated.

© University of Pretoria
• Doctor Kirsty Bastable: Thank you for the methodological assistance at various stages
of this study, and for the stimulating academic conversations over coffee.

• Professor Kerstin Tönsing, Doctor Alecia Samuels, Doctor Ensa Johnson, Professor
Juan Bornman, my PhD group and Mrs Liza Siefe: Thank you for your assistance and
support over the years. You all have made this journey incredibly memorable.

• Doctor Christina Sotiropoulou Drosopoulou, Charene van der Walt and Alex
Stipnovich: Thank you for your assistance with piloting the surveys.

• Doctor Karin Van Niekerk and Doctor Daleen Casteleijn: Your advice regarding the
research methodology is acknowledged and greatly appreciated.

• Mrs Isabel Claassen: Thank you for your meticulous language editing of this thesis.

• I am hugely appreciative of each professional participant for volunteering their time and
expertise to this study. I sincerely value your contribution.

• To all the persons with dementia whom I have met at different phases of this study,
thank you for teaching me what it means to enjoy the present moment.

• I hereby acknowledge the financial assistance of the National Institute for the
Humanities and Social Sciences (NIHSS), in collaboration with the South African
Humanities Deans Association (SAHUDA) towards this research. Opinions expressed
in this report and conclusions arrived at are those of the author and are not necessarily to
be attributed to the NIHSS and SAHUDA.

• I am grateful to the Andrew W. Mellon Foundation, through the grant-holder (Prof.

Shakila Dada), for the financial assistance towards this research project.

• I am also thankful to the University of Pretoria, for the financial support towards this
research study.

ii

© University of Pretoria
 DEDICATION

This thesis is dedicated to my grandmother, Mrs Elizabeth Reddy. Our very last moment
together was the most profound. In the absence of words, we shared an indescribable
connection communicated through eye gaze, touch and a smile. The writing of this thesis was
inspired by that moment and will forever be deeply cherished.

iii

© University of Pretoria
 ABBREVIATIONS

AAC : Augmentative and Alternative Communication

AACOM-PwD : Augmentative and Alternative Communication Outcome
Measure for Persons with Dementia
DAT : Dementia of the Alzheimer’s Type
EBP : Evidence-Based Practice
PCC : Person-Centred Care
PC-AAC : Person-Centred Augmentative and Alternative Communication
PPA : Primary Progressive Aphasia
PwD : Person with Dementia
PwPPA : Persons with Primary Progressive Aphasia
WHO : World Health Organization

iv

© University of Pretoria
 TABLE OF CONTENTS

ABBREVIATIONS ............................................................................................................................................. iv

LIST OF TABLES ............................................................................................................................................. viii

LIST OF FIGURES.............................................................................................................................................. ix

LIST OF APPENDICES ........................................................................................................................................ x

ABSTRACT ....................................................................................................................................................... xi

CHAPTER 1: PROBLEM STATEMENT AND RATIONALE ..............................................................................1

1.1. Introduction ......................................................................................................................................... 1
1.2. Problem statement and rationale........................................................................................................ 1
1.3. Terminology ......................................................................................................................................... 3
1.4. Outline of chapters .............................................................................................................................. 5
1.5. Summary.............................................................................................................................................. 7

CHAPTER 2: LITERATURE REVIEW.............................................................................................................8

2.1. Introduction ......................................................................................................................................... 8
2.2. Interventions for persons with dementia ............................................................................................. 8
2.2.1. Defining dementia .......................................................................................................................... 8
2.2.2. Impact of dementia on interpersonal interaction.......................................................................... 9
2.3. Person-centred care approach........................................................................................................... 10
2.3.1. Background of person-centred care ............................................................................................ 10
2.3.2. Life story work and person-centred care ..................................................................................... 12
2.3.3. Critical debates on person-centred care ...................................................................................... 13
2.4. Evidence-based practice in dementia interventions .......................................................................... 16
2.4.1. Evidence-based practice in health care ....................................................................................... 16
2.4.2. Evidence-based practice in augmentative and alternative communication ............................... 17
2.5. Phase 1 Research review: AAC interventions for persons with dementia .......................................... 18
2.5.1. Systematic search ......................................................................................................................... 19
2.5.2. AAC strategies and techniques .................................................................................................... 30
2.5.3. AAC outcomes for persons with dementia .................................................................................. 33
2.5.4. AAC outcome measures for persons with dementia ................................................................... 35
2.6. Secondary review: Scoping review ..................................................................................................... 37
2.7. Current Study ..................................................................................................................................... 38
2.8. Summary............................................................................................................................................ 39

CHAPTER 3: Development of a PC-AAC intervention for PWD ................................................................40

3.1. Introduction ....................................................................................................................................... 40
3.2. Phase 2: Development of a PC-AAC intervention for PwD (Overview) ............................................... 40
3.3. Component 1: Person-centred co-production process ....................................................................... 44
3.3.1. Consultation with multiple stakeholders ..................................................................................... 44
3.4. Component 2: Electronic life story conversational support ............................................................... 47
3.4.1. Customisation of the electronic conversational support ............................................................. 47

© University of Pretoria
 3.4.2. Scaffolded PC-AAC interpersonal interaction ............................................................................ 50
3.4.3. Procedural checklists ................................................................................................................... 55
3.5. Component 3: AAC outcomes for persons with dementia ................................................................. 56
3.5.1. AAC outcomes with the electronic life story conversational support ......................................... 56
3.6. Component 4: AAC outcome measure for persons with dementia .................................................... 56
3.6.1. The AACOM-PwD description .................................................................................................... 57
3.7. Summary............................................................................................................................................ 59

CHAPTER 4: METHODOLOGY .................................................................................................................60

4.1. Introduction ....................................................................................................................................... 60
4.2. Research aims .................................................................................................................................... 60
4.3. Multimethod research design ............................................................................................................ 61
4.3.1 Research phases ................................................................................................................................ 62
4.4. Ethical considerations for the overall study ....................................................................................... 64
4.5. Phase 3A: Q-methodology ................................................................................................................. 65
4.5.1 Step 1: Concourse development ................................................................................................... 66
4.5.1.1. Step 1.1: Professional participants .......................................................................................... 67
4.5.1.2. Step 1.2: Recruitment of professionals ..................................................................................... 68
4.5.1.3. Step 1.3: Description of professionals ..................................................................................... 68
4.5.1.4. Step 1.4: Concourse development procedures ......................................................................... 70
4.5.1.5. Step 1.5: Concourse development data analysis ...................................................................... 71
4.6. Step 2: Q-set (Statements) ................................................................................................................. 71
4.7. Step 3: Q-sort (ranking of statements) ............................................................................................... 74
4.8. Factor analysis ................................................................................................................................... 76
4.9. Phase 3B: Case study ......................................................................................................................... 78
4.10. Phase 3B: Participants ....................................................................................................................... 78
4.11. Materials ........................................................................................................................................... 83
4.12. Phase 3B: Procedures ........................................................................................................................ 88
4.12.1. Data collection procedures...................................................................................................... 88
4.12.2. Baseline phase ......................................................................................................................... 89
4.12.3. Intervention with PC-AAC scaffolding strategies .................................................................. 90
4.12.4. Stakeholder perspective ........................................................................................................... 90
4.13. Reliability ........................................................................................................................................... 90
4.13.1. Procedural integrity ................................................................................................................ 90
4.13.1.1. Procedural integrity of the baseline phase .............................................................................. 91
4.13.1.2. Procedural integrity in the intervention phase......................................................................... 91
4.13.2. Data reliability: Scoring the AACOM-PwD ........................................................................... 91
4.13.3. Intra-rater reliability ............................................................................................................... 91
4.13.4. Inter-rater reliability ............................................................................................................... 92
4.14. Data analysis ..................................................................................................................................... 92
4.15. Summary............................................................................................................................................ 93

CHAPTER 5: RESULTS .............................................................................................................................94

5.1. Introduction ....................................................................................................................................... 94
5.2. Terms ................................................................................................................................................. 95
5.3. Phase 3A: Q-Methodology results ..................................................................................................... 96

vi

© University of Pretoria
 5.3.1. Factor extraction.......................................................................................................................... 96
5.3.2. Factor rotations............................................................................................................................ 97
5.3.3. Characteristics of three extracted factors.................................................................................... 98
5.3.4. Factor interpretation .................................................................................................................. 100
5.3.5. Consensus statements ................................................................................................................ 111
5.4. Phase 3B: Case study results ............................................................................................................ 113
5.4.1. Reliability ................................................................................................................................... 113
5.4.1.1.1. Procedural reliability ........................................................................................................ 113
5.4.1.1.2. Procedural integrity in the baseline and intervention phase ............................................. 113
5.4.2. Data reliability: Scoring of the AAC Outcome Measure for PwD ........................................... 114
5.4.2.1.1. Intra-rater reliability ......................................................................................................... 114
5.4.2.1.2. Inter-rater reliability ......................................................................................................... 118
5.5. Stakeholder’s perspective: Participant with dementia .................................................................... 120
5.6. Overview of results: Phases 3A and 3B ............................................................................................ 121
5.7. Summary.......................................................................................................................................... 122

CHAPTER 6: DISCUSSION ...................................................................................................................... 123

6.1. Introduction ..................................................................................................................................... 123
6.1. A person-centred care approach in AAC interventions for persons with dementia ......................... 124
6.2. PC-AAC scaffolding strategies .......................................................................................................... 125
6.3. Electronic life story conversational support ..................................................................................... 127
6.4. AAC outcomes for Persons with Dementia ...................................................................................... 131
6.5. AAC outcome measurement for Persons with Dementia................................................................. 132
6.6. Reliability ......................................................................................................................................... 133
6.6.1. Procedural reliability ................................................................................................................. 133
6.6.2. Reliability of scoring the AAC Outcome Measure for PwD (AACOM-PwD) .......................... 134
6.7. Summary.......................................................................................................................................... 136

CHAPTER 7: CONCLUSION, EVALUATION AND RECOMMENDATIONS ................................................... 137

7.1. Introduction ..................................................................................................................................... 137
7.2. Summary of the findings .................................................................................................................. 137
7.3. Clinical implications of the study ..................................................................................................... 138
7.4. Evaluation of the study .................................................................................................................... 139
7.4.1. Strengths .................................................................................................................................... 139
7.4.2. Limitations ................................................................................................................................. 140
7.5. Recommendations for future research ............................................................................................ 141
7.6. Summary.......................................................................................................................................... 142

References ................................................................................................................................................... 143

vii

© University of Pretoria
 LIST OF TABLES

Table 2.1: Summary of studies on AAC interventions for persons with dementia (PwD):
Unaided AAC Systems
Table 2.2: Summary of studies on AAC interventions for PwD: Electronic AAC systems
(high technology)
Table 2.3: Summary of studies on AAC interventions for PwD: Non-electronic AAC
systems
Table 2.4: Summary of studies on AAC interventions for PwD: Combined AAC systems
(unaided, non-electronic and electronic)
Table 3.1: Involvement role and AAC principles embedded in the four components
Table 3.2: Summary of adapted interviews with multiple stakeholders
Table 3.3: Component 2: Sub-components and materials developed
Table 3.4: Evidence-based PC-AAC conversational partner scaffolding strategies
Table 3.5: Application of the PC-AAC scaffolding strategies
Table 3.6: Component 4: Sub-components and materials developed
Table 3.7: Existing outcome measures to inform development of the AACOM-PwD
Table 3.8: AACOM-PwD scoring on scoring sheet (example)
Table 4.1: Description of professionals
Table 4.2: Piloting of online Q-sort
Table 4.3: Synopsis of Q-sort data
Table 4.4: Selection criteria for case study participant
Table 4.5: Selection criteria for familiar conversational partner
Table 5.1: Factor extraction decision-making protocol
Table 5.2: Overview of seven default factors initially generated
Table 5.3: Factor rotation comparisons
Table 5.4: Characteristics of the three extracted factors
Table 5.5: Factor loadings for the three rotated factors
Table 5.6: Factor A: Description of professionals
Table 5.7: Factor A: Highest- and lowest-ranked statements
Table 5.8: Factor A: Distinguished statements
Table 5.9: Factor B: Description of professionals
Table 5.10: Factor B: Highest and lowest statements
Table 5.11: Factor B: Distinguished statements
Table 5.12: Factor C: Description of professionals
Table 5.13: Factor C: Highest- and lowest-ranked statement
Table 5.14: Factor C: Distinguished statements
Table 5.15: Factor Summary: Factor A, Factor B and Factor C
Table 5.16: Consensus statements across factors
Table 5.17: Procedural reliability percentages: Baseline and intervention phases
Table 5.18: Intra-rating reliability of scoring the AACOM-PwD: Domain level
Table 5.19: Intra-rating reliability of scoring the AACOM-PwD: Sub-domain level
Table 5.20: Inter-rater reliability of scoring the AACOM-PwD: Domain level
Table 5.21: Inter-rater reliability of scoring the AACOM-PwD: Sub-domain level
Table 5.22: Stakeholder’s perspectives: Self-reported enjoyment of person with dementia

viii

© University of Pretoria
 LIST OF FIGURES

Figure 2.1: Evidence-based AAC framework (Schlosser & Raghavendra, 2004)

Figure 2.2: Flow diagram of the screening process in reviewing literature on AAC
interventions for PwD
Figure 3.1: Overview of a PC-AAC intervention for PwD developed in this study
Figure 3.2: Organising framework of the PC-AAC intervention for PwD
Figure 3.3: AAC device (iPad) supported by a universal stand
Figure 3.4: Schematic representation of the turn taking opportunities within the life story
interpersonal interaction
Figure 4.1: Three research phases of this study
Figure 4.2: Orientation to Q-methodology employed in this study
Figure 4.3: Q-sort (pre-sorting of statements) on QMethod Software
Figure 4.4: Q-sort (final sorting of statements) on QMethod Software
Figure 4.5: Schematic representation of Phase 3B data collection procedures and data
analysis
Figure 5.1: Schematic illustration of the results chapter
Figure 6.1: Schematic illustration of the discussion chapter

ix

© University of Pretoria
 LIST OF APPENDICES

Appendix A: Co-production Process: Observation Form

Appendix B: Co-production Process: Participant’s Personal Information
Appendix C: Life Story Framework and Themes
Appendix D: Life Story Vignette
Appendix E: Electronic Life story Conversational Support on AAC Device
Appendix F1: AAC Outcome Measure for a Person with Dementia (AACOM-PwD)
Appendix F2: AACOM-PwD Operational Definitions
Appendix G1: Ethics Approval (original study)
Appendix G2: Ethics Approval (amended study)
Appendix H: Concourse Development: Letter of Information to Professionals
Appendix I: Concourse Development: Online Questionnaire for e-Delphi Panel
Appendix J: Concourse Development: Content Analysis
Appendix K1: Q-set: Statements (pre-pilot)
Appendix K2: Q-set: Statements (post-pilot)
Appendix L: Q: Sort: Letter of Information to Professionals
Appendix M: Online Q-sort
Appendix N: Permission Letter to Dementia Care Home
Appendix O: Recruitment Flyer (for persons with dementia)
Appendix P: Case study: Letter of Consent to Legal Guardian of the Participant
Appendix Q: Case study: Letter of Consent to a Familiar Conversational Partner
Appendix R: Case study: Letter of Consent to Participant with Dementia (adapted)
Appendix S: Observer to Informed Consent Form
Appendix T: On-going Informed Consent Form and Picture Booklet
Appendix U: Biographical Questionnaire: Participant with Dementia
Appendix V: Biographical Questionnaire: Familiar Conversational Partner
Appendix W: Procedural Checklist: Baseline Phase
Appendix X: Procedural Checklist: Intervention Phase
Appendix Y: Stakeholder Perspective Checklist
Appendix Z: Declaration of Originality
Appendix AA: Declaration from Language Editor

© University of Pretoria
 ABSTRACT

Background: The current mandate in dementia research is to develop interventions that are
evidence-based and person-centred. However, person-centred augmentative and alternative
communication interventions in dementia are limited. Method: A multimethod research
design was conducted in three Phases. Phase one reviewed current research evidence to
inform the development of a person-centred intervention. Phase two developed a person-
centred augmentative and alternative communication intervention package with four
components, (a) elements of person-centred care; (b) an electronic conversational support; (c)
augmentative and alternative communication outcomes; and (d) an outcome measure. Phase
three tested the feasibility of these four components by considering them in two ways (phase
3A and phase 3B). In 3A, Q-methodology was employed to obtain professionals’ viewpoints.
In 3B, a case study was used to test the feasibility of the researcher’s implementation of an
electronic conversational support with a participant with dementia and the scoring reliability
of an outcome measure developed in this study. Results: Q-methodology results revealed
professional views on (i) person-centred authorship, (ii) person-centred language style using
supportive strategies (iii) adapted participatory methods with persons with dementia in their
interventions. Case study results indicated that an electronic conversational support could be
reliably implemented. Inter-rater reliability was not at an acceptable level for the multimodal
domain of an outcome measure developed in this study. Conclusion: This study highlights
the importance of using person-centred scaffolding strategies and exploring the potential of
adapted participatory methods to develop and implement evidence-based, person-centred
interventions in persons with dementia.

Keywords:
Augmentative and Alternative Communication, co-production, dementia, evidence-based
practice, person-centred care, Q-methodology

xi

© University of Pretoria
 CHAPTER 1: PROBLEM STATEMENT AND RATIONALE

1.1. Introduction

This chapter provides the problem statement and background to this study. The
purpose of the study is stated, followed by explanations of the frequently used terminology
and a list of abbreviations. The seven chapters of this thesis are subsequently outlined.

1.2. Problem statement and rationale

Interpersonal interaction is essential for human connection. It provides a context for

developing a relationship with others by sharing stories, exchanging ideas and responding to
humour (Person & Hanssen, 2015). Persons with dementia experience significant difficulty in
maintaining interpersonal interaction with their conversational partners (Hall, Lund, Young,
Okell, & Van Steenbrugge, 2018). They have challenges with the understanding of spoken
language and retrieval of words in conversations (Hickey & Bourgeois, 2018). Confusion
with word categories is often apparent, which affects the meaning of the messages that
persons with dementia wish to express. For this reason, they have difficulty with initiating
and contributing meaningfully to conversations (Samuelsson & Hydén, 2017).
As a major neurocognitive disorder, dementia brings about progressive and permanent
changes in persons with dementia. With deterioration in verbal language, persons with
dementia withdraw from interpersonal interaction, which increases social isolation and
loneliness (Hickey & Bourgeois, 2018). Close relationships are fundamentally affected, and
over time, the relational gap between persons with dementia and those around them widens
(Webb, 2017). Conversational partners play a vital role in supporting the conversations of
persons with dementia. However, conversational partners are unaware of how to use
interaction strategies to support persons with dementia effectively (Alsawy, Mansell,
MeEvoy & Tai, 2017). As a result, they have expressed a need for personalised interventions
to enable them to facilitate meaningful interaction with persons with dementia (Wynn &
Khayum, 2015).
Dementia-related communication difficulties have real-world implications for the 50
million people living with dementia. In fact, an exponential escalation of 131,5 million is
probable by 2050 (World Health Organization [WHO], 2017). Moreover, the human
population is said to be on the verge of reaching an extraordinary ageing milestone due to
increased longevity. Accordingly, there is a notable acceleration in the number of persons

© University of Pretoria
older than 65 years, compared to the younger segment of the global populace (United
Nations, 2019). This global mega-trend has concurrently increased the risk of developing
dementia.
To address this extraordinary health phenomenon, the WHO (2017) has appealed for
evidence-based interventions that improve the wellbeing of persons with dementia. Given
this call, there is a need for interventions that can be substantiated by the best available
scientific research and that require mandatory involvement with persons with dementia and
the collaborative efforts of multiple stakeholders (WHO, 2017).
The availability of evidence-based interventions to support interpersonal interaction in
persons with dementia is limited (Morris, Horne, McEvoy, & Williams, 2017; Vasse,
Vernooij-Dassen, Spijker, Rikkert, & Koopmans, 2010). In particular, academic scholars
have raised concerns that the majority of available communication interventions for persons
with dementia are underpinned neither by theory nor a well-defined conceptual framework
(Morris et al., 2017).
The current mandate for dementia research is to develop interventions that are
evidence-based, and person-centred (Lanzi, Burshnic, & Bourgeois, 2017; Mahendra, 2019).
Person-centred care is promoted as best practice in the context of working with persons with
dementia (Brossard Saxell, Ingvert, & Lethin, 2019). In person-centred care, there is an
emphasis not only on the preferences of persons with dementia but also in encouraging their
involvement in decision-making regarding their choices. Although accepted as best practice
for persons with dementia, there appears to be uncertainty about the definition of person-
centred care and its implementation in clinical practice (Dewing & McCormack, 2017).
Person-centred care has recently been emphasised in the Augmentative and
Alternative Communication (AAC) literature. McNaughton, Light, Beukelman, Klein, Nieder
& Nazareth (2019) have recently highlighted the importance of person-centred planning.
Person-centred planning in AAC underscores the centrality of the person who uses AAC in
all phases of AAC assessment and intervention (McNaughton et al., 2019; Williams,
Krezman, & McNaughton, 2008). Furthermore, AAC scholars have called for an agenda to
expand person-centred research for adults with acquired communication disabilities (Fager,
Fried-Oken, Jakobs, & Beukelman, 2019).
The evidence on person-centred care as it applies to AAC interventions in persons
with dementia is still developing (Lanzi et al., 2017). There is research evidence related to the
use of non-electronic aids (e.g., communication cards) to enable persons with dementia to
express their preferences and choices (Bourgeois, Camp, Antenucci, & Fox, 2016). However,
2

© University of Pretoria
the evidence on how person-centred care is applied in the development of electronic
conversational aids, namely general consumer-level mobile technologies with specific
applications, with persons with dementia is scarce. This highlights the need to catalyse
research evidence on person-centred electronic AAC interventions.
Alongside the escalating dementia prevalence and the associated real-world
implications, is the critical need for AAC professionals to expand their understanding of
person-centred AAC interventions. In sum, it is incumbent on AAC researchers to develop
interventions that not only consider the best empirical research but also integrate clinical
expertise with the perspectives of relevant stakeholders (Schlosser & Raghavendra, 2004).
The novel contribution made by this study aims to respond to a pressing global need for
innovative dementia research. This study should address the existing gaps in the available
knowledge by providing evidence on a person-centred AAC intervention to support
interpersonal interaction in a person with dementia.

1.3. Terminology

The following frequently used terminology are explained below:

• Augmentative and Alternative Communication
Augmentative and Alternative Communication (AAC) refers to the scientific field of
research and clinical practice. AAC offers communication support through systems that are
either unaided or aided (American Speech-Language-Hearing Association [ASHA], 2019).
While unaided systems require no technology (e.g., gestures), aided systems comprise of
non-electronic, paper-based solutions (e.g., memory books) or electronic, highly
technologically dependent devices (e.g., computer technology) (Waller, 2019).
• AAC intervention
In this study, an AAC intervention refers to the implementation of evidence-based AAC
scaffolding techniques and strategies (e.g., aided modelling) to support auditory
comprehension of spoken messages, word finding, conversational initiation, topic
maintenance, and turn taking in a person with dementia.
• Communication supports
Communication supports refer to any strategy, tool or technique used by a conversational
partner to enable persons with neurodegenerative conditions to participate in activities.
Communication supports can range from positive attitudes of conversational partners to
policies that promote communication participation (Fried-Oken, Mooney, & Peters, 2015).

© University of Pretoria
 • Electronic life story
In this study, an electronic life story was a life story of a participant with dementia
presented on an electronic AAC device. The electronic device consisted of an Apple iPad (4th
generation) as the hardware platform using the GoTalk NOW application. The purpose of the
electronic life story was to increase enjoyment when eliciting conversations about the
participants’ personalised stories.
• Evidence-based practice
In this study, evidence-based practice is defined as “the integration of the best and current
research evidence with clinical/educational expertise and relevant stakeholder perspectives”
(Schlosser & Raghavendra, 2004, p. 3).
• Familiar conversational partner
A familiar conversational partner, in this study, refers to an adult (spouse, child, family
member, friend, or caregiver) who has background knowledge on the life history of a
participant with dementia. The role of the familiar conversational partner is to assist the
participant with dementia in selecting materials for the co-produced electronic life story
conversational support (Kindell, Burrow, Wilkinson, & Keady, 2014).
• Interpersonal interaction
In this study, the term interpersonal interaction refers to the relational aspects of
interactive communication. The focus of an interpersonal interaction is to establish or
maintain a relationship with a person with dementia. There is emphasis to create a sense of
social connection between a person with dementia and conversational partner, and to
experience a level of shared enjoyment between the conversational partners (Davidson,
Worrall, & Hickson, 2008).
• Life story
A life‐story is a biographical approach that enables a person with dementia to
communicate about their memorable past experiences and reminisce about their life history
(McKeown, Clarke, & Repper, 2006). Life‐story work can be presented in various formats,
e.g., non-electronic books with photograph collages. In this study, the researcher developed
the participant’s life story from three main themes related to life events, family and friends,
and values and beliefs (Skinner, Bonnet, Schlundt & Kalekar, 2019).
• Person-centred care
Person-centred care is a philosophical approach that emphasises that personhood is
achieved in relation to and interaction with others (Kitwood, 1997). Person-centred
4

© University of Pretoria
interventions highlight the unique life history, values, preferences and needs of persons with
dementia, which is realised in mutual collaboration and understanding with those with whom
they interact with (Savundranayagam, 2012).
• Person-centred AAC intervention for person with dementia
This study developed a Person-Centred Augmentative and Alternative Communication
(PC-AAC) intervention package to support interpersonal interaction in a person with
moderate to severe dementia. The PC-AAC intervention for a person with dementia consists
of four interlinked components. Component 1 refers to the elements of a person-centred care
approach; Component 2 relates to an electronic life story conversational support; Component
3 relates to AAC outcomes for persons with dementia; and Component 4 developed an AAC
outcome measure for persons with dementia. Each of the four components contain sub-
components, underlying processes, materials and evidence-based strategies. In this study, the
PC-AAC intervention for a person with dementia was developed, and the feasibility of the
components were tested in two ways. Firstly, the views of professionals (based on their
clinical and educational expertise) were obtained on all four components Secondly,
Component 2: electronic life story conversational support was implemented in a real-life
context with a person with dementia. The procedural reliability of implementation, as well as
Components 3 and 4 were evaluated.
• Person with dementia
An individual diagnosed with dementia, specifically a person who experiences
communication difficulties arising from a neuro-degenerative dementia (e.g., dementia of the
Alzheimer’s type). Dementia is a major neurocognitive disorder in the Diagnostic and
Statistical Manual of Mental Disorders, Fifth Edition (DSM-V) (American Psychiatric
Association [APA], 2013). This study describes one person with moderate to severe dementia
with probable dementia of the Alzheimer’s type.

1.4. Outline of chapters

This study report contains seven chapters.

• Chapter 1 provides the problem statement and background to this study. The frequently
used terminology is defined, a list of abbreviations is provided, and an outline is given
of the content of the different chapters.
• Chapter 2 offers an overview of literature related to the historical and theoretical
underpinning of person-centred care, as well as current debates on the concept of

© University of Pretoria
 person-centred care. An evidence-based practice framework is presented as the
conceptual framework for the current study. The chapter includes a research review of
AAC interventions in persons with dementia. Specifically, the research review focused
on electronic AAC strategies and techniques, AAC outcomes, AAC outcome measures
and emerging trends as they relate to person-centred AAC interventions for persons
with dementia. The chapter concludes with a summary of a secondary review on
intelligent assistive technology devices for persons with dementia to supplement the
information obtained in the research review. Both reviews of the literature were used to
inform the development of a person-centred AAC intervention package for a person
with dementia, described in Chapter 3.
• Chapter 3 describes the development of the person-centred AAC (PC-AAC)
intervention package developed from the literature reviewed in Chapter 2. The PC-
AAC intervention package is described in relation to its four main components, sub-
components, underlying processes and developed materials. A co-production
framework for developing an electronic life story conversational support (Component
2) with a person with dementia is subsequently depicted. An AAC outcome measure for
persons with dementia (Component 4) that was developed in this study to measure turn
taking within an interpersonal interaction is also presented.
• Chapter 4 outlines the research methodology employed in this study. The rationale for
selecting the research design is provided, followed by the main aim and sub-aims and
the three research phases. Phase 3 – a feasibility testing phase of the four components
of the PC-AAC intervention package developed in Chapter 3– is reported on in two
parts. Phase 3A describes Q-methodology and addresses the four steps involved in Q-
methodology data collection procedures related to concourse development, selection of
statements, ranking of statements by professionals, and factor analysis. Phase 3B
presents a case study in which two components of the PC-AAC intervention package
were implemented with a participant with moderate to severe dementia in a real-life
setting. A description of the participant and setting, the participant selection criteria, in-
person data collection procedures, recruitment materials, general procedures and ethical
considerations is presented.
• Chapter 5 presents the results of the feasibility testing phase in two parts. In Phase 3A,
AAC professional’s viewpoints are attained through in-person factor analysis using Q-
methodology data. This is followed by Phase 3B, which reports on the procedural

© University of Pretoria
 integrity of implementing Component 2: electronic life story conversational support
and the reliability of scoring the AAC outcome measure for PwD (Component 4).
Qualitative data related to the self-reported enjoyment (Component 3) of the participant
with dementia is presented. Collectively, these results represent the feasibility testing
results of the four components of the PC-AAC intervention package developed in this
study (in Chapter 3).
• Chapter 6 discusses and interprets the results of the study. In each section, the findings
of this study are compared to previous research studies. Pertinent literature is also
consulted to highlight important implications of this study.
• Chapter 7 presents a summary of the research that was conducted. This is followed by
an evaluation of the study, highlights of the clinical implications, and recommendations
for future research.

1.5. Summary

This chapter presented the problem statement and rationale for the study. The need for
evidence-based, person-centred interventions for persons with dementia, underpinned by a
theoretical or conceptual framework, was highlighted. The researcher emphasised the need to
extend the knowledge base to improve interpersonal interaction using electronic AAC
interventions within the field of AAC. The definitions of frequently used terminology were
provided, and to conclude, the seven chapters comprised in this thesis were discussed briefly.

© University of Pretoria
 CHAPTER 2: LITERATURE REVIEW

2.1. Introduction

The chapter begins with a discussion of dementia and its impact on interpersonal
interaction. This is followed by a discussion of the elements of a person-centred care
approach and the current critical debates that surround this concept of care for persons with
dementia. Next follows a discussion of evidence-based practice as applied in interventions
within the field of dementia. The evidence-based practice conceptual framework as applied in
this study is subsequently presented. The chapter concludes with an overview of the two
reviews of literature related to current AAC interventions, strategies and techniques that are
used to support communication in persons with dementia.

2.2. Interventions for persons with dementia

2.2.1. Defining dementia

Dementia is classified as a major neurocognitive disorder in the Diagnostic and

Statistical Manual of Mental Disorders, Fifth Edition (DSM-V) (American Psychiatric
Association [APA], 2013). Neurocognitive disorder was introduced as an alternative term to
the word dementia, which may carry a stigma due to its Latin translation of being ‘out of
one’s mind’. Despite the change in nomenclature, the term dementia is commonly used in
research and clinical practice, and is still widely accepted (APA, 2013).
Dementia describes a cluster of irreversible and pervasive cognitive changes that
occur in different parts of the brain. Notably, one or more cognitive domains – i.e., memory,
language, executive function, learning and social cognition – may be progressively affected
(APA, 2013). Dementia symptoms arise from various diseases affecting different parts of the
brain. Accordingly, there are various dementia subtypes such as dementia of the Alzheimer’s
type (DAT), vascular dementia, dementia with Lewy bodies, and frontotemporal dementia
(Klimova & Kuca, 2016). The most common dementia subtype, DAT, is caused by
Alzheimer’s Disease that involves a progressive build-up of neuritic plaques and
neurofibrillary tangles that degenerate different cortical areas (DeTure & Dickson, 2019).
Furthermore, vascular dementia, as the second most common dementia subtype,
results from a progressive narrowing of cerebral blood vessels. Frontotemporal lobar atrophy
manifests in semantic dementia (also referred to as fluent primary progressive aphasia).

© University of Pretoria
Dementia subtypes can frequently co-exist and consequently present with a unique set of
mixed symptoms (DeTure & Dickson, 2019; Hopper, Hickey & Bourgeois, 2018).
Irrespective of the dementia subtype, persons with dementia (PwD) experience a significant
decrease in cognitive functioning that gradually affects their independence in daily activities.
In defining dementia, is it essential to understand that the syndrome in itself is not an
inevitable outcome of growing older. Although an advancement in age is a key risk factor for
developing dementia, younger individuals are not exempt from developing dementia. This is
indicated in prevalence statistics in which persons below 65 years of age account for 9% of
young-onset dementia (WHO, 2020).
Although pharmacological intervention may assist in reducing behavioural symptoms
and delaying the progression of cognitive deterioration, there is no available pharmacological
intervention to cure dementia. This means that interventions that do not use medication but
aim to enhance the psycho-social well-being and quality of life of persons with dementia are
pivotal for enabling persons to live well with dementia symptoms.

2.2.2. Impact of dementia on interpersonal interaction

The innate human need for social connection is vital for one’s quality of life. Social
connection is a subjective feeling of relatedness to another individual that may be created
through friendship and shared humour (Perion & Steiner, 2017; Sabat & Lee, 2012). A sense
of social connection is facilitated and maintained through verbal and non-verbal
communication as persons co-construct meaning with each other within an interpersonal
interaction (Birt, Griffiths, Charlesworth, Higgs, Orrell, & Poland, 2020; Perion & Steiner,
2017). However, in persons with dementia interpersonal interactions are compromised.
Whilst the expressive language of persons with dementia may appear grammatically and
phonologically correct, difficulties with auditory comprehension of spoken language such as
narratives and complex instructions are common (Mahendra, Hickey, & Bourgeois, 2018).
Interpersonal interaction is a collaborative process. As such, a person with dementia
and a conversational partner contribute to an interpersonal interaction through a sequence of
turns that facilitates shared meaning and common ground (Alant, 2017; Clark, 1996;
Samuelsson & Hydén, 2017). Although the structural mechanism of turn taking within
interpersonal interactions is generally preserved in persons with dementia, word-finding
difficulties and longer processing time may result in difficulties with turn taking (Ekström,
Lindholm, Majlesi, & Samuelsson, 2017).

© University of Pretoria
 Furthermore, word-finding and attentional difficulties affect their ability to sustain
reciprocal responses (Bourgeois & Hickey, 2018). With a progression in communication
challenges in both comprehension and expression of language, persons with dementia may
withdraw from interpersonal interaction altogether. In this regard, Ward, Vass, Aggerwal,
Garfield and Cybyk (2008) point out that persons with severe dementia may appear
unresponsive and, in turn, their conversational partners may underestimate their need for
interpersonal interaction.
The literature reveals that conversational partners may perceive that interpersonal
interaction with persons with dementia, especially those with moderate to severe dementia, is
futile (de Medeiros, Saunders, & Sabat, 2012; McEvoy & Plant, 2014). In fact, this
perception is often shared by rehabilitation and medical professionals working with persons
with dementia. Several authors raise concern over the notion of ‘therapeutic nihilism’ as a
negative professional assumption that persons with dementia do not benefit from intervention
(Giezendanner et al., 2019; Hopper, 2003; Mahendra, Scullion, & Hamerschlag, 2011;
Mahendra, Fremont, & Dionne, 2013).
On the other hand, the perspectives of persons with dementia may differ from those of
conversational partners and professionals. Despite the irreversible cognitive-communicative
changes, persons with dementia preserve the need for interpersonal interaction and for
maintaining social connection through companionships with those around them (Alant, 2017;
Sabat & Lee, 2012; Savundranayagam, 2013). This is supported in qualitative evidence from
a systematic review of 23 studies which indicated that persons with dementia believe that
interpersonal interactions with supportive conversational partners are important to foster their
wellbeing (Patterson, Clarke, Wolverson & Moniz-Cook, 2017). In parallel with these
findings, Birt et al. (2020) synthesised 13 qualitative studies and indicated that social
connection was an important priority for persons with dementia. Such connection is achieved
through positive identity-supporting interpersonal interactions. According to Haapala, Carr
and Biggs (2019), persons with dementia self-reported that their highest priorities were those
of social inclusion and a sense of feeling connected in interpersonal interaction with others.

2.3. Person-centred care approach

2.3.1. Background of person-centred care

Person-centred care is advocated as best practice in care for persons with dementia.
Person-centred care is a philosophy of care that emphasises the remaining strengths and

10

© University of Pretoria
abilities of a person with dementia rather than to focus on the trajectory of neurodegeneration
(Kitwood, 1997). Person-centred care is founded on the humanistic theory of Carl Rogers
(1958) who focused on client-centred therapy. The work of Rogers (1958) provided a person-
centred care framework for the development of therapeutic relationships. The propositions of
Rogers’ person-centred care theoretical framework (1958) underscore the relationship that
develops between a therapist and individual receiving therapy, based on a therapists’
empathetic understanding, transparency, and perspective taking. Rogers (1958) further
argued that it was the application of these principles that would enable a client to reach their
full potential (Edgar, Wilson, & Moroney, 2020).
Kitwood (1997) built on the work of Rogers (1958) and introduced the concept of
person-centred care for persons with dementia. Traditionally, interventions for persons with
dementia were framed from a bio-medical lens, in which the trajectory of neurodegeneration
was emphasised by loss of a sense of self, decline, and deficits (Kogan, Wilbur, Mosqueda,
2016; Sabat, 2019; Terkelsen, Petersen, & Kristensen, 2020). Kitwood (1997) in opposition
to a bio-medical view, challenged this notion of dementia, and its associated assumption that
neuro-cognitive decline was solely responsible for the loss of a sense of self in persons with
dementia. Social psychological scholars assert that an interplay of neuro-cognitive and social
psychological factors influences whether personhood is maintained (Kitwood, 1997; Sabat
2019; Sabat & Harre, 1992). In this regard, Kitwood and Bredin (1992) referred to the term
‘malignant social psychology’ in which negative conversational partner behaviours such as
invalidation, patronisation, disempowerment and mockery not only lead to depersonalised
interpersonal interactions but threatens the personhood of persons with dementia (Sabat,
2019).
From a socio-psychological theoretical perspective, Kitwood (1997) proposed that person-
centred care was central to the idea of personhood and could be defined as “a standing or
status that is bestowed upon one human being, by others, in the context of relationship and
social being” (Kitwood, 1997, p. 7). While the concept of personhood is complex, Dewing
(2008b) asserts that it could be seen as the unique attributes that make up an individual.
Accordingly, person-centred care can be realised through supportive interpersonal
interactions in which conversational partners place value on the uniqueness, preferences,
choices, and autonomy of persons with dementia (Brooker, 2003; Kitwood, 1997; Fazio,
Pace, Flinner, & Kallmyer, 2018).

11

© University of Pretoria
 Furthermore, person-centred interpersonal interactions entail both verbal and non-
verbal communication supports that lead to outcomes that affirm the individuality of the
person with dementia by recognising, validating and responding empathetically to their
feelings (Alant, 2017; Lanzi et al., 2017; McEvoy & Plant, 2014). Conversely, Kitwood
(1997) warned that personhood is undermined when these characteristics are not recognised
and supported by individuals in the social environment of the person with dementia. In
particular, the use of life stories as a person-centred intervention has shown to positively
enhance the relational aspects of interpersonal interactions in persons with dementia (Astell,
Savundranayagam, Kelson, Purves, & Phinney, 2018).

2.3.2. Life story work and person-centred care

Interventions focused on the life story of persons with dementia are fundamental to
supporting person-centred care in clinical practice (Cooney & O’Shea, 2018; Doran, Noonan,
& Doody, 2018). Research shows that person-centred care can be maintained through
interpersonal interactions that are based on the personal life story, cultural background and
experiences of the person with dementia (Cooney, & O’Shea, 2018; McKeown, Clarke,
Ingelton, Ryan, & Repper, 2010; Villar, Serrat, & Bravo-Segal, 2019). Thus, reminiscence-
based interactions in the personal life story of the person with dementia and the significance
of their lived experiences come to the fore (Brooker & Latham, 2015; Dewing, 2008b). Life
story work is a type of reminiscence-based approach that focuses on enabling a person with
dementia to share their past history and maintain their personal identity in an attempt to
preserve a social connection with those around them (Cooney & O’Shea, 2018; Gridley,
Birks, & Parker, 2020; Kindell et al., 2014).
While the body of knowledge on life story work in dementia is still growing, evidence
suggests the positive outcomes of using paper-based life story books to increase the well-
being of persons with dementia by providing opportunities for reciprocity and by enhancing
communication in care interactions (Gridley, Brooks, Birks, & Baxter, 2016; Grøndahl,
Persenius, Bååth, & Helgesen, 2017; McKeown et al., 2010). According to Subramaniam and
Woods (2016), electronic life story formats, using participatory design processes, increases
enjoyment in persons with dementia, and the authors suggest that an electronic life story
format may serve as a communication aid as dementia symptoms progress. Unfortunately, the
evidence on electronic life story interventions for persons with dementia is underexplored
(Astell et al., 2018; Critten & Kucirkova, 2019; Hashim, Rias, & Kamarauzaman, 2013). A

12

© University of Pretoria
recent systematic review of 14 studies highlighted that while there may be various formats of
life stories, electronic life stories in particular are scarce (Elfrink, Zuidema, Kunz, &
Westerhof, 2018). Moreover, Gridley et al. (2020) posit that there is scant evidence on life
story work from the perspective of persons with dementia.

2.3.3. Critical debates on person-centred care

Despite being recommended as a philosophy of care for persons with dementia,

person-centred care is highly debated (Dewing, 2019; Edgar et al., 2020). There appears to be
ambiguity in its definition. Kogan et al. (2016) performed a systematic literature search of
studies related to person-centred care in older adults from 1990 to 2014. The authors found
17 different definitions of person-centred care. Importantly, Kogan et al. (2016) affirmed that
a consensus definition of person-centred care would not only improve classification of
evidence-based person-centred interventions, but also the replication thereof in clinical
practice. The literature further highlights a contention that the definition of person-centred
care is not fully understood in current practice because it is condensed to a few basic
attributes, such as being compassionate or just valuing patient preferences (Dewing &
McCormack, 2017; Edgar et al., 2020; McCance, McCormack, & Dewing, 2011;
McCormack, Dewing, & McCance, 2011). Further advancing this argument, Røsvik,
Brooker, Mjroud and Kirkevold (2013) suggest that person-centred care is still a theoretical
idea and remains elusive in clinical practice.
The need for an agreed-upon definition of person-centred care has been argued to
have presented a range of person-centred care models with limited empirical support and
value in practice (Dewing & McCormack, 2017; Pluut, 2016). Yet, at the same time, Gibson
et al. (2019) affirm that various models of person-centred care could explicate the
complexities of the concept and its application in various fields of clinical practice.
To operationalise the concept of person-centred care in clinical practice, researchers
have developed different person-centred care frameworks. Advancing the work of Kitwood
(1997), Brooker (2004) summarised four essential elements of person-centred care in her
VIPS model of person-centred care. According to Brooker (2004), the VIPS model
emphasises value for the person with dementia, individualised interventions, recognition of
the perspective of the person with dementia and the importance of the social environment
(Passalacqua & Harwood, 2012; Røsvik et al., 2013). In contrast to a focus on relational
elements, McCormack and McCance (2017) developed the Person-Centred Practice

13

© University of Pretoria
Framework in the field of nursing with five interrelated elements (e.g., macro context,
person-centred processes, pre-requisites by professionals, the care environment) that lead to
person-centred outcomes.
In the field of speech-language therapy, Hickey, Kinder, Khayum, Douglas and
Bourgeois (2018) are of the opinion that speech-language therapists should be prepared to
provide interventions that are person-centred. Yet, it cannot be assumed that speech-language
therapists are prepared or are adequately skilled to provide person-centred interventions for
persons with dementia. In a recent survey, Maltese speech-language therapists expressed a
distinct need for training in person-centred care as well as processes of how to involve
persons with dementia in decision-making (Saccasan & Scerri, 2020). In this
regard, Douglas, Brush, and Bourgeois (2018) provided practical clinical guidance for the
assessment and intervention of persons with dementia by recommending a person-centred
approach underpinned by Montesorri For Aging and Dementia.
Although various person-centred frameworks have been developed to assist in the
application of person-centred care, Swaffer (2019) asserts that person-centred care is still not
translated into clinical practice because it remains merely an afterthought to a perceived best
practice of working with persons with dementia. Such disagreement in the literature impacts
not only the way in which person-centred care is applied and delivered through interventions,
but also how research is practically conducted with persons with dementia (Edgar et al.,
2020; Hennely & O’Shea, 2017; Hung & Chaudhury, 2011).
Furthermore, there are challenges associated with the practical implementation of
person-centred care in clinical practice. Gibson et al. (2019) argue that the implementation of
person-centred care in the design and delivery of interventions is challenging amid competing
priorities of time constraints.
Several researchers have raised concern that persons with dementia are often not
involved in, consulted with or informed of decisions related to the interventions they receive
(Harding et al., 2019; Hubbard, Downs, & Tester, 2003; Lepore, Shuman, Weiner, & Gould,
2017). Accordingly, in negating the core elements of person-centred care, persons with
dementia are being divested of their identity, independence and inclusion in matters that
directly influence their lives (Bosco, Schneider, Coleston-Shields, & Orrell, 2019).
Involving persons with dementia as end users in the design and development of
interventions is positioned as a re-thinking of traditional ways of working with persons with
dementia (McLaughlin, 2020). Swaffer (2015) stresses that failure to include persons with
dementia in intervention research may not only promote stigmatisation, but may also cast
14

© University of Pretoria
doubt on the validity of research findings. Therefore, the literature shows that dementia
advocates are contending for a distinction between the perspectives of persons with dementia
and those of health professionals and caregivers (Bartlett, 2014; Bartlett, Windemuth-
Wolfson, Oliver, & Denning, 2017; Yeandle, Kröger, & Cass, 2012).
There is also an implicit expectation that researchers should involve persons with
dementia in intervention research that concerns them, rather than to rely solely on family
proxies (Keady, Hydén, Johnson, & Swarbrick, 2018). Innovation is needed to ensure that
adapted methodologies be used that intentionally involve persons with dementia in
intervention research (Hydén, Swarbrick, Johnson, & Keady, 2018). A growing body of
research illuminates the importance of conducting research with persons with dementia
instead of for them (Dewing, 2008a; Kindell, Keady, Sage & Wilkinson, 2017; Murphy &
Oliver, 2013). In the same vein, Nedlund and Bartlett (2018), as well as Hydén and Antelius
(2017) argue for the need to involve persons with dementia as collaborators in intervention
research.
The literature shows evidence of various adapted participatory research methods such
as ethnography (i.e., the scientific description of peoples and cultures with their customs and
habits) (Antelius, Kiwi, & Standroos, 2018) and walking interviews (to generate rich data and
knowledge about participants’ connections to their environments) (Kullberg & Odzakovic,
2018). By using these participatory strategies, persons with dementia and other stakeholders
are enabled to participate meaningfully in intervention research (Hydén et al., 2018). This
emphasises important current research priorities not only about how intervention research is
conducted, but also about whom it is conducted with.
Bosco et al. (2019) propose that person-centred care can be achieved through the
active involvement of persons with dementia in the production of interventions. The
involvement of these persons, together with assertions of their choices and autonomy, is in
line with social and health care movements that encourage patient and public involvement
(Alsawy, Tai, McEvoy, & Mansell, 2019; Terkelsen et al., 2020). Participatory research
methodologies such as the co-production of interventions may be seen as an empowerment
process that re-conceptualises persons with dementia to be active contributors to
interventions that are developed for their use (McConnell et al., 2019; Realpe & Wallace,
2010). Co-production is one methodology recognised by the World Health Organization as a
strategy aimed at achieving Integrated People-Centred Health Services (2015) in which
researchers and relevant stakeholders work together to achieve person-centred outcomes
(Kickbusch & Gleicher, 2012; WHO, 2015). Central to the idea of co-producing interventions
15

© University of Pretoria
is the notion that persons with dementia, their family and relevant stakeholders are
experience-based experts who can contribute much knowledge and expertise to complement
the knowledge and expertise of researchers (Boivin et al., 2018; Davies et al., 2019).
Despite the advantages of participatory methods such as co-production, persons with
dementia are still perceived as passive recipients within person-centred care interventions
(Bosco et al., 2019). The literature highlights that the preferences concerning a fully informed
client is an important component of evidence-based practice (Dodd, 2007; Dollaghan, 2004).
By not involving persons with dementia in the process of developing interventions through
the use of participatory methods, hampers the advancement of the evidence-base (Hydén et
al., 2018; Roulstone, 2015).

2.4. Evidence-based practice in dementia interventions

2.4.1. Evidence-based practice in health care

Evidence-based practice is endorsed as the recommended framework on which all

healthcare professionals should base their clinical decision making (Dodd, 2007). Evidence-
based medicine (EBM) was formally introduced to the field of clinical medicine as the
“conscientious, explicit, and judicious use of current best evidence in making decisions about
the care of individual patients” (Sackett, Rosenberg, Gray, Haynes, & Richardson, 1996,
p. 71). EBM was expanded and widely adopted as evidence-based practice (EBP) in an
attempt to enhance professional practices across various disciplines. Dollaghan (2007, p. 2)
offered an expanded interpretation of evidence-based practice as an integration of (a) “best
available external evidence from systematic research, (b) best available evidence internal to
clinical practice, and (c) best available evidence concerning the preferences of a fully
informed patient”.
Recent literature shows a trend of heightened interest in evidence-based interventions
for persons with dementia across various fields, namely nursing (Skela-Savič et al., 2020),
occupational therapy (Smith & D’Amico, 2019), psychological care (Teri, Logson, McCurry,
Pike, & McGough, 2020) and speech-language therapy (Ruggero, Croot, & Nickels, 2020).
Since evidence-based practice is applied to various professional disciplines and can be
defined differently by professionals (Rousseau & Guina, 2015), it is essential for the purposes
of this study to understand evidence-based practice as applied in the field of Augmentative
and Alternative Communication (AAC).

16

© University of Pretoria
2.4.2. Evidence-based practice in augmentative and alternative communication

Augmentative and alternative communication (AAC) is a scientific field of evidence-

based research and clinical practice that aims to maintain the quality of life, participation and
engagement of persons with dementia by using a variety of compensatory strategies,
techniques and devices (American Speech-Language-Hearing Association [ASHA], 2019).
The literature shows that AAC researchers have encouraged the use of evidence-based
practice in AAC interventions to not only show accountability in intervention design, but also
to promote the best intervention outcomes for all AAC users (Granlund & Olsson, 1999;
Schlosser, 2006). In line with the framework of EBM (Sackett et al., 1996) and evidence-
based practice suggested by Dollaghan (2007), Schlosser and Raghavendra (2004)
emphasised the importance of evidence-based decision making in the field of AAC. The
triadic integration of best and current research evidence with clinical expertise and the
perspectives of relevant stakeholders are the three evidence-based practice cornerstones
required in decision making for a specific direct stakeholder (Schlosser & Raghavendra,
2004). An evidence-based practice framework (Schlosser & Raghavendra, 2004) as applied
in the current study is shown in Figure 2.1.

Perspective of a person with dementia:

• Phase 2 (co-produced electronic conversational support)
• Phase 3B (case study, implement electronic
conversational support, self-reported enjoyment)

Best and current research EBP in AAC

evidence: (applied to Professional’s clinical and
persons with educational expertise:
• Phase 1 (research review
and scoping review of dementia) • Phase 3A (Q-
the literature) methodology)

Figure 2.1: Evidence-based practice in AAC (Schlosser & Raghavendra, 2004)

Evidence-based practice has been regarded as a disciplined approach towards making

clinical decisions that lead to the best desired outcomes for all persons with health conditions
(Rousseau & Guina, 2016). Importantly, the integration of all three cornerstones is important
in professional decision making for an individual person with dementia (Dollaghan, 2007;
Schlosser & Sigafoos, 2009). The rationale for incorporating the perspectives of a person

17

© University of Pretoria
with dementia is not only in line with current discourses in the literature about including the
voices of persons with dementia, but also ensures that the outcomes of AAC interventions are
meaningful (Alant, 2017; Johnson, 2020). Involving stakeholders at various stages in the
development of AAC systems is of vital importance (Waller, Balandin, O’Mara, & Judson,
2005). Since persons with dementia as direct stakeholders are affected by all decision making
regarding their use of AAC systems, their opinions should be highly valued (Schlosser &
Raghavendra, 2004). The clinical and educational experience of professionals are pivotal to
validate different components of AAC interventions (Schlosser, 1999). AAC professionals
are obligated to apply reliable and rigorous procedures in the implementation of their
interventions (Gast & Ledford, 2018). The use of reliable measurement tools is a critical
component of demonstrating the effect of AAC interventions in persons with dementia
(Ledford, Dane, & Gast, 2018). For the advancement of evidence-based practice, research
evidence demonstrating the real-life outcomes that AAC interventions have on interpersonal
interaction in persons with dementia, is essential (Gast & Ledford, 2018; Smith & Murray,
2011; Spriggs, Lane & Gast, 2018).
Previous experimental research evidence that emerged from systematic reviews of
literature constitutes an important cornerstone of evidence-based practice and shows the
effect of AAC interventions (Roulstone, 2015; Schlosser & Raghavendra, 2004). However,
Smith (2016) points out that not all AAC research focuses on the experimental effects of an
intervention – it also focuses on the views and life experiences of persons who use AAC.

2.5. Phase 1 Research review: AAC interventions for persons with dementia

Synthesised summaries of research evidence are essential to assist AAC professionals in

evidence-based decision making (Grant & Booth, 2009; Schlosser & Sigafoos, 2008). Whilst
previous reviews offer insights into the benefit of AAC interventions for persons with
dementia, AAC studies are scattered across various reviews and limited due to the focus of
each enquiry. For example, Swan, Hopper, Wenke, Jackson, Till and Conway (2018)
evaluated the evidence for speech-language interventions in persons with moderate to severe
dementia and retrieved two AAC interventions that improved conversation. Morello et al.
(2017) reviewed studies on language and communication interventions for persons with
Alzheimer’s disease and found two studies that used memory cards in conversation. It is clear
that a more systematic approach is required to synthesise the current knowledge base of AAC
interventions for persons with dementia.

18

© University of Pretoria
2.5.1. Systematic search

May, Dada and Murray (2019) conducted a research review to identify and describe the
current evidence-base pertaining to AAC interventions for persons with dementia. A research
review method (Grant & Booth, 2009) was employed by May et al. (2019) as Phase 1 of the
current study. The aim of the research review was to synthesise the AAC research related to
AAC systems and strategies, AAC outcomes and AAC outcome measures for persons with
dementia. A research review employs a systematic approach to the identification and synthesis
of studies but, it does not include an appraisal of the literature as completed in a systematic
review (Grant & Booth, 2009). The research review in Phase 1 of the current study used the
following search terms:
“dementia” or Alzheimer* “AND” “augmentative and alternative communication” or
“AAC” or augment* or “communication support” OR communication aid* OR
"communication system*" OR “speech generating device*” OR “voice output
communication aid*” OR gesture* OR sign* OR “graphic symbol” OR total
communication OR Vocal* "AND" communicat* or “interaction” or conversat*
A multifaceted search strategy was adopted. The nine electronic databases that were
searched for published studies and dissertations, were Cumulative Nursing and Allied
Health Literature (CINAHL), PsycINFO, PsycARTICLES, Academic Search Complete,
MEDLINE, Linguistics and Language Behaviour Abstracts (LLBA), ProQuest Dissertations
and Theses Global, Scopus, and IEEE Xplore digital library. The database searches were
restricted to temporal (1990-2018), linguistic (English) and source type (academic journals
and dissertations) limiters. Furthermore, the journals of Augmentative and Alternative
Communication and Communication Disorders were hand searched (1999-2018).
Additional searches included an ancestral search of studies that met the inclusion criteria,
and forward citations of included studies on Google Scholar.
The electronic studies were exported to Covidence, an online software application that
facilitated the management of the review process (Veritas Health Innovation, n.d.).
Pre-defined study inclusion criteria were agreed upon by all three reviewers (i.e.,
authors of the review), and were based on the following four eligibility criteria:
i. Report on adult participants, 18 years and older, with a degenerative dementia, e.g.,
Alzheimer’s dementia, vascular dementia, dementia with Lewy bodies, frontotemporal
dementia (primary progressive aphasia (PPA)/semantic dementia). People with PPA
(without dementia) prior to 2013 were also included (APA, 2013).

19

© University of Pretoria
 ii. Primary research on AAC strategies, techniques or technologies to support receptive or
expressive language and memory for interaction purposes.
iii. Utilising a quantitative, qualitative or mixed-method design (including single case
studies).
iv. Study published in an academic journal, or a master’s thesis or doctoral dissertation.
Studies were excluded if they reported on persons with mild cognitive impairment and
if data could be not isolated for persons with dementia.
Once the studies were exported to Covidence, the first two reviewers (i.e., authors)
blind-reviewed each study at title and abstract level, and at full-text. The two reviewers
applied the pre-defined eligibility criteria for each study, and a ‘yes’, ‘no’ or ‘maybe’
response was selected at title and abstract level in Covidence. A study was excluded, if both
reviewers selected ‘no’, and were included at full-text if a reviewer selected ‘yes’, or
‘maybe’. Furthermore, at full-text, each reviewer selected a reason from a list of drop-down
options in Covidence to support the exclusion of a study.
The two reviewers followed a consensus-building process during screening
disagreement. Subsequent to reading each study, detailed data was extracted independently
by the first author. Fifty-two percent of the extracted data was independently checked by the
remaining authors as per the following parameters: participants, research design, data
collection method, setting, instructional format and administration of AAC technologies,
description of AAC components, outcome measures, communication outcomes, reported
AAC benefit, and key findings. As per the aim of the review, the included studies were
further synthesised into four main categories: (i) AAC systems and strategies, (ii) AAC
outcomes and (iii) AAC outcome measures for persons with dementia.
A flow chart of the study selection process is illustrated in Figure 2.2.

20

© University of Pretoria
Figure 2.2: Flow diagram of the screening process in reviewing literature on AAC
interventions for persons with dementia

As illustrated in Figure 2.2, altogether 613 studies were identified. At full text, the
majority of the studies were excluded on account of the following: not having a
communication or interaction outcome (n=22); not relating to AAC (n=9); incorrect
publication type (n=8); unavailable online (n=5); did not focus on persons with dementia
(n=4); incorrect study design (n=4); duplicated copy of study (n=2); foreign language (n=1).
A corpus of 39 studies that met the inclusion criteria is summarised in Tables 2.1, 2.2, 2.3
and 2.4.

21

© University of Pretoria
 Table 2.1: Summary of studies on AAC interventions for persons with dementia (PwD): Unaided AAC Systems
Participant diagnosis,
Author(s), Design severity, age, AAC Study focus AAC outcome AAC outcomes
year, conversational partner description measures
country (CP)
Ellis & Single- DAT, severe (n=5) Eye gaze, Non-verbal Standardised Increased reciprocity when
Astell subject 77-89 years gestures, communication 1) Direct observation of participants’ communication
(2017) design CP: Researcher vocalisations, repertoires behaviour behaviours were imitated by the
UK facial expressions researcher. Increased enjoyment
Non-standardised and laughter in the interaction.
2) Communicative
behaviours coded and
counted
Hydén Case study DAT, severe (n=1) Vocalisations, Non-verbal Non-standardised Participant attempted to initiate
(2011) 85 years body movements, vocalisations in social 1) Frequency and types interaction using non-verbal
Sweden CP: Familiar CP (n=2) gaze direction interaction of non-verbal cues and vocalisations.
vocalisations identified

Kindell et Case study Semantic dementia, mild Enactment using Everyday Non-standardised Participant used enactment
al. (2013) (n=1) body posture, conversations 1) Conversational strategy to contribute to
UK 71 years pointing, facial patterns observed conversations. Reciprocal
CP: Various (n=3) expressions laughter within dyad.

22

© University of Pretoria
 Table 2.2: Summary of studies on AAC interventions for PwD: Electronic AAC systems (high technology)
Authors, year, Design Participant diagnosis, AAC Study focus AAC outcome AAC outcomes
severity, age, description
country measure
conversational
partner (CP)
Aitken (2015) Single-subject DAT, VD, mild to Digital Quality and Non-standardised No improvement in on-topic
New Zealand design moderate (n=4) memory book quantity of 1) Conversational statements and utterances.
61-88 years conversations utterances and AAC facilitated easier
CP: Family member (n=4) statements coded and conversation between the CPs
and researcher counted and PwD.
2) Family
questionnaire
Alm et al. (2004) Quantitative Dementia*, moderate (n=9) CIRCA Interest and Non-standardised Increased choice of
UK group 65-95 years involvement of 1) Evaluation conversational topics for the
comparisons CP: Caregiver/ care staff the PwD in questionnaires PwD.
(n=9) interaction Enjoyment of interaction.
Astell et al. Observational DAT, mixed severity CIRCA Nature of dyadic Non-standardised PwD showed independence in
(2010) study (n=11) interaction 1) Checklist of choosing conversational topics.
UK (qualitative) 65-95 years interactional activity Increased laughter within the
CP: Caregiver/ care staff dyad.
(n=11)
Davis & Shenk QUANT-qual DAT, moderate (n=10) Multimedia Engagement that Standardised More comments and smiles
(2015) age unspecified videos promotes talking 1) Observational with personal videos.
USA CP: Researcher (n=8) Measure of Greater diversity of language
Engagement with generic videos.
Non-standardised
2) Language patterns
identified from
conversation analysis
Dynes (2018) Within- DAT, mild-moderate (n=7) Electronic Person-Centred Non-standardised Interactions became more
Canada participants, 52-86 years conversational Communication 1) Utterances coded as person-centred and enjoyable.
prospective CP: Family member (n=7) memory aid (PCC) per adapted PCC CPs supported the preferences
design coding chart of PwD.
Ekström et al. Case study DAT, severity unspecified Digital Communication Non-standardised Increased conversational length
(2017) (n=1) communication characteristics 1) New conversational and time spent on talking about
Sweden 52 years book domain or conversation the device.

23

© University of Pretoria
Authors, year, Design Participant diagnosis, AAC Study focus AAC outcome AAC outcomes
severity, age, description
country measure
conversational
partner (CP)
CP: Family member (n=1) initiation identified and AAC did not generate new
counted topics within the interaction.
2) Conversational
length recorded
Fried-Oken et al. Quantitative DAT, moderate (n=30) Digitised voice Conversational Non-standardised Paucity of language, fewer
(2009) group 50-94 years output behaviours with 1) Utterances counted elaborations, overall quantity of
USA comparisons CP: Researcher digitised voice and coded utterances reduced.
output
Fried-Oken et al. Quantitative, DAT, moderate (n=30) Pilot 1: Conversational Non-standardised No improvement in quantity
(2012) group pilot 1, Digitised voice performance Conversations coded and type of utterances.
USA comparisons CP: Researcher output (with and using a social Voice output reduced
without AAC communication conversational performance.
aid, and voice framework
output)
Hamel et al. Mixed- DAT, severity unspecified Mobile Feasibility and Non-standardised Device was a focal point to
(2016) methods (n=18) reminiscence utility of a 1) Checklists share experiences and enhance
USA design Mean age 84 years aid reminiscence aid 2) Themes from semi- conversations.
CP: Familiar CP (n=14) in interaction structured interviews Increased enjoyment of
and direct observation interaction.
Mooney et al. Single-subject PPA, severity unspecified Mobile Lexical retrieval Non-standardised Improved lexical retrieval skills
(2018b) design (n=6) technology skills during 1) Number of target and conversational confidence
USA 62-80 years activity retell in words recorded and (PwD).
CP: Familiar CP (n=6) conversation counted CPs scaffolded conversations.
2) Direct user feedback
Purves et al. Observational Dementia*, moderate (n=3) CIRCA Regionally Non-standardised AAC used to initiate and
(2015) study 81-90 years adapted 1) Interactions maintain topics.
Canada (qualitative) CP: care staff (n=1) programme for transcribed Companionable silences.
dyadic interaction

24

© University of Pretoria
 Table 2.3: Summary of studies on AAC interventions for PwD: Non-electronic AAC systems
Participant diagnosis,
Authors, Design severity, age, AAC Study focus AAC outcome AAC outcomes
year, country conversational partner description measures
(CP)
Andrews-Salvia Single-subject DAT, Dementia*, severe Memory book On-topic facts in Non-standardised Increased on-topic facts
et al. (2003) design (n=4) severe dementia 1) On-topic facts coded evident in all participants.
USA 90-94 years and counted.
CP: Researcher
Bourgeois Single-subject DAT, moderate (n=3) Communication Quality of Non-standardised Improved factual statements
(1990) design 59-66 years wallet conversational 1) Utterances coded and and fewer ambiguous
USA CP: Familiar CP (n=3) content counted utterances.
2) Satisfaction Rating No changes in conversational
Form behaviours noted by CPs.
Bourgeois Single-subject DAT, moderate-severe Memory Conversational Non-standardised Some participants increased
(1993) design (n=6) wallet/book content and social 1) Utterances and social on-topic statements,
USA 74-88 years skills of dyad behaviours (turn taking) elaborations and turn taking.
CP: Another PwD (n=6, i.e., coded and counted CPs reduced ambiguous
3 dyads) 2) Social validation utterances.
rating scale
Bourgeois & Single-subject DAT, mixed severity (n=4) Memory wallet Conversational Non-standardised The PwD increased appropriate
Mason (1996) design 74-80 years content (PwD), 1) Utterances coded and statements, decreased repetitive
USA CP: Volunteer staff (n=3) conversational counted statements.
behaviours (CP) 2) Satisfaction rating Conversational behaviour of
CP improved.
Bourgeois et al. Within-subjects Dementia*, moderate- VoiceMyChoice™ Preference and Non-standardised The PwD were able to
(2016) design severe (n=37) choice making in 1) Preference communicate preferences.
USA 67-96 years interaction Assessment Nursing aides’ understanding
CP: Nursing aide (n=33) Questionnaire (PAQ) of the preferences of PwD
improved.
Bourgeois et al. Quantitative Dementia*, moderate Memory book Quality and Non-standardised Increased duration of speaking
(2001) group (n=66) quantity of 1) Utterances and time, number of utterances, and
USA comparisons Mean 85 years naturalistic statements coded and conversational topics.
CP: Nursing aide (n=66) interaction counted CP reduced number of
2) Duration of prompts.
verbalisations, memory
book use

25

© University of Pretoria
 Participant diagnosis,
Authors, Design severity, age, AAC Study focus AAC outcome AAC outcomes
year, country conversational partner description measures
(CP)
Bourgeois et al. Quantitative Dementia*, moderate Memory book Communication Non-standardised Low frequency of memory
(2004) (Pretest – post- (n=125) skills training 1) Frequency of memory book use during care
USA test) 75-86 years programme book use by nursing interactions.
CP: Nursing aide (n=126) aides calculated

Chang (2011) Single-subject Dementia*, mixed severity Memory book Quality and Non-standardised Increased on-topic statements
USA design (n=3) quantity of 1) Utterances coded and of facts, decreased ambiguous,
82-88 years conversations counted unintelligible, and
CP: Researcher perseverative utterances.
Chang (2015) Within-subjects DAT, Dementia* mild- Decisional Decisional Non-standardised Participants increased
USA design moderate (n=20) (visual) aid capacity 1) Adapted decisional understanding of treatment
76-97 years capacity recording form options.
CP: Researcher 2) Quality of vignette Improved clarity of statements.
statement form
3) Social validation
rating scale
Fried-Oken et Quantitative, DAT, moderate Pilot 2: Topical Non-standardised AAC priming with spaced
al. (2012) group (n=11) pilot 2 Communication vocabulary and 1) Utterances coded for retrieval training increased
USA comparisons 50-94 years board (without references to topical vocabulary. references to device and the
(Pilot study) CP: Researcher voice output) AAC 2) Number and number of target words used.
percentage of target
words, references to
AAC device calculated.
Fried-Oken et QUAL-quant DAT, PPA, mild-moderate Communication Conversational Non-standardised Selected conversational topics
al. (2015) (n=109 board topics selected by 1) Structured related to life experiences and
USA Mean 75 years PwD conversations with PwD. personal narratives.
CP: Familiar CP (n=109), 2) Topics thematically
researcher coded
Gomès-Taibo et Single-subject DAT, mixed severity (n=3) Memory book Quantity and Non-standardised Increased self-identity
al. (2014) design 86-87 years quality of 1) Utterances coded and statements, reduced ambiguous
Spain CP: Researcher conversational counted statements, improved
content, and skills conversations quality.

26

© University of Pretoria
 Participant diagnosis,
Authors, Design severity, age, AAC Study focus AAC outcome AAC outcomes
year, country conversational partner description measures
(CP)
Hoerster et al. Single-subject DAT, VD, severe (n=4) Memory book Conversational Non-standardised The PwD increased their
(2001) design 83-90 years content (PwD), 1) Utterances coded and factual statements. Nursing
USA CP: Nursing aide (n=4) communication counted aides’ communicative
behaviours (CP) 2) Social validation behaviour improved post-
rating scale training.
3) Post-interview
questions
Johnson (2003) Single-subject Dementia*, mild-moderate Sensory cues Quality and Non-standardised No increase in the quantity or
USA design (n=5) quantity of 1) Utterances coded and quality of conversations.
73-88 years conversations counted
CP: Nursing aide 2) Social validation
(n=1) rating scale
McPherson et Single-subject DAT, VD, severe (n=5) Memory aids Quality of Non-standardised Some participants spent a
al. (2001) design 73-90 years conversations 1) Topic-related and non- higher percentage of time on
USA CP: Caregiver/care staff topic-related topic-related utterances.
(n=5) conversation coded and
calculated
Murphy et al. Quantitative Dementia*, mixed severity Talking Mats™ Expression of Non-standardised Increased on-task behaviours,
(2010) group (n=31) views related to 1) Effectiveness involvement and
UK comparisons 54-90 years well-being framework of Functional conversational length.
CP: Researcher Communication Perseverative behaviours
2) Percentage of ‘on- decreased.
task’ behaviour
3) Perseverations
4) Interview time
Murphy & QUALI-quant Dementia*, mixed severity Talking Mats™ Discussions on Non-standardised The PwD increased their
Oliver (2013) (n=18) managing 1) Conversations involvement in decision
UK 60-86 years activities of daily thematically coded making.
CP: Family member (n=18) living 2) Involvement Measure The dyad felt satisfied with the
3) Satisfaction discussion.
questionnaire
Reitz & Cross-over DAT, mild-moderate (n-6) Talking Mats™ Shared decisions Standardised The PwD increased
Dalemans design 84-90 years (Dutch version) and language use 1) The OPTION Scale involvement in decision
(2016) CP: Family member (n=6) making.

27

© University of Pretoria
 Participant diagnosis,
Authors, Design severity, age, AAC Study focus AAC outcome AAC outcomes
year, country conversational partner description measures
(CP)
Netherlands Non-standardised CP understood the PwD.
2) Utterances coded and No improvement in language
counted use.
Ruiz (2015) Pretest – post- DAT, moderate Memory book Quantity of Non-standardised The PwD increased on-topic
Puerto Rico test (n=1), 75 years conversational 1) Utterances and use of responses.
CP: Family member (n=1), content grammar coded and AAC did not improve the use
researcher counted of grammar.
Spilkin & Case study DAT, moderate to severe Memory book Quality of Non-standardised The CP scaffolded the
Bethlehem (n=1) interaction 1) Interaction structure interaction.
(2003) 85 years structure coded (topic The PwD improved topic
South Africa CP: Family member (n=1) management, repair) maintenance, decreased topic
2) CP quality of perseveration.
interaction rating scale

28

© University of Pretoria
 Table 2.4: Summary of studies on AAC interventions for PwD: Combined AAC systems (unaided, non-electronic and electronic)

Participant diagnosis,
Authors, Design severity, age, AAC Study focus AAC outcome AAC outcomes
year, conversational description measures
country partner (CP)
Broughton et Pretest – Caregiver/care staff (n=52) Unaided + non- Memory and Non-standardised Staff’s knowledge of
al. (2011) post-test electronic AAC communication 1) Knowledge of support communication strategies
Australia strategies strategies test improved.
2) Post-training survey Staff self-reported greater respect
and empathy for the PwD.
Cress & Case PPA, severity unspecified Unaided + non- Cued Non-standardised Cued comprehension ineffective
King (1999) study (n=2) electronic AAC comprehension and 1) Comprehension of with unfamiliar listeners.
USA 59-60 years augmented symbols tallied by CP Familiar CPs increased success in
CP: Family member (n=4) expression cueing new topics.

Mooney et Pretest – PPA, severity unspecified Unaided, non- Multimodal Non-standardised Dyads learned to match AAC
al. (2018a) post-test (n=5) electronic + communicative 1) Modes of strategy to their communication
USA 63-73 years electronic AAC interactions Communication Survey needs.
CP: Familiar CP (n=6)

Trahan et al. Single- Dementia, vascular Unaided + non- Picture-based Non-standardised Participants learned to exchange a
(2014) subject dementia, mild-moderate electronic AAC communication 1) Frequency of picture card for a highly preferred
USA design (n=3) responses, skill independent card activity.
85-87 years maintenance exchanges and vocal
CP: Researcher responses

Wong et al. Case Semantic dementia, mild Unaided + non- Communicative Non-standardised Participants showed improvement
(2009) study (n=1) electronic AAC effectiveness 1) Modified in expressing opinions.
USA 61 years communication framework The CP scaffolded interactions with
CP: Family member (n=1) 2) Codified ideas verbal and nonverbal support.

29

© University of Pretoria
 As reflected in Tables 2.1, 2.2, 2.3 and 2.4, studies included had been published in the
past 28 years, between 1990 and 2018. The majority (n=20) were conducted in the USA,
seven in the UK, three in Canada, two in Sweden and one each in Australia, New Zealand,
Spain, the Netherlands, Puerto Rico and South Africa. Of the 39 studies included, 33 were
published as journal articles, four master’s theses and two doctoral dissertations. The
majority of the studies (n=23) sampled participants with DAT, dementia of an unspecified
subtype (n=13) and vascular dementia (n=4). Persons with frontotemporal dementia, i.e.,
PPA or semantic dementia, were researched in only five studies, while some samples
included persons with two dementia subtypes (DAT and vascular dementia). Persons with
Lewy body dementia were not encountered in any of the studies. The information obtained
from the 39 studies are summarised in the next section to highlight trends and gaps related to
AAC strategies and techniques, AAC outcomes and AAC outcome measurement.

2.5.2. AAC strategies and techniques

AAC intervention research for persons with dementia began in 1990 (Bourgeois, 1990;
Bourgeois, 1993; Bourgeois & Mason, 1996). In a series of intervention studies using single-
subject experimental designs, early research for persons with dementia documented the
benefit of using non-electronic AAC systems (e.g., memory books, communication passports
and communication wallets) with generic or personalised photographs, and autobiographical
sentences (Andrews-Salvia, Roy, & Cameron, 2003; Bourgeois, 1990, Bourgeois 1993,
Bourgeois & Mason 1996, Bourgeois, Dijkstra, Burgio, & Allen-Burge, 2001).
From an interactional perspective, seminal AAC research focused on supporting
conversational topics, and improving the quantity and quality of utterances in persons with
mild to moderate dementia. For instance, researchers demonstrated the use of paper-based
memory wallets in increasing turn taking, on-topic statements, topic elaborations and
reducing off-topic statements in persons with dementia (Andrews-Salvia et al., 2003;
Bourgeois, 1990; Bourgeois, 1993; Bourgeois & Mason, 1996; Cress & King, 1999).
Cress and King (1999) used two case studies as a research method to emphasise the
benefit of multimodal AAC strategies. For example, augmented input and the use of facial
expressions and gestures within natural communicative contexts were used to support persons
with primary progressive aphasia (semantic dementia) (Cress & King, 1999). With a growing
evidence-base, research began to demonstrate the benefit of non-electronic AAC systems for
persons with severe dementia. In a pilot study with a single participant, researchers
highlighted that a person with severe dementia increased the proportion of time spent on on-

30
© University of Pretoria
topic utterances when supported with a simple, paper-based memory aid (McPherson,
Furniss, Sdogati, Cesaroni, Tataglini, & Lindesey, 2001). In a related study, Hoester, Hickey
and Bourgeois (2001) demonstrated that the equity of conversational turn taking increased
between a person with milder dementia and a conversational partner when using a memory
aid, however this effect was not evident in persons with more advanced dementia.
Interest in electronic AAC systems using technologically dependent devices has grown
steadily in view of a global research agenda for innovative social and health technologies to
support persons with dementia (Baxter, Enderby, Evans, & Judge, 2012; Bourgeois, Fried-
Oken, & Rowland, 2010; WHO, 2017). The reviewed literature shows that research into
electronic AAC systems for persons with dementia began in 2004 and has increased slowly,
albeit incrementally since 2010.
With the advent of technology and mobile computing, researchers from the USA
(Davis & Shenk, 2015; Hamel, Sims, Klassen, Havey, & Gaugler, 2016; Fried-Oken et al.,
2009; Fried-Oken et al., 2012; Mooney, Bedrick, Noethe, Spaulding, & Fried-Oken, 2018b),
Canada (Crete-Nishihata et al., 2012; Dynes, 2018; Purves, Phinney, Hulko, Puurveen, &
Astell, 2015), Sweden (Ekström, Ferm, & Samuelsson, 2017), New Zealand (Aitken, 2015)
and the United Kingdom (Alm et al., 2004; Astell et al., 2010) have offered preliminary
evidence on the use of electronic AAC system options to support interpersonal interactions in
persons with dementia. Given that approximately 60% of persons with dementia live in non-
Westernised low- and middle-income countries (WHO, 2018), the complete lack of research
on electronic AAC systems for persons with dementia living in these countries is alarming.
Electronic AAC systems with computer-based memory aids and mobile technologies
appear to contribute to enriched interpersonal interaction for persons with dementia (Alm et
al., 2004). Research has demonstrated multimedia communication tools support conversation
and relationships in persons with Alzheimer’s disease (Astell et al., 2010). Despite noting
possible pitfalls of electronic AAC support in their study, Ekström et al. (2017) found that a
personalised AAC application had the potential to encourage interpersonal interaction with a
person with dementia.
There appears to be an interest in exploring AAC technology with natural language
processing abilities as noted in a study by Mooney et al. (2018b). These authors used a
specific AAC research application, CoChat, by employing features of natural language
processing (NLP), just-in-time principles and social media. Multimedia digital life history
interventions consisting of past memories and wearable computing technology (SenseCam)
(Crete-Nishihata et al., 2012) are beginning to be explored by researchers (Davis & Shenk,

31
© University of Pretoria
2015). Empirical data shows that reminiscence-based activities with multimedia content are
not only enjoyable activities for persons with dementia but also enhance interpersonal
interactions between conversational partners (Alm et al., 2004; Astell et al., 2010; Hamel et
al., 2016; Purves et al., 2015). The development of electronic life stories can also be useful to
enhance person-centred care with the addition of recorded voice and digital photographs
(Astell et al., 2018).
However, in the reviewed studies, the use of voice output was found to have a negative
effect on interpersonal interaction in persons with dementia. This was evident in two pilot
studies on digitised voice output that had been embedded in customised communication
boards and in which a label was spoken out each time a person with DAT touched a picture
(Fried-Oken et al., 2009; Fried-Oken et al., 2012).
AAC interventions for persons with dementia suggest an interest in person-centred
care. Non-electronic aids enable persons with dementia to express their opinions in
conversations about their wellbeing by using the Talking Mats™ communication framework
(i.e., line drawings on a visual scale) (Murphy, Gray, Van Achterberg, Wyke, & Cox, 2010;
Murphy & Oliver, 2013; Reitz & Dalemans, 2016). Likewise, AAC interventions with non-
electronic communication cards (i.e., VoiceMyChoice™) (Bourgeois, Camp, Antenucci, &
Fox, 2016) and decisional aids (Chang, 2015) assist persons with dementia to voice their
preferences and choices in decision-making conversations.
Whilst the implementation effect of AAC interventions is important, the process of
developing interventions with persons with dementia as co-producers is also essential
(Swarbrick, 2015). Crete-Nishihata et al. (2012) supported a person with dementia in
authoring his own life story intervention with an electronic aid. The positive person-centred
outcomes of this study included affirmation of self-identity, independence in sharing stories
and enjoyment in aided-conversations (Crete-Nishihata et al., 2012). The direct inclusion of
persons with dementia in selecting conversational topics and authoring their own life stories
represents a shift towards a person-centred approach within AAC-supported interactions.
Person-centred communication is used by a conversational partner through statements
that recognise and affirm the identity of a person with dementia (Alsawy et al., 2019;
Savundranayagam, Basque, & Johnson, 2020; Savundranayagam & Moore-Nielsen, 2015;
Savundranayagam, Sibalija, & Scotchmer, 2016). Emphasis is thus placed on validating the
feelings of a person with dementia (Williams et al., 2017). Alant (2017) refers to empathetic
listening in which a conversational partner listens and responds with the intent to understand
the person with dementia at a feeling level. In so doing, a conversational partner focuses on

32
© University of Pretoria
building a sense of connection within the interpersonal interaction in collaboration with a
person with dementia (Kindell et al., 2017).
In line with this view, Dynes (2018) specifically trained family members to use
person-centred communication to validate the feelings of persons with dementia and to
negotiate with them during interactions with an electronic aid. As an outcome of the
intentional focus on person-centred communication, family members perceived
interpersonal interactions to be more meaningful. Similarly, albeit from a different
perspective, Broughton et al. (2011) found that nurses’ empathy and validation of
personhood improved AAC-supported care interactions. Purves et al. (2015) also adapted
computer-based multimedia to be culturally, linguistically or socially relevant to persons
with dementia who are from Chinese, Hispanic or multicultural backgrounds.
Interpersonal interaction can be supported by using the unique set of non-verbal
communicative behaviours of persons with severe DAT and semantic dementia. These
included imitated communication behaviours termed ‘adaptive interaction’ (e.g., eye gaze)
(Ellis & Astell, 2017); non-verbal vocalisations and non-verbal cues (e.g., eye rubbing)
(Hydèn, 2011); and ‘enactment’ as a compensatory strategy that includes direct reported
speech with body posture, pointing and facial expressions (Kindell, Sage, Keady &
Wilkinson, 2013).

Finally, persons with dementia may benefit from multimodal AAC interventions
(Fried-Oken et al., 2015). The benefit of multimodal AAC strategies was evident in the
reviewed studies that focused on the use of unaided AAC and non-electronic aids (Broughton
et al., 2011; Cress & King, 1999; Trahan, Donaldson, McNabney, & Kahng, 2014; Wong,
Anand, Chapman, Rackley, & Zientz, 2009). Furthermore, the literature indicates that the
combined use of various non-electronic aids (PPA wallet cards), and high-technology
(mobile technology) as well as AAC strategies (augmented input through keywords and/or
written choice) support conversations in persons with PPA (Mooney, Beale & Fried-Oken,
2018a).

2.5.3. AAC outcomes for persons with dementia

Companionship and connection with others are unmet needs for persons with
dementia (Hancock, Woods, Challis, & Orrell, 2006; van der Roest et al., 2009). Within the
reviewed studies, outcomes that enhanced companionship within a dyad were expressed as
laughter, smiles, enjoyment of interaction (Alm et al., 2004; Astell et al., 2010; Hamel et al.,

33
© University of Pretoria
2016; Purves et al., 2015; Wong et al., 2009), feelings of social closeness, increased
engagement (e.g., eye contact) (Davis & Shenk, 2015) or an expressed desire to interact with
others (e.g., imitation of communication behaviours) (Ellis & Astell, 2017; Hydén, 2011;
Kindell et al., 2013).
The self-reported outcomes from the perspective of persons with dementia provides
important evidence on the effectiveness of AAC interventions (Broomfield, Haarop, Judge,
Jones, & Sage, 2019; Cohen & Hula, 2020). Within the reviewed AAC interventions, these
were noted in the self-reported increase in confidence of PwPPA when interacting with others
(Mooney et al., 2018a; Mooney et al., 2018b), and in the affirmation of self-identity in a
person with dementia (Gomèz-Taibo, Amado, Dominguez, Iglesias, & Real, 2014).
The majority of outcomes of AAC interventions focused on language-based outcomes
related to the quantity and quality of conversational content (e.g., increased on-topic
statements, and decreased ambiguous, unintelligible and repetitive utterances) (Andrews-
Salvia et al., 2003; Chang, 2011; Gomèz-Taibo et al., 2014; McPherson et al., 2001; Ruiz,
2015). The quantities of utterances, topic initiations and elaborations were reduced when a
digital memory book (Aitken, 2015) or communication board programmed with an embedded
voice output (Fried-Oken et al., 2009) was used in conversations with persons with mild to
moderate DAT. Conversely, PwPPA increased their lexical retrieval skills when using mobile
technology (Mooney et al., 2018b) and persons with DAT increased their conversational time
when interacting with a digital communication or memory book (Dynes, 2018; Ekström et
al., 2017). While most of the conversational time focused on the usage of the digital
communication book or its content, there was no evidence that the electronic aid supported
the initiation of new conversational topics (Ekström et al., 2017).
Within the context of training conversational partners, outcomes related to balanced
turn taking, increased duration of speaking time, and improved facilitative behaviours such as
acknowledgments by communication partners have been noted (Bourgeois, 1990; Bourgeois
& Mason, 1996). However, these outcomes were not consistently observed as functional
communication changes based on judgements of individuals unfamiliar to the intervention
(Bourgeois, 1990; Bourgeois, 1993; Bourgeois & Mason, 1996; Chang, 2015; Hoerster et al.,
2001; Johnson, 2003).
In a pilot study reported by Fried-Oken et al. (2012), persons with DAT underwent a
training procedure prior to interacting with a communication board, and as a result, increased
references to the aid and a greater use of targeted vocabulary were evident.

34
© University of Pretoria
 AAC outcomes have focused on interactive coaching and modelling of
communication strategies or use of AAC techniques with conversational partners.
Communication outcomes included improved caregiver topic elaborations (which in turn
improved topic maintenance and decreased perseverations by the person with dementia)
(Spilken & Bethlehem, 2003) and conversational partners’ improved use of cued
comprehension strategies for PwPPA (Cress & King, 1999, Mooney et al., 2018a).

2.5.4. AAC outcome measures for persons with dementia

Smith and Murray (2011) emphasise that, in line with EBP, AAC professionals should
have outcome measures to demonstrate the effectiveness of AAC interventions. Developing
and validating outcome measurement tools have been noted in AAC for persons with
aphasia (Brock, Koul, Corwin, & Schlosser, 2019) and persons with mild neurocognitive
disorders (Lanzi, Bourgeois, & Dedrick, 2020).
Within the reviewed studies, researchers developed non-standardised outcome
measures for their specific intervention studies. These included a Preference Assessment
Questionnaire (PAQ) (Bourgeois et al., 2016); evaluation questionnaires; interaction
checklists; and questions for family interviews (Alm et al., 2004; Astell et al., 2010; Crete-
Nishihata et al., 2012; Hamel et al., 2016). An Involvement Measure (Murphy & Oliver,
2013) and a decisional capacity questionnaire (Chang, 2015) were developed by adapting
questions from other tools to measure the decision-making skills of persons with dementia
in interactions.
As part of training programmes, non-standardised tests and pre-post training surveys
or questionnaires were used to measure nurses’ knowledge of memory and communication
support strategies (Broughton et al., 2011), as well as the use of multimodal communication
strategies by PwPPA and their communication partners (Mooney et al., 2018a).
Three standardised outcome measures that were developed included the following three
instruments: Observational Measure of Engagement (Cohen-Mansfield et al., 2009) to
measure engagement that promoted talking in a persons with dementia (Davis & Shenk,
2015); Direct Observation of Behaviour (Bowie & Mountain, 1993) to measure nonverbal
communicative behaviours of persons with dementia (Ellis & Astell, 2017); and the OPTION
Scale (Elway, 2003) to measure shared decision-making skills in persons with dementia and
their communication partners when using a communication aid (Reitz & Dalemans, 2016).
The majority of AAC intervention studies for persons with dementia relied on
standardised, quantitative instruments to measure conversational content (e.g., ambiguous

35
© University of Pretoria
utterances, on-topic statements, repetitions, target vocabulary) that had been coded
according to pre-determined criteria and counted (Aitken, 2015; Andrews-Salvia et al.,
2003; Bourgeois et al., 2001; Chang, 2011; Dynes, 2018; Fried-Oken et al., 2009; Fried-
Oken et al., 2012; Gomèz-Taibo et al., 2014; McPherson et al., 2001, Mooney et al., 2018b;
Reitz & Dalemans, 2016; Ruiz, 2015). The specific aspects of conversational content that
were measured depended on the communication focus of each study.
Granlund and Blackstone (1999) proposed five necessary domains of an AAC
outcome measure, of which one was a measure of satisfaction. Within the reviewed studies,
there was limited evidence that the satisfaction of a person with dementia was measured as
an outcome. Some researchers used standardised, quantitative measurements, but included a
subjective social validation procedure or satisfaction rating scale by means of which persons
unfamiliar with the intervention detected functional changes in the targeted communicative
behaviours (Bourgeois, 1990; Bourgeois, 1993; Bourgeois & Mason, 1996; Chang, 2015;
Hoerster et al., 2001; Johnson, 2003). While these were standardised measures within
specific studies, their clinical utility was limited to the scope of the respective studies.
Although the majority of aforementioned studies employed quantitative measures,
conversational analysis as a qualitative research method provided salient details of AAC
supported interactions. For instance, Spilken and Bethlehem (2003) used conversational
analysis to analyse turn taking, repair strategies, and interaction structure (e.g., topic
maintenance) to improve a conversational partners’ use of a memory aid with a person with
dementia. Hydén (2011) utilised conversational analysis to identify the frequency and types
of verbal and non-verbal vocalisations in persons with severe dementia. By analysing the
transcriptions of verbal utterances and vocalisations, Hydén (2011) raised the importance of
responding to the vocalisations of persons with severe dementia as meaningful interactional
attempts. Kindell et al. (2013) examined the interactional strengths of a person with semantic
dementia by transcribing and analysing naturally occurring interactional data. Accordingly,
the authors identified nuanced detail of the use of enactment as an adaptive interaction
strategy in a person with semantic dementia. Likewise, through analysis of interaction,
researchers examined the ‘communication initiatives’ (i.e., introducing a new topic without a
partner prompt or topic initiation after a lapse of silence) and conversational length of a
person with dementia who used an electronic device in an interaction with a conversational
partner (Ekström et al., 2017).
Lastly, modified or adapted functional communication frameworks were used to code
interactions according to utterance types, communication functions or conversational

36
© University of Pretoria
behaviours (e.g., engagement during interviews) (Fried-Oken et al., 2012; Murphy et al.,
2010; Wong et al., 2009).
The use of outcome measures with established reliability or validity data appears to be
more favourable in clinical practice compared to tools that require subjective judgements
(Zaga, Cigognini, Vogel, & Berney, 2020). Evidence from the reviewed literature reveals
that there are currently no available psychometrically validated AAC outcome measures to
measure interpersonal interaction in persons with dementia.

2.6. Secondary review: Scoping review

A scoping review was conducted in an independent study by Dada, Hyman, May and
Murray (2020). A summary of the findings is described in this study as it supplements the
information on electronic AAC systems (described in Par 2.5.2). The scoping review
followed scoping review methodology (Arksey & O’Malley, 2005) to map the currently
available research on intelligent artificial assistive technology devices (IATDs).
The aims of the scoping review were as follows: Firstly, to provide an overview of the
literature on IATDs. Secondly, to report on the benefits and limitations of IATDs, and
thirdly, to identify trends and gaps in the literature. The scoping review applied a
multifaceted search strategy; and used databases and search terms similar to those employed
by May et al. (2019). However, Dada et al. (2020) included additional terms specifically
related to “assistive technology”, “assistive device”, and “intelligent”. Following screening
procedures, the included studies (n=105) were described according to (a) the types of IATDS,
(b) the cognitive and communicative domains targeted and (c) their commercial availability.
The results showed that the majority of research on IATDs, currently remain at the
conceptual stage of prototype development (n = 41). Importantly, persons with dementia
were not involved in the development of technologies designed for their use. The majority of
reviewed studies using artificial intelligence entail companion robots (n=63) to support
memory and attention difficulties in persons with dementia (Takayanagi, Kirita, & Shibata,
2014). Studies (n=33) on social robots, to a lesser extent, could interpret and respond to the
utterances and facial expressions of persons with dementia to reinforce turn taking (Hendrix,
Feng, van Otterdijk, & Barakova, 2019; Moyle et al., 2017, 2015).
Most research using IATDs focus on the memory and attention difficulties in persons
with dementia. AAC scholars have advised that innovative technology with artificial
intelligence can benefit individuals who use AAC (Light et al., 2019). The results of this

37
© University of Pretoria
scoping review highlight that intelligent assistive technology devices have the potential to
respond to a broad range of communication needs and capabilities of persons with dementia.
Therefore, the need for AAC interventions to consider the role of intelligent assistive
technology in persons with dementia is pertinent.

2.7. Current Study

From the reviewed AAC literature in the previous sections (Par 2.5, and 2.6), it is
evident that important research findings have developed the evidence-base for persons with
dementia. It is clear that research has shown the evolving nature of AAC interventions from
seminal studies using non-electronic aids, to the current potential of artificial intelligence to
support communication in persons with dementia. However, in light of the contemporary
discourses related to person-centred care, it appears that person-centred AAC research for
persons with dementia is limited.
Although persons with dementia have been involved in different aspects of designing
their AAC interventions (e.g., selecting materials or topics of conversation), this has been
performed in a limited way. AAC researchers have not fully explored the range of adapted
participatory methods that could be employed to include persons with dementia in developing
their own AAC interventions. Given the rise in the prevalence of dementia, there is potential
for professionals to consider innovative methods to develop AAC interventions with persons
with dementia. However, there are limited guidelines for AAC professionals on how to apply
such adapted participatory methods in research.
Due to the critical debates related to person-centred care in the literature, there is a need
to understand the views of professionals toward this evolving concept and its
operationalisation in AAC research and practice. Furthermore, there is a paucity of research
guidance in the literature on what elements of person-centred care could be included in AAC
interventions for persons with dementia. More specifically, how AAC strategies could be
integrated within a person-centred care approach to support interpersonal interaction in
persons with dementia, is currently underexplored.
While quantitative experimental designs were necessary to demonstrate the effect of
interventions in persons with dementia, the use of multiple methods or mixed-method
research designs are underexplored in the literature. In this regard, studies that utilise a
multiple method research design have the potential to enhance the robustness of AAC
findings for persons with dementia.

38
© University of Pretoria
 Lastly, multimodal AAC outcome measures with established or preliminary
psychometric properties to measure interpersonal interaction in persons with dementia was
not located in the reviewed literature. Based on the aforementioned limitations highlighted in
the reviewed AAC literature, the current study aimed to develop and test the feasibility of a
person-centred AAC intervention for a person with dementia. The intervention was
developed by employing adapted participatory methods and was underpinned by a person-
centred care approach comprising four components extrapolated from the literature. The
feasibility of implementing the intervention was tested by obtaining the views of
professionals on its four components. Additionally, the person-centred AAC intervention was
implemented with a person with dementia and an outcome measure developed in the study
was assessed for scoring reliability.

2.8. Summary

This chapter discussed the available literature as it pertains to person-centred care in

supporting interpersonal interaction in persons with dementia. Although emphasised in the
AAC literature, person-centred care and its operationalisation in developing electronic AAC
interventions with persons with dementia are only emerging. Given the impetus of
participatory research, the current literature points to important gaps that are potentially
unexplored and that should be taken into consideration in developing person-centred,
evidence-based interventions for persons with dementia. The reviewed research evidence
outlined in this chapter informed the development of a person-centred AAC intervention
described in the next chapter.

39
© University of Pretoria
 CHAPTER 3: DEVELOPMENT OF A PC-AAC INTERVENTION FOR PWD

3.1. Introduction

This chapter provides an outline of a PC-AAC intervention for PwD developed in this
study. First, an overview is presented of the four components that comprise the PC-AAC
intervention for PwD. Next, each component is described individually in relation to its sub-
components, underlying processes, and materials. The first and second components are
described as an electronic life story conversational support co-produced with a person with
moderate to severe dementia. Linked to the two preceding components, the third and fourth
components relate to AAC outcomes and an AAC outcome measure for PwD, after which the
chapter is concluded.

3.2. Phase 2: Development of a PC-AAC intervention for PwD (Overview)

The PC-AAC intervention for PwD consisted of four interlinked components. As

depicted in Figure 3.1, these four interlinked components integrated concepts and processes
related to a person-centred care approach; electronic life story conversational support; AAC
outcomes for PwD; and an AAC outcome measure for PwD.

Four main components of a PC-AAC intervention for PwD

1. 2. 4.
3.
Person- Electronic AAC
AAC
centred life story outcome
conversational outcomes measure for
care
support for PwD PwD
approach

Figure 3.1: Overview of a PC-AAC intervention for PwD developed in this study

The four components were identified as existing gaps or unexplored areas within the
existing body of AAC research for persons with dementia (see Par. 2.5 and Par. 2.6). Each
component was guided by six key AAC principles for research and intervention
recommended by Blackstone, Williams and Wilkins (2007) and was tailored for a specific

40
© University of Pretoria
participant in Phase 3B of the main study, i.e., a person with moderate to severe dementia
(see Par. 4.10.3.1). In adherence to the ethical protocol of this study (see Par. 4.4), a
pseudonym – Mrs Anna Brown – was used to refer to the participant to protect her
anonymity. Aligned with the principle of the co-production of research, the participant was
involved in different roles as listener, co-thinker, decision maker and indirect advisor. These
roles were adapted from an Involvement Matrix by Smits, Meeteren, Klem, Alsem and
Ketelaar (2020) and a definition of the different roles and AAC principles (Blackstone et al.,
2007) as embedded in each of the four components is summarised in Table 3.1.

Table 3.1: Involvement role and AAC principles embedded in the PC-AAC intervention
for PwD

Component Involvement (role) AAC principles

(Smits et al., 2020) (Blackstone et al., 2007)
Component 1 Listener: The person with dementia • Principle 1: Adapted methodologies
PCC listens and is present at meeting with enable a person with dementia and
participatory family. stakeholders to participate in AAC
processes intervention research.

Component 2 Co-thinker: The person with • Principle 2: Theoretical constructs and

Electronic dementia is asked for their opinion the selection of interpersonal
life story on their life story, and their choice of interaction strategies are based on
conversational AAC symbols. empirical evidence.
support Decision maker: The person with • Principle 3: AAC device features are
dementia is involved in decision selected in accordance with the
making related to the customisation strengths and capabilities of a person
of electronic life story conversational with dementia.
support. • Principle 4: on the role of
conversational partners in scaffolding
an interpersonal interaction with a
person with dementia is emphasised.

Component 3 Indirect advisor: Self-reported • Principle 5: Conversational supports

AAC outcomes meaningful outcomes from offer a person with dementia
qualitative studies are reviewed. opportunities to participate in
interpersonal interaction and to
Co-thinker: The person with maintain social connection with their
dementia is asked for their opinion conversational partners.
and feedback on the AAC outcomes
(enjoyment).
Component 4 • Principle 6: Meaningful AAC
AAC outcome outcomes are measured from the
measure perspective of a person with dementia.

All four components, including the roles played by the participant and AAC
principles were subsequently incorporated into an organising framework to enable the

41
© University of Pretoria
researcher to operationalise all four components in clinical practice in a systematic and
principled manner. The organising framework is illustrated in Figure 3.2.

42
© University of Pretoria
 Component 1: Person-centred participatory design process
Co-production process with a person with moderate to severe dementia

Sub-component 1.4.
Refinement

Sub-component 1.1.
Sub-component 1.3. Consult with stakeholders:
Clinical reflection Sub-component 2.1.
• Naturalistic observations
Electronic life story conversational support:
Adapted interviews with: Listener
Customisation procedures:
• Family member and PwD
Sub-component 1.2. • AAC symbol selection Co-thinker
Participant feedback • Familiar conversational partner
• Life story choices
• Health and social care staff
• Choice of voice output
• Choice of legacy message Decision
• Choice of personalised music maker
• Seating and positioning

Indirect
Sub-component 2.2. Advisor Component 4
Researcher’s Component 3
expertise: Scaffolded PC-AAC interpersonal AAC outcome measure
AAC outcomes for PwD
Material interaction for PwD
Co-thinker
development
*Feasibility testing
(Phase 3 of main study)

Figure 3.2: Organising framework of the PC-AAC intervention for PwD

43
© University of Pretoria
3.3. Component 1: Person-centred co-production process

As shown in Figure 3.2, Component 1 was an iterative co-production process that consisted
of four sub-components: consultation with multiple stakeholders; participant feedback;
clinical reflection; and refinement of the electronic life story conversational support. Before
observations were made and the person-centred co-production process began, ethics approval
was obtained from the Research Ethics Committee at the Faculty of Humanities, University
of Pretoria (Appendix G1). The participant and all stakeholders were made aware that their
participation in the research study was voluntary (McMillan & Schumacher, 2014). Should
they have wished to withdraw from the study at any time, they were permitted to do so
without penalty. The researcher ensured that the participant with dementia did not undergo
any form of harm or undue distress and was treated with respect throughout the research
study.
The personal information of the participant with dementia as well as all information
obtained from the familiar communication partner, family member and other stakeholders
were kept strictly confidential following the University of Pretoria’s ethical research policy
on data management. The participant’s personalised life story content on the AAC device
was used only for the purpose of the study by the researcher. After the completion of data
collection, the participant with dementia and her family member were given a printed version
of the digital life story as a memento. Full details of the ethics approval and informed consent
procedures are further elaborated on in Par. 4.4 and Par 4.10.1 respectively.
The section below presents the procedures related to the consultation with multiple
stakeholders and clinical reflection. Sub-components related to participant feedback,
customisation and refinement of the electronic life story conversational support (Component
2.1), although interlinked with Component 1, are described in Par. 3.4.1.

3.3.1. Consultation with multiple stakeholders

3.3.3.1. Observations in everyday contexts
Naturalistic observations were performed to obtain an understanding of the
interpersonal interaction of the participant in daily activities with different conversational
partners at the care home. Observations were performed by means of unobtrusive and
‘participant’ observations. During unobtrusive observations, the researcher observed the
participant from afar (Atelius et al., 2018). The participants’ preferences for interpersonal

44
© University of Pretoria
interaction at specific times of the day, and the level of support provided by various
conversational partners were captured on an observation form (Appendix A) developed by
the researcher. ‘Participant’ observations were performed not only to obtain observational
information, but also to establish rapport with the participant prior to the co-production
process (Atelius et al., 2018). To do so, the researcher participated in the daily routine of the
participant. This entailed joining the weekly music group, arts and crafts activities, and sitting
in the dining area during mealtimes with the participant. Subsequent to the ‘participant’
observations, the researcher recorded information on the observation form (Appendix A)
related to the participants’ preferences, her frequently used phrases, and the activities she
particularly enjoyed. Overall, observations took place at a minimum of three times per week
over a period of two weeks.

3.3.3.2. Adapted stakeholder interviews

The researcher interviewed multiple stakeholders to gain different perspectives on the
life story, preferences and interests of the participant. The four stakeholder groups consisted
of the participant’s closest family member, her familiar conversational partner, nursing staff,
her occupational therapist, and her music therapist. The researcher was sensitive to the time
constraints and work priorities of each stakeholder, and therefore the setup of the interviews
was adapted. The walking interview was employed as a participatory research method as an
alternative to a structured interview for persons with dementia (Kullberg & Odzakovic,
2018). This method was also adapted to involve nursing staff without imposing the need for a
traditional sit-down interview. All information obtained from the adapted stakeholder
interviews was documented on an observation form (see Appendix A). A summary of the
materials, procedures and outcomes of the adapted interviews appears in Table 3.2.

45
© University of Pretoria
 Table 3.2: Summary of adapted interviews with multiple stakeholders

Stakeholder Data collection Materials Procedure Outcome

Family member Face-to-face Participants’ A personal information form was The family member completed and returned the
Participant informal family emailed to the family member in personal information sheet and brought along the
Familiar discussion in a photographs; advance. Via email, a meeting time participant’s family photographs. Additional
conversational social setting personal was arranged in a relaxed social supplemental life story information (e.g., scrapbooks
partner information setting (coffee shop) at the research and family biographies written in the participants own
forms site. This was attended by the words) was supplied. The family member provided an
(Appendix B) family member who granted in-depth background of the family’s religious beliefs
consent, the participant herself and and Jewish traditions. The participant listened and
her familiar conversational partner. contributed to the discussion in a reminiscing way.
Participant Walking Observation The familiar conversational partner One walking interview took place and unfolded as a
interview form wheeled the participant in her naturally occurring conversation. The researcher
(Appendix A) wheelchair through the garden of followed up on conversations that the participant
the care home. Natural directed about trees and flowers, and her family
conversations arose with the stories. The familiar conversational partner contributed
researcher. to these conversations.

Nursing staff Walking Observation Nursing staff were informally Nursing staff provided information on the participant’s
interview form asked questions while ‘walking and personal fashion style, and personal grooming choices.
(Appendix A) talking’.

Occupational Informal Observation The occupational therapist was The occupational therapist provided information about
therapist interview form informally interviewed at the the participant’s favourite topics and personal
(Appendix A) research site. preferences (e.g., food, traditions).
Music therapist Informal Observation The music therapist was informally The music therapist provided information on music
interview form interviewed following a music that the participant enjoyed and suggested music to be
(Appendix A) group session. included in the life story conversational support.

46
© University of Pretoria
3.3.3.3. Clinical reflection
Subsequent to each observation, the researcher engaged in intentional reflection by
writing in a reflective journal to reflect on the lessons learned about the participant (May,
Dada, & Murray, 2020). This provided a channel for the researcher to evaluate some of her
own assumptions held in relation to cultural differences and the strength of persons with
dementia in general (Mahendra et al., 2013).

3.4. Component 2: Electronic life story conversational support

The electronic life story conversational support consisted of three sub-components,

namely customisation of the electronic conversational support, PC-AAC scaffolding
strategies, and procedural implementation. The sub-components and materials are reflected in
Table 3.3.

Table 3.3: Component 2: Sub-components and materials developed

Component 2 Sub-components Materials developed

Electronic life Sub-component 2.1: Life story content
story Customisation procedures • Two personalised life stories
conversational Customised electronic AAC system choices:
support • AAC symbol selection
• Voice output
• Legacy message
• Personalised music
• Seating and positioning
AAC device and familiarisation:
• iPad
• GoTalk Now application
Sub-component 2.2: PC-AAC scaffolding strategies:
Scaffolded PC-AAC interpersonal • Aided modelling
interaction • Prosodic emphasis
• Attentional cues
• Expectant delays
• Person-centred communication
PC-AAC interpersonal interaction script
Sub-component 2.3: Procedural • Procedural checklists
implementation

3.4.1. Customisation of the electronic conversational support

3.4.1.1. Life story content
Life story work supports person-centred care as it aims to value the unique identity
and personal life history of persons with dementia (McKeown et al., 2010). The researcher

47
© University of Pretoria
collated all information that emerged from the stakeholder interviews (see Par. 3.3.1),
including all the photographs supplied by the family member (see Table 3.2). The
photographs were then collaboratively selected by the participant and her familiar
conversational partner. With each photograph selected, the familiar conversational partner
assisted by facilitating a reminiscing conversation with the participant. This provided the
researcher with an opportunity to record supplemental information for the life story. This
process took place during two separate sessions lasting approximately 25 minutes each.
Next, the researcher wrote two life stories by integrating seven to ten facts related to
the photographs and information extracted from family scrapbooks or family biographies.
The researcher put together two draft life stories, based on preferred topics stated by the
family member and the information gathered in consultation with multiple stakeholders (see
Par. 3.3.1).
The life story was structured with a life story title, based on each life story topic. An
integrative review of 13 studies on life story work in older persons by Doran et al. (2018)
revealed that there is no prescribed approach to developing a life story. Based on the
information collated from stakeholders in Par. 3.3.1, the researcher arranged the participant’s
information into a framework of three main life story themes related to the milestones and
highlights of the participant’s life (Skinner et al., 2019). The researcher developed the
participant’s life story from three main themes related to life events, family and friends, and
values and beliefs (Skinner et al., 2019) (Appendix C).
From these life story themes; two life stories were written. Each life story consisted of
six to ten short and simple sentences. These sentences highlighted happy moments,
milestones or memorable events in the participant’s life and were associated with a highly
salient context-relevant photograph. A vignette of one life story is presented in Appendix D,
and screenshots of the life story on the electronic aid are found in Appendix E. Generic
pictures from Google Images that have personal relevance to the participant were also used to
supplement the life story where family pictures were unavailable.
3.4.1.2. Customised electronic AAC system choices
The participant played the role of decision maker in selecting the AAC symbols
(photographs selected in Par. 3.4.1.1) that she preferred to be included in her life story. To
facilitate conversations related to the customisation of the electronic life story conversational
support, the researcher supported the conversations using AAC strategies (e.g., pictures
cards, aided modelling and pointing). The researcher offered the participant choices
supported by pictures related to her preference for colour or black-and-white photographs, the

48
© University of Pretoria
use of voice output in the form of a legacy message and music, and if voice output was
preferred, then the type of legacy messages and type of music. Legacy messages are
personalised words or phrases that are considered to be a trademark of an individual recorded
in their own unique voice (Costello, 2016).
The participant independently participated in decision making with regard to her
electronic life story conversational support by using a combination of pointing and verbal
utterances. She indicated her preferences and chose colour photographs, preferred voice
output using her own messages/phrases, and she selected classical music. Due to difficulties
in executive functioning, the participant experienced difficulty in spontaneously recording
her own legacy message. Therefore, the researcher audio-recorded naturally occurring
conversations with the participant using the audio-recording function on an Apple iPhone 6.
This enabled audio extraction of specific words and phrases to be utilised later as voice
output on the participant’s AAC device.
3.4.1.3. AAC device and familiarisation
The AAC device consisted of an Apple iPad (4th generation) as the hardware platform
using the GoTalk NOW application (app). GoTalk NOW is an AAC app for persons with
communication difficulties that can be customised with sound, internet images, personal
photographs and video clips (Attainment Company Inc., 2020). The GoTalk NOW app was
selected for its accessibility and affordability for research with persons with dementia in the
South African context. The Apple iPad was selected as the family member reported to have
previously used the device for playing memory games and scrolling through digital family
photographs with the participant.
A universal, adjustable table computer stand was used to support the AAC device to
increase the participant’s comfort and ease of viewing the AAC intervention content (see
Figure 3.3). The participant’s selected photographs were scanned using a colour scanning
device and then they were uploaded electronically to the AAC device. The researcher
included a familiarisation procedure of the AAC device when the PC-AAC scaffolding
strategies were applied (see Par. 3.4.2.2. and Table 3.5). The familiarisation entailed the
researcher demonstrating three navigational points on the AAC device: light, single touch
with pad of finger on the AAC device; press the Home button, and press the GoTalk NOW
app icon.

49
© University of Pretoria
Figure 3.3: AAC device (iPad) supported by a universal stand

Customisation of the electronic life story conversational support overlapped with

Component 1 as stated in Figure 3.2. The process was iterative and dependent on the
participant’s feedback. Following a first draft of the life story (see Par. 3.4.1.1) on the AAC
device, the researcher trialled the electronic life story conversational support with the
participant. This created an opportunity for the researcher to make refinements based on the
participant’s feedback and preferences. Although music was played through the in-built
speakers of the AAC device, a portable Bluetooth speaker was also connected to the AAC
device to increase the audibility of voice output. The volume was adjusted to suit the
participant’s hearing comfort. The participant expressed a preference to remain seated in her
wheelchair seat, instead of a regular seat. This was noted for subsequent sessions to increase
her comfort during the use of the electronic life story conversational support.
Previous studies showed that visual scenes resulted in an increased number of
conversational turns for persons with aphasia (Brock, Koul, Corwin & Schlosser, 2017).
Therefore, a visual scene in this study included a static but highly contextual, personal
photograph or a personally relevant generic photograph to discuss the characters within the
life story context. In certain instances, the visual scene was embedded with music or voice
output (in the form of pre-recorded voice messages) (see Par. 3.4.1.2).

3.4.2. Scaffolded PC-AAC interpersonal interaction

3.4.2.1. Identification of PC-AAC scaffolding strategies
A life story interpersonal interaction was scaffolded by the researcher with application
of PC-AAC scaffolding strategies. Theoretically, this was underpinned by Bruner’s (1978)
scaffolding theory which asserts the role of graded support by an adult to assist a child
(novice) to solve a problem on their own (Bakhurst & Shanker, 2001). Similarly, within the
context of an electronic life story support with the participant, the role of the conversational
partner (researcher) was to scaffold the interpersonal interaction using person-centred AAC
strategies (i.e., PC-AAC scaffolding strategies).

50
© University of Pretoria
 The literature demonstrates the positive impact of aided modelling (as an augmented
input strategy) to enhance the ability of a person using AAC to understand verbal messages
(Dada, Stockley, Wallace, & Koul, 2019; Dada, Flores, Bastable, & Schlosser, 2020; Dietz,
Knollman-Porter, Toth, & Brown, 2014; Wallace, Dietz, Hux, & Weissling, 2012; Wallace,
Knollman-Porter, Brown, & Hux, 2014). In this study, aided modelling was applied as an
input strategy to read the life story content to the participant verbally and simultaneously
point to AAC symbols (photographs) or written text on the AAC device. In this way aided
modelling was used to increase the auditory comprehension of the spoken life story (see
Section 3.4.1.1). Furthermore, person-centred communication related to validation,
recognition, validation, and empathy were integrated with the AAC scaffolding strategies
(Alant, Samuelson, & Ogle, 2015; Alsawy et al., 2019; McEvoy & Plant, 2014;
Savundranayagam, et al., 2016; Savundranayagam & Moore-Nielsen, 2015; Williams et al.,
2017).

Table 3.4 presents a summary of studies that the researcher consulted, the authors
involved, and key aspects of supporting interpersonal interaction with AAC strategies and
person-centred communication.

51
© University of Pretoria
Table 3.4: Evidence-based PC-AAC scaffolding strategies

Evidence-based Authors Description of Population Main techniques, strategies extracted

scaffolding evidence
strategies source
Aided AAC Dada et al. (2019); Dada et al. (2020); Dietz et Intervention Persons with Auditory comprehension supported by
modelling al. (2014); Wallace et al. (2012); Wallace et al. studies aphasia supplementary spoken language and
(augmented input (2014) pointing to AAC symbols. The
techniques) conversational partner highlights
information with multiple modalities
(written text, gestures) during interpersonal
interaction.
Multimodal cues Diehl & Wallace (2018); Rebstock & Wallace Individual Adult neuro- Prosodic emphasis to supplement spoken
Attentional cues (2020); Wallace et al. (2012) intervention cognitive language and assist in comprehension.
Prosodic emphasis studies communication Attentional cues to gain attention
disorders
Expectant delay O’Neill, Light & Pope (2018) Meta-analysis Aided input combined with expectant delay
(28 studies) (giving the person with dementia time to
Kent-Walsh, Murza, Malani, & Binger (2015) Meta-analysis respond).
(17 studies)
Pennington, Goldbart, & Marshall (2004) Systematic Interactional partners’ responsiveness.
review (four
studies)
Person-centred, and Alant, Samuelson, & Ogle (2015); Alsawy et Individual Persons with Relates to statements:
empathetic al. (2019); McEvoy & Plant, 2014; intervention dementia Empathetic responsivity: Responds at a
communication Savundranayagam et al. (2016); studies feeling level.
Savundranayagam & Moore-Nielsen (2015); Personal recognition: Communicates a
Williams et al. (2017) well-known trait of the person with
dementia.
Personal elaboration: Elaborates a personal
aspect of the person with dementia.

52
© University of Pretoria
3.4.2.2. Turn taking with the implementation of PC-AAC scaffolding
To facilitate turn taking, the evidence-based PC-AAC interaction strategies (see Table
3.4) – aided modelling; expectant delay; attentional cues; prosodic emphasis; person-centred
communication (personal recognition, personal elaborations, validation, empathetic listening)
– were applied in six steps. The purpose of these steps was to provide PC-AAC scaffolding to
support the participant to also take a turn. The participant’s turn taking responses could be
verbal (V), gestural (G), aided (A) or multimodal (MM) (e.g., a combination of verbal
utterance and a gesture, such as a head nod ‘yes’ in agreement). The purpose of using PC-
AAC scaffolding was to support the participant to use multimodal interaction to increase her
turn taking ability and to interact in the electronic life story (described in Par. 3.4.2.1). The
six steps are explained in detail in Figure 3.4.

2. 3. 6.
1. Person-Centred 4. 5.
Initiate Close
Offer life Communication Expectant Accept
interaction interaction
story PR, PE, ER delay responses
(with aided (respond
choice (with attentional (pause and (of any
modeling, contingently
(with aided cues and/or wait 5 modality)
affirm self- with aided
modeling) prosodic seconds)
identity) modeling)
emphasis)
Turn taking opportunity: From initiation to closure of interaction
Key
Indicates the sequence of PC-AAC strategies implemented by the researcher.
Indicates turn taking within a turn taking opportunity that should be scored on the companion
AAC outcome measure for PwD
Indicates 5 seconds after the closure by the researcher, but the participant responds.
ER Empathetic response: Validates the participant at a feeling level or provides affirmation.
PE Personalised elaboration: Communicates an expansion or elaboration of a personal aspect of
the life story sentence.
PR Personal recognition: Communicates a recognised/well-known key trait, personalised action
performed by the participant or family member in the AAC symbol (photographs).
Attentional A cue to direct the attention of the participant (e.g., using the participant’s name or the word
cue “so, in this picture”…).
Prosodic Raised intonation of spoken utterance to draw emphasis to meaning of a word.
emphasis

Figure 3.4: Schematic representation of the turn taking opportunities within the life story
interpersonal interaction

Table 3.5 contains a description of each of the six steps and an example of an
interpersonal interaction script in order to illustrate the application of the life story
interpersonal interaction with the participant.

53
© University of Pretoria
 Table 3.5: Application of the PC-AAC scaffolding strategies and conversational partner
script
INTERVENTION Application of PC-AAC scaffolding OUTPUT
STEP 1-6 (with an example of the researcher’s script) (participant’s
(begin after (only applied in the intervention phase) possible responses)
familiarisation with V G A MM
the AAC device)
1. Offer life story Aided modelling: Point to life story topics on AAC
choice device while offering each life story choice.

2. Initiate interaction Aided modelling: Point to AAC symbols while x x x x

reading life story sentence.
Researcher reads life story sentence: (e.g., “I am Anna
Brown. I was an actress and model when I was
younger.”

3. Person-centred Attentional cue +Aided modelling: Attentional cue x x x x

communication (name/“so”..) + point to AAC symbols (photograph)
(use participant’s while verbalising PCC comment or question.
name+ Personal Contextual support of people, object, location.
Recognition [PR] Prosodic emphasis +Aided modelling: To
comment/question supplement spoken speech
[PE], [ER] Researcher says: “Mrs Brown in this picture you
(prosodic emphasis) are on the front cover of a
magazine.” “That must have been a special moment?”
4. Expectant delay.
Wait 5 seconds. If no response after 5 seconds:
(a) Researcher repeats the script once with aided modelling. Wait 5 seconds.
(b) If no response after (a), researcher makes a choice, confirms with participant.
(c) If inappropriate response, researcher responds contingently, re-focuses participant back to topic,
then performs (a) and or (b) accordingly.

5. Multimodal responses x x x x
6. Close interaction Respond contingently to the participant’s responses x x x x
with aided modelling while pointing to AAC symbols.
Prosodic emphasis: To supplement spoken speech.
Researcher says: e.g. “Yes, Mrs Brown, being on the
front cover of a magazine must have been a special
moment…”

Before the researcher applied the PC-AAC scaffolding strategies, the participant was
familiarised with the AAC device (see Par. 3.4.1.3). As was subsequently shown in Table 3.5,
the researcher adhered to a semi-structured conversational partner script and applied the PC-
AAC conversational partner strategies. Collectively, these procedures were undertaken in six
intervention sessions with the participant.
3.4.2.2.1. Offer life story choice
Aligned with a person-centred approach, the participant was offered two choices of life
story topics (see Par. 3.4.1.1). The researcher used aided modelling to point to the labels of

54
© University of Pretoria
two life story topics, while simultaneously naming each topic verbally. The researcher
throughout maintained eye contact with the participant and waited for 5 seconds for a
response.
3.4.2.2.2. Initiate interaction
The interaction began by the researcher initiating the life story with an attentional cue
(e.g., using the participant’s name), followed by a person-centred comment or question with
an attentional cue (e.g., “in this picture”). She then waited for a response and concluded by a
modelling prompt of the participant’s responses. Next, the researcher used aided modelling to
either prompt the participant to provide a response or support them in elaborating, in order to
create further turn taking opportunities.
3.4.2.2.3. Person-centred communication
The use of person-centred communication in the form of comments or questions was
applied as a personal recognition, personal elaboration or empathetic response.
3.4.2.2.4. Expectant delay
The researcher paused and waited to allow the participant time to respond. If the
participant did not make a topic choice, the researcher repeated the AAC intervention.
3.4.2.2.5. Accept any responses
Since the outcomes were intended to create enjoyment and satisfaction with the
interpersonal interaction (rather than to focus on the quality of the language), off-topic
responses were not corrected. Hence, off-topic responses were counted as a turn and coded
accordingly.
3.4.2.2.6. Close interaction
Aided AAC modelling was used at the closure of the interaction to contingently
respond to the participant. The researcher paused for five seconds after the closure of one
turn taking opportunity and before beginning the next, so as to allow the participant to
respond. Any turn taking responses by the participant within this five-second interval were
scored (see Section 3.6.1.1 and Table 3.9).

3.4.3. Procedural checklists

Two procedural checklists were developed to implement the electronic life story
conversational support in a real-life context.
3.4.3.1. Procedures checklist: Baseline phase
A general procedures checklist (Appendix W) with a score sheet was developed to
rate the procedures applied in a baseline phase during which the researcher did not apply the

55
© University of Pretoria
PC-AAC scaffolding strategies with the participant. Further details of this checklist are found
in Par. 4.11.3.
3.4.3.2. Procedures checklist: Intervention phase
A general procedures checklist (Appendix X) was developed to rate the procedures
for applying the PC-AAC scaffolding strategies as outlined in Par. 3.4.2.2. This checklist
ensured that the researcher provided a familiarisation procedure (see Par. 3.4.1.3) of the AAC
device prior to the application of six PC AAC scaffolding strategies. Further description of
this checklist is presented in Par. 4.11.3.

3.5. Component 3: AAC outcomes for persons with dementia

3.5.1. AAC outcomes with the electronic life story conversational support

The interpersonal interaction of the participant with the electronic life story conversational
support (Component 2 in Par. 3.4) has a relational focus. The goal is to maintain relationships
or to create a sense of social connection between a person with dementia and their
conversational partner with the use of the electronic life story conversational support. The
outcomes are intended to be on two levels (intra-personal and interpersonal). On an intra-
personal level, the intended outcome is enjoyment, and satisfaction with interaction that
unfolds when the electronic life story conversational support is used with a conversational
partner. As such, these are best rated from the direct perspective of the person with dementia.
On an interpersonal level, the outcomes are an increase in turn taking and they are measured
using the AAC outcome measure developed in this study (see Par. 3.6).

3.6. Component 4: AAC outcome measure for persons with dementia

The Augmentative and Alternative Communication Outcome Measure for Persons

with Dementia (AACOM-PwD) was developed in this study as a companion outcome
measure to the electronic life story conversational support described in Par. 3.4.

Table 3.6: Component 4: Sub-components and materials developed

Component 4 Sub-components Outputs developed in the study

AAC outcome AAC Outcome Measure AACOM-PwD
measures for Persons with - Four turn taking domains
Dementia - 11 turn taking sub-domains
(AACOM-PwD) - Turn taking coding scheme
- Turn taking score sheets

56
© University of Pretoria
 Since an AAC outcome measure for persons with dementia does not exist, the
researcher reviewed five outcome measures with established psychometric properties from
the field of nursing (Williams, Newman, & Hammar, 2018; Jones, Sung, & Moyle, 2018),
speech-language therapy (Brady et al., 2018) and psychogeriatrics (Mabire, Gay, Vrignaud,
Garitte, & Vernooji-Dassen, 2016). Table 3.7 provides an outline of the existing outcome
measures, the authors and the main interpersonal domains of each measure that was
inspected.

Table 3.7: Existing outcome measures that informed the development of the AACOM-
PwD

Outcome measure Authors, year, field Main domains of each measure

Verbal and Non- Williams, C., Newman, • Social (nonverbal)
verbal Interaction D., Hammar, L.M. (2018) • Unsocial (non-verbal)
Scale Nursing • Social (verbal)
• Unsocial (verbal)

Communication Brady, N.C., Fleming, K, (Expressive) Communication levels

Complexity Scale Romine, R., Holbrook, • Pre-intentional communication
A., Muller, K., & Kasari, (e.g., oriented to an object with non-words
C. (2018) vocalisation or gestures)
• Intentional-non-symbolic
Speech-Language (intentional gestures, e.g., pointing with eye
Pathology gaze)
• Intentional-symbolic (e.g., spoken words or
use of AAC symbol selection, signs)

Social Observation Mabire, Gay, Vrignaud, • Facial expressions (e.g., eyebrow raise)
Behaviours Resident Garitte, & Vernooji- • Verbal interaction (e.g., requests, answers)
Index (social Dassen (2016) • Quasi-linguistic interactions (e.g., pointing)
interaction between • Interactive behaviours (body orientation
people with Psychogeriatrics towards speaker)
dementia)
Engagement of a Jones, Sung, & Moyle • Verbal sub-scale (initiates, participates or
Person with (2018) maintains verbal conversation or sounds in
Dementia Scale response to activity)
(EPWDS) Nursing • Visual subscale (maintains eye contact with
activity)
• Behaviour subscale (responds to an
activity/person by approaching, touching)
3.6.1. The AACOM-PwD description

Following the review of existing outcome measures in Table 3.7, the AACOM-PwD
was developed to measure turn taking. Jones, Sung & Moyle (2018) developed the
Engagement of a Person with Dementia Scale (EPWDS) which has established psychometric

57
© University of Pretoria
properties. The researcher considered measuring engagement by using the EPWDS (Jones et
al., 2018) as a complementary outcome measure to the AACOM-PwD. However, in the
context of exploratory probing into the AACOM-PwD with a single case study participant,
the researcher decided that the AACOM-PwD would be piloted first as an outcome measure
to measure concrete turn taking domains.
In this study, turn taking was operationalised as the exchanges that occur between the
dyad and that are independent of the form used by the conversational partners (Granlund &
Wilder, 2006; Wilder, 2008). The AACOM-PwD was developed as an observation-based
measure to observe the turn taking of a person with dementia with a conversational partner
(the researcher) during a ten-minute video-recorded interpersonal interaction. The turn taking
domains are measured as the participant’s responses during the use of the electronic life story
conversational support before (i.e., baseline) and after (see Par. 3.4.1.2 and Figure 3.4) the
application of the PC-AAC scaffolded strategies.
The AACOM-PwD categorises turn taking into four domains: verbal turns (V),
gestural turns (G), aided turns (A) and multimodal turns (MM). The turn taking process is set
within the boundary of a mutual focus of attention (i.e., electronic life story conversational
support on the AAC device). Each domain is divided further into eleven sub-domains to
specify the type (form) of turn taking within the respective domain. The domains and sub-
domains are summarised in Table 3.8, and the full operational definitions of the AACOM-
PwD domains are found in Appendix F2.

3.6.1.1. AACOM-PwD coding scheme and scoring

Aligned to each life story topic (see Par. 3.4.1.1), there are a minimum of six turn
taking opportunities. Turn taking is scored according to the six PC-AAC scaffolding steps to
support the participant in taking turns (see Par. 3.4.2.2). Turn taking is coded using a turn
taking coding scheme (see Table 3.8), and a full score sheet is found in Appendix F1.

58
© University of Pretoria
Table 3.8: AACOM-PwD scoring on scoring sheet (example)
Turn taking Domains
Interaction 1 Verbal (V) Gestural (G) Aided (A) Multimodal (MM) Turns
Turn taking Sub-domains
a b a b c a b a b c d

Offer Choice ✓
PCC+wait ✓
Model choice ✓ ✓

Subtotal 2 1 1 4
Subtotal 3 1
Turn taking Domains: (V) Verbal turns: turn taking with vocalisations or verbal utterances, (G) Gestural turns: turn taking
with body movements, head nods or facial expressions, (A) Aided turns: turn taking with or without the use of the AAC
device, (MM) Multimodal turns: a combination of two or more of the aforementioned turns.
Turn taking sub-domains: Va (Vocalisations), Vb (Verbalisation), Ga (Body movement), Gb (Head nod), Gc (Facial
expressions), Aa (aided turn with assistance), Ab (aided turn without assistance), MMa (Verbal + Gestural), MMb (Verbal +
Aided), MMc (Gestural + Aided), MMd (Verbal + Aided + Gestural)

As shown in Table 3.8, a score is assigned in each interpersonal interaction opportunity

for the turn taking domain (e.g., verbal) and the sub-domain, i.e., specific modality of verbal
turn taking (e.g., a verbal turn using a vocalisation). Scoring is linked to the operational
definitions in Appendix F2. The domain and sub-domain turns are tallied within each turn
taking opportunity. At the end of the entire 10-minute video recording, turn taking is scored
on three levels: the total number of turns taken by the participant, and the type of turn taking
at a domain level and sub-domain level.

3.7. Summary

Chapter 3 described the PC-AAC intervention package for PwD as it was developed with
a person with moderate to severe dementia in this study. The intervention comprised four
main components, and each component was described individually in relation to its sub-
components, underlying processes, developed materials and evidence-based strategies. The
first and second components described the roles the participant played in co-producing her
electronic life story conversational support. Next followed a description of the AAC
outcomes and of the development of an AAC outcome measure to measure turn taking with
the electronic life story conversational support (second component). The chapter concluded
with an outline for scoring the AAC outcome measure for PwD. The feasibility of testing and
implementing various components of the PC-AAC intervention package for PwD will be
described in the next chapter.

59
© University of Pretoria
 CHAPTER 4: METHODOLOGY

4.1. Introduction

This chapter discusses the research methodology used in this study. First, the main aim
and sub-aims are presented. This is followed by a description of the research design, the
rationale for this study, and an overview of the three research phases. Next, research Phase 3
is detailed in two parts: Phase 3A describes the four steps involved in Q-methodology related
to online data collection methods, the recruitment of professional participants, and data
analysis procedures. Phase 3B presents a case study with a description of the participant and
setting, participant selection criteria, in-person data collection procedures, recruitment
materials and data analysis procedures.

4.2. Research aims

4.2.1. Main aim

The main aim of this study was to explore the development and feasibility testing of a
person-centred AAC intervention package to support interpersonal interaction in persons with
dementia. Hereafter, this is referred to as a PC-AAC intervention for PwD.

4.2.2. Sub-aims

In order to fully explore the main research aim, four sub-aims were delineated:
i. To review the current literature in a systematic manner to identify AAC interventions in
persons with dementia. A research review was conducted by May et al. (2019) and
described in Par. 2.5.1. A supplementary scoping review was performed by Dada et al.
(2020) and is described in Par. 2.6.
ii. To develop a PC-AAC intervention for PwD based on the findings in (i). The
development process is described in Chapter 3.
iii. To identify and describe the patterns of expert professionals’ viewpoints on four
components of the PC-AAC intervention for PwD developed in (ii).
iv. To implement the PC-AAC intervention for PwD developed in (ii) in a real-life context
with a participant with dementia in order to:
a) Assess the procedural reliability of implementing an electronic life story
conversational support (i.e., Component 2) with a participant with moderate to
severe dementia.

60
© University of Pretoria
 b) Assess the intra-rater and inter-rater reliability of scoring an AAC outcome
measure for PwD (i.e., Component 4) of the PC-AAC intervention for PwD
developed in (ii). This assessment can be found in Par. 4.13 and Par. 5.4.
c) Determine the self-reported enjoyment (Component 3) of a participant with
moderate to severe dementia when using a co-produced electronic life story
conversation support.

4.3. Multimethod research design

This study used a multimethod research design (Hunter & Brewer, 2006). According
to Pearce (2015), a multimethod research design is a hybrid method as it entails a
combination of different types of research methods, data collection strategies and data
analysis procedures that are employed within a single research study (Hunter & Brewer,
2015). A multimethod research design adopts a pragmatic philosophical approach (Hesse-
Biber, 2015) that centres around the research aim. Pragmatism orients itself to solving
practical solutions in the real world, and pragmatic epistemology infers that knowledge is
based on experience, actions and beliefs (Kaushik & Walsh, 2019). Furthermore, a pragmatic
philosophical approach utilises multiple methods in order to meet the specific research
purpose (Kaushik & Walsh, 2019). The rationale for selecting a multimethod research design
for the current study was as follows:
The initial research aim of this study was to determine the effect of an electronic PC-
AAC intervention for interpersonal interaction in persons with dementia by means of
collecting data directly from them. However, after a single day, data collection for the
original study was abruptly halted in March 2020 due to the COVID-19 pandemic. The
subsequent lockdown measures in South Africa directly impacted the original research
method in two ways:
i. The intended participant sample had to be changed. Elderly persons aged 60 years and
older, including persons with dementia living in frail care and in care homes for the
elderly, were classified as a high-risk group for contracting COVID-19 due to the
higher prevalence of underlying or co-occurring health conditions in this age group
(Department of Health, Republic of South Africa, 2020).
ii. The geographical location of data collection was also affected. The Western Cape, the
province in which data collection had been planned, was the first province to record the
highest COVID-19 infection rates per capita in South Africa (Western Cape

61
© University of Pretoria
 Government, 2020). Consequently, in-person data collection with persons with
dementia at the original research sites was suspended indefinitely.
Given the unprecedented circumstances brought about by the COVID-19 pandemic and
its direct impact on the original research study, flexible and innovative online research
strategies had to be sought (Smith, Ostinelli, & Cipriani, 2020; Nicol, Piccirillo, Mulsant, &
Lenze, 2020). Hence, the research aim was amended to be exploratory by employing a
multimethod research design. Overall, the multimethod research design adopted in this study
included three methods: a research review methodology and a scoping review (Grant &
Booth, 2009); Q-methodology (Brown, 1980); and a case study with a single participant with
dementia (Yin, 2011). Each research method yielded its own dataset and required its own
data analysis procedures. However, there was an overall level of complementarity, as each
method informed the main aim of the study (Morse, 2003).

4.3.1 Research phases

This study comprised of three research phases (see Figure 4.1) mapped onto the three
cornerstones of evidence-based practice (Schlosser & Raghavendra, 2004):
i. Phase 1 relates to the current research evidence on AAC for persons with dementia.
ii. Phase 3A relates the viewpoints of professionals on the four components of the PC-
AAC intervention for PwD based on their clinical and educational expertise, and
iii. Phase 3B relates to the direct stakeholder perspectives of a person with dementia in
using the electronic life story conversational support (component 2) with the researcher.
The direct stakeholder perspective is interlinked with Phase 2 (stakeholder involvement
in the development of Component 2).

PHASE 1
EBP: Research evidence
(sub-aim i)
Research method 1: Research review methodology (Grant & Booth, 2009)
Research review of AAC interventions in persons with dementia (May et al., 2019) (Chapter 2)
Secondary review: Scoping review on intelligent assistive technology devices for persons with
dementia (Dada et al., 2020) (Chapter 2)

PHASE 2
Development of a PC-AAC intervention for PwD (sub-aim ii)

PC-AAC intervention for Component 1: Person-centred care in AAC for PwD

Component 2: Electronic life story conversational support
PwD
Component 3: AAC outcomes for PwD
(Chapter 3) Component 4: AAC outcome measure for PwD
62
© University of Pretoria
 PHASE 3
Feasibility testing of the PC-AAC intervention for PwD

PHASE 3B
PHASE 3A

EBP: Clinical and educational EBP: Direct stakeholder’s

expertise: Professionals’ perspective
viewpoints
Sub-aim iv
Sub-aim iii Research Method 3: Case study (Yin, 2011)
Research Method 2: Q-Methodology
(Brown, 1980) Stakeholder perspectives: Participant with dementia
Researcher implements the electronic life story
Professionals involved: AAC researchers, conversational support (component 2) with the case
academic teachers, expert clinicians study participant to assess:
Q-methodology steps: (Chapter 4) a) Procedural reliability of implementation
Step 1: Concourse development b) Reliability of scoring agreement of
• Evidence source (i): review of the AACOM-PwD: (component 4)
literature (Chapter 2) • Intra-rater reliability
• Evidence source (ii): e-Delphi Panel • Inter-rater reliability
(Chapter 4) Quantitative dataset (Chapter 5)
Step 2: Q-set (37 statements)
Step 3: Q-sorting (ranking of statements) c) Self-reported enjoyment from the direct
Step 4: Factor analysis (Chapter 5) perspective of the participant with dementia
Qualitative dataset (Chapter 5)

Figure 4.1: Three research phases of this study

Figure 4.1 illustrates the three research phases in this study. In Phase 1, a research
review methodology (Grant & Booth, 2009) was used to undertake a review of AAC
interventions in a systematic manner. This was the primary review for the current study
conducted by May et al. (2019). The methodological process of conducting the research
review, together with a summary of the results, are presented in Chapter 2. A secondary
review by Dada et al. (2020) was performed as an independent scoping review. However, it
was integrated into the current study as it contributed supplementary literature on electronic
AAC systems for persons with dementia.
In Phase 2 (see Par. 3.2), a 4-component PC-AAC intervention for PwD was developed
for this study. The PC-AAC intervention for PwD comprised of various sub-components,
underlying processes and materials related to person-centred care in AAC for persons with
dementia, electronic life story conversational supports, AAC outcomes for persons with
dementia, and the development of the AACOM-PwD (an outcome measure for person with

63
© University of Pretoria
dementia). Following the development of the PC-AAC intervention package for PwD, Phase
3 (feasibility testing) was conducted in two parts.
In Phase 3A, feasibility testing focused on obtaining the viewpoints of expert
professionals on the four components and underlying sub-components of the PC-AAC
intervention package for PwD. Q-methodology (Brown, 1980) was employed as a research
method to be executed in four steps to examine professionals’ viewpoints in a systematic
manner. Step 1 involved developing a concourse from various sources of information to
identify current discourse related to the research topic. In this study, evidence sources
included two reviews of the literature (conducted in Phase 1). Additionally, an international
e-Delphi panel of expert professionals provided their opinion on three broad areas of the PC-
AAC intervention for PwD. Themes generated via content analysis were combined with
literature-derived information to build a set of statements (Step 2). These statements were
then ranked by a second group of professionals in Step 3 through an online Q-sorting process.
Following Step 3, Q-data was analysed quantitatively using factor analysis (Step 4).
In Phase 3B, a case study was employed (Yin, 2011). Two components of the PC-AAC
intervention for PwD were implemented with a person with moderate to severe dementia in a
real-life setting. The threefold purpose of Phase 3B was to assess the procedural reliability of
implementing the electronic life story conversational support, assess inter-rater and intra-rater
reliability of scoring the AACOM-PwD and determine the participants’ self-reported
enjoyment in using the conversational support with the researcher.

4.4. Ethical considerations for the overall study

This study received ethical approval from the Research Ethics Committee of the
Faculty of Humanities, University of Pretoria (see Appendices G1 and G2). All participants
who took part in both Phases 3A and 3B were assured of their confidentiality and anonymity.
Their biographical information was obtained for descriptive purposes only and was not
disclosed to any person outside of this study. Participants were ensured that, should their
responses be included in the thesis for the interpretation of the study results, a generic non-
identifiable coding system would ensure that all responses would be anonymised.
Participation in this study was completely voluntary, and all participants were free to
withdraw from the study at any time without providing a reason. Participants were informed
that all the study data was to be stored on a password-protected computer and would be
accessible only to the student researcher and supervisors of this study. They were also

64
© University of Pretoria
informed that once the research study had been completed, all data would be stored at the
Centre for AAC, Faculty of Humanities, for a period of 15 years, in accordance with policy at
the University of Pretoria. There were no tangible or financial incentives attached to their
voluntary participation in this study.

4.5. Phase 3A: Q-methodology

The aim of Phase 3A was to explore the viewpoints of professionals with relevant
clinical and educational experience on the components of the PC-AAC intervention for PwD
developed in this study (Chapter 3). To address the aim of Phase 3A, a Delphi method was
considered as a possible strategy to obtain professionals’ views. According to Hasson,
Keeney and McKenna (2000), the purpose of a Delphi method is to perform multiple rounds
of iterative feedback from professionals to obtain consensus on a research topic. Since the
purpose of Phase 3A was to explore and describe professionals’ viewpoints, rather than to
obtain consensus of professional opinion, Q-methodology was selected as an appropriate
research method.
Q-methodology is an integrated approach that is used to understand human
subjectivity (Brown, 1980). Based on an individual’s subjectivity, viewpoints can be
extracted to highlight aspects that may be relevant or important to an individual at a specific
point in time (Watts, & Stenner, 2012). In this study, understanding the viewpoints of AAC
professionals on components of the PC-AAC intervention for PwD contributes an
understanding of what may currently be important for or contentious in AAC research and
practice with persons with dementia.
Q-methodology is unique as it obtains professionals viewpoints qualitatively, but
employs quantitative research techniques to systematically evaluate qualitative data (Brown,
1980; Yang, 2016; Zabala & Pascual, 2016). Q-methodology has the advantage that it
enables a more nuanced and detailed understanding of patterns of professionals’ viewpoints,
which is uncommon in conventional survey methods (Yang, 2016). Furthermore, in
comparison to traditional survey designs, small sample sizes do not affect the research
findings in Q-methodology studies. This is due to the methodology being well-suited for
exploratory research, which renders generalisability irrelevant to Q-methodology studies
(Brown, 1980; Yang, 2016). Q-methodology is conducted as a sequential multistep process
(McKeown & Thomas, 2013), and in this study, four sequential steps were involved. These
were concourse development, developing a set of statements called a Q-set, ranking the set of

65
© University of Pretoria
statements though a process called Q-sort, and factor analysis and interpretation of Q-data.
The details of the four steps are depicted in Figure 4.2.

• Concourse development (collection of evidence on the research topic)

•Research review of AAC interventions for persons with dementia (Chapter 2)
•Scoping review of IATDs for persons with dementia (Chapter 2)
Step 1
•Expert opinions via an e-Delphi panel

• Q-set (preparation of statements to be ranked by professionals)

•Identification of initial statements from the concourse (content analysis)
•Pilot statements online
Step 2 •Final set of statements

• Q-sort (professionals sort and rank statements via Q-sorting)

•Participant (professional) recruitment
Step 3 •Q-data collection

• Factor analysis and interpretation

•Factor anaysis of Q-data (Chapter 5)
Step 4 •Factor interpretation (Chapter 5)

Figure 4.2: Orientation to Q-methodology employed in this study

Next, an overview is given of the Q-methodology steps as shown in Figure 4.2, based
on the methodological procedures applied within each step.

4.5.1 Step 1: Concourse development

In Q-methodology, the concourse is the foundation of broader scientific evidence and

the range of current discourses related to the research topic, extracted from various evidence
sources (McKeown & Thomas, 2013). In this study, three main sources of evidence added to
the comprehensiveness and richness of the concourse. Firstly, the researcher conducted a
review of AAC interventions for persons with dementia in a systematic manner undertaken in
Phase 1 of the main study (Par.2.5.1). This information was supplemented by a second
scoping review on intelligent assistive technology devices for persons with dementia. In Q-
methodology, data from in-person focus groups and professional interviews is typically used
in developing the concourse (Newman & Ramlo, 2013). However, given the imposed
COVID-19 social distancing regulations of 2020, in-person focus groups were not possible.

66
© University of Pretoria
Therefore, the third evidence source was obtained from a geographically dispersed panel of
international professionals who provided their expert opinion on three broad, open-ended
questions related to the research topic (i.e., a single- round international e-Delphi panel).
Collectively, all three evidence sources formed the basis of statements developed in Step 2.
The next section describes the e-Delphi panel participants, and general procedures followed
by the e-Delphi panel.

4.5.1.1. Step 1.1: Professional participants

To participate in the e-Delphi panel, professionals were selected based on three

selection criteria.
i. In alignment with the research aim, participants had to be professionals from the
disciplines of AAC, speech-language therapy or a related health science discipline by
virtue of their scholarly work in the AAC and dementia-related communication
literature.
ii. In this study, an expert was defined as a professional who met two pre-requisites. The
professionals had to hold a degree in the aforementioned professional disciplines, and
they had to be involved in AAC and/or dementia-related research, clinical work or
teaching in their respective professional disciplines. As a measure to verify
professionals’ educational qualifications, the researcher undertook a three-pronged
review process. This process entailed the review of the author details of potential
professional participants as these appeared in their recent research publications, as well
as checking their Open Researcher and Contributor Identity (ORCID) profiles and their
online clinical profiles (where applicable).
iii. Professionals were also selected based on their area of research interest within the
scope of adult cognitive-communicative disorders, interpersonal interaction research in
general or dementia-related communication in particular, person-centred care in
dementia studies, communication technology, or instrument development. This resulted
in a diversity of professional perspectives (Avella, 2016). Professionals’ final inclusion
in this study depended on their voluntary consent to take part in the e-Delphi panel and
their submission of a completed questionnaire before the expiry date of the e-Delphi
panel round.

67
© University of Pretoria
4.5.1.2. Step 1.2: Recruitment of professionals

A non-probability, purposive sampling strategy was employed to recruit professionals

for the e-Delphi panel. The aforementioned selection criteria were applied, which resulted in
a list of 63 potential professionals who were invited to participate.
The potential e-Delphi panel of professionals were recruited by obtaining their email
addresses from their publicly available online professional profiles. An invitation letter
(described in Par. 4.5.1.4) was personally addressed and emailed to each potential
professional participant. This was sent from the email accounts of the two supervisors of this
study. Temporal and work-related factors were the reason for making initial email contact in
this manner. This was because recruitment procedures overlapped with the mid-year summer
vacation of potential professional participants in the northern hemisphere, and the increased
academic and clinical demands of potential participants had to be considered owing to the
COVID-19 pandemic (Sahu, 2020). As such, the chances of potential participants flagging
the researcher’s email as unsolicited or spam were much higher than usual under these
circumstances. In total, 37 professionals were volunteered to participate in the e-Delphi panel
and are described next.

4.5.1.3. Step 1.3: Description of professionals

Professionals contributed to two datasets in Step 1 and Step 3 of the Q-methodology

process. As shown in Figure 4.2, professional opinions were obtained in Step 1 (i.e.,
concourse development, source 2, e-Delphi panel) and professionals performed the Q-sort in
Step 4. The latter group of professionals are referred to as the P-set (McKeown & Thomas,
2013). For ease of reference, in describing the professionals, both sets of professionals are
presented in Table 4.1.

68
© University of Pretoria
Table 4.1: Description of professionals
e-Delphi panel P-set
(N= 31) (N=33)
Gender Female 23 29
Male 7 4
Prefer not to say 1 n/a
Years of 0-5 years 1 1
experience 6-10 years 4 7
11-15 years 5 1
16-20 years> 21 24
Region Australia 2 1
Brazil 1 1
Canada 0 1
Finland 1 0
Germany 1 1
Ireland 2 1
South Africa 1 13
Sweden 4 3
United Kingdom 14 10
United States 5 3
Profession Speech-language therapy 17 25
Psychology 8 3
Nursing 1 0
Other (communication technology, human 5 5
rights, education)
Area of AAC 11 19
specialisation AAC and dementia 5 0
Dementia studies 5 6
Communication technology * 2
AAC, dementia and technology 0 1
Other 11 5
Focus of Research 16 5
work Clinical work 0 6
Research and clinical work 7 1
Research and teaching 6 10
Research, clinical work and teaching 0 10
Other 2 1
Area of Instrument development 2 2
interest Adult cognitive-communicative disorders 4 **
(general)
Dementia-related communication (specific) 6 2
Interpersonal interaction (general) 10 12
Person-centred care 0 9
Other (AAC technology) 9 1
**Patient Adults with acquired communicative disorders 16
population Children with disabilities 5
served Adults with acquired communicative disorders 12
and children with disabilities
**This category was only specified for Step 4 (Q-sort)

In total, 31 professionals participated in the e-Delphi panel (response rate of 49%) and
33 performed the online Q-sort (response rate of 54%). Overall, the vast majority of these

69
© University of Pretoria
professionals were speech-language therapists (n= 17 and n=25 respectively) who specialised
in the field of AAC (n=11; n=19). Professionals on the e-Delphi panel were primarily
involved in research (n=16), followed by research and clinical work (n=7), and research and
teaching (n=6). Professionals in the P-set were equally involved in research and teaching
(n=10), and in a combination of research, clinical work and teaching (n=10). The main
interests of the majority of professionals were interpersonal interaction (n=10 and n=12
respectively). Participants were geographically dispersed across ten countries, with the
majority from the United Kingdom (n=14 in the e-Delphi panel) and South Africa (n=13 in
the P-set). The professionals, mainly female, were overall well experienced, with work
experience of between 16 and 20 years or more. Professionals’ expertise focused on
interaction with adults with acquired communicative disorders (n=16) or a combination of
both adults and children with communication disabilities (n=12), while five participants
focused primarily on interaction with children with disabilities only (n=5).

4.5.1.4. Step 1.4: Concourse development procedures

An invitation and informed consent letter was sent to each potential professional
participant to inform them of the reason for conducting an e-Delphi panel, the purpose of the
study, how their email addresses had been obtained, and what was expected of their
participation as an e-Delphi panellist (see Appendix H).
The e-Delphi panel questionnaire (Appendix I) comprised two sections, namely the
biographical information of professionals, and three open-ended questions related to the
research topic:
i. AAC strategies and communication supports for PwD
ii. Person-centred care within AAC-supported interactions
iii. Outcomes to include in an AAC interpersonal interaction outcome measure for PwD
Open-ended questions ensured that professionals’ responses were not prompted nor
led by the researcher, and thus prevented response bias (Sinha, Smyth, & Williamson, 2011).
Unlimited free text gave professionals the liberty to provide as much input as they preferred.
Participants were asked a final question on whether they would consider continuing with the
study in the Q-sort by ticking yes, maybe or no. The online questionnaire was developed on
and administered via Qualtrics (www.qualtrics.com) as a dedicated online survey software
program.

70
© University of Pretoria
4.5.1.5. Step 1.5: Concourse development data analysis

e-Delphi panel data was downloaded from Qualtrics.com and analysed with qualitative data
analysis software, ATLAS. tiTM for MAC, version 8.4. The purpose of analysing the e-Delphi
panel data was to contribute to the development of a set of statements to be ranked in the Q-
sort (Step 3). Data was analysed using a combined inductive-deductive approach to content
analysis (Fereday & Muir-Cochrane, 2006). Data was analysed for content that was
observable and measureable to obtain a frequency count of instances in which codes occurred
(Graneheim & Lundman, 2004). Codes were assigned deductively by assigning a pre-
determined code extracted from the literature (i.e., literature-derived). Data that could not be
assigned to a literature-derived code were created as a new code in an inductive manner
(Fereday & Muir-Cochrane, 2006).
Coding of data took place in four waves. Firstly, the researcher coded the dataset
independently several times. Next, an independent checker, a PhD student, checked the entire
dataset for accurate application of the coding rules and appropriateness of coding categories
and broad themes. Disagreements were discussed and agreement was reached on the final set
of codes, categories and themes. Thirdly, the researcher re-coded the complete dataset twice
to ensure all final coding was correctly applied. Finally, trustworthiness of the data coding
was assessed by a second independent coder, a post-doctoral fellow in AAC, who coded a
randomly selected portion of the dataset to assess inter-rater reliability of coding. Coding
disagreements were discussed by re-examining the meaning unit in question and deliberating
until consensus was reached.
The final themes used for the generation of the concourse are shown in Appendix J.

4.6. Step 2: Q-set (Statements)

The Q-set consists of statements derived from the concourse and ranked by each
professional. Based on the themes developed in Step 1.5 (Par. 4.5.1.5), an initial set of 50
statements was generated by the researcher. Through an iterative process and in consultation
with the supervisors of the study, the researcher reduced the statements to 37 preliminary
statements which were piloted by two pilot testers as outlined in the section below.

4.6.1. Step 2.1. Piloting testing of the online Q-set

The preliminary 37 statements (Appendix K1) were ranked online on QMethod

Software (Lutfallah & Buchanan, 2019), a computer-based Q-methodology application
purchased by the researcher via a subscription package. The two pilot testers assessed the

71
© University of Pretoria
representativeness of the statements against the concourse and assessed the feasibility of
ranking the statement on QMethod Software (Lutfallah & Buchanan, 2019). Both pilot testers
were speech-language therapists: one was an AAC doctoral student with extensive experience
of adult-acquired cognitive-communicative disorders, and the second was an AAC Master’s
student whose research is based on assistive technology and persons with dementia.
Cognitive interviews, considered to be best practice in testing survey content prior to
its use, were used as a technique to improve the validity of the online Q-sort (Behr,
Meitinger, Braun, & Kaczmierk, 2017; Howlett, McKinstry, & Lannin, 2017). The purpose
of a cognitive interview is to understand the cognitive processes used by, in this instance, the
pilot testers. Specifically, it explored the testers’ (i) comprehension of the Q-sort questions,
(ii) retrieval of information from memory, (iii) how they made their decisions and (iv) how
they selected their final survey responses (Howlett et al., 2017). The cognitive interviews
were conducted retrospectively. In other words, immediately after the pilot testers had
completed their online Q-sorts, the researcher telephoned each pilot tester individually.
Following guidelines by Howlett et al. (2017), the researcher used a combination of pre-
planned and spontaneous verbal probes to ask the pilot testers questions. These questions
related to the comprehension of each written question, how pilot testers arrived at answers,
the clarity of statements, and the technical challenges experienced with the online platform.
The researcher captured the pilot testers’ responses on a table in Microsoft Word© 2016. Both
sets of responses were integrated and utilised to inform revisions to the final Q-set. The
findings of the cognitive interviews and adaptations that were made are summarised in Table
4.2. A final Q-set of 37 statements is evident in (Appendix K2).

72
© University of Pretoria
Table 4.2: Piloting of the online Q-sort: Aim, procedures, findings and adjustments
Aim Procedures Findings Adjustments
To determine whether The pilot testers Biographical information: • What is your area of interest within
biographical were emailed a • One question has two embedded sub-questions, namely research, teaching or clinical work?
information was link to the online population and area of interest. • Which population does your research,
adequate. Q-sort. teaching or clinical work focus on?

To assess the • Reiterate the time required to complete the Q-sort Instructions adjusted as follows:
comprehensiveness of Remind participants of the following: • “Rank the statements (from the pre-sorted
instructions. • The grid is forced-choice distribution. Provide an piles) accordingly to a forced-choice
explain of what this means. distribution grid that contains 37 cells. This
• Clarify: one statement in one cell. means each statement is placed
individually into one cell on the grid (i.e.,
one cell=one statement).”
Statements: • Two redundant statements were found. • Redundant statements were removed.
• To assess the • There were too few statements on non-electronic and • Dementia severity was inserted in specific
representativeness unaided AAC strategies. statements.
of the statements • Specific dementia severity was not stated. • Potential clinicians were recruited and
against the • Statements have high clinical utility. Clinicians should added to the invitation list for the Q-sort in
concourse. be invited as part of the Q-sorting. Step 3.
• To identify the • Too few negative statements had been included. Pilot • Ten positive statements were changed to
clarity of wording of testers suggested to change some positive statements to negative statements.
statements. negative ones, as there were too few statements to
disagree with.
To test the feasibility of The two pilot • None • None
online administration testers
related to technicalities/ conducted the Q-
concerns/ participant sorting on
code login, etc. various devices
(laptop, smart
phone, desktop
computer).

73
© University of Pretoria
4.7. Step 3: Q-sort (ranking of statements)

In Q-methodology, the Q-sort refers to the process in which participants rank a set of
statements according to a forced-choice distribution grid while considering all other
statements. In this study, the Q-sort was facilitated electronically through QMethod Software
(Lutfallah & Buchanan, 2019) and the results of the Q-sort data provided the main data Phase
3A of this study. Q-methodology does not enforce a priori meaning to the statements that are
ranked (Watts & Stenner, 2005), which encourages participants to rate the significance of
statements based on their unique perspectives.

4.7.1. Step 3.1: Q-sorting – general procedures

Given the low response rate from the e-Delphi panel (Par. 4.5.1.3), and based on
recommendations from the pilot testers of the Q-sort (see Table 4.2), professionals who had
not been invited to the e-Delphi panel were invited to participate in the online Q-sort. The
same participant selection criteria as had been applied to the e-Delphi panel (Par. 4.5.1.1)
were applied to professionals invited as the P-set. Potential professionals in the P-set were
sent a letter of invitation (Appendix L) to state the purpose of the study, what was expected
of them in the Q-sorting process and to inform them that their participation was voluntary. In
total, 61 professionals were invited to participate as the P-set. The description of the P-set
appears in Table 4.1.
For the online Q-sort administration, each professional was emailed a link to the
online Q-sort as well as a unique participant login code. Upon clicking and entering the
landing page on QMethod Software (Lutfallah & Buchanan, 2019), participants completed a
biographical questionnaire, followed by detailed two-step Q-sorting instructions (Appendix
M).
Professionals were able to start the Q-sort and continue at a later time if preferred.
They had two weeks in which to complete the Q-sort, with a weekly reminder. During this
time, there was an unexpected automatic upgrade to the software, and some professionals
experienced technical difficulties in accessing their online Q-sort. Therefore, Q-sort data
collection was extended for a further two weeks (i.e., four weeks in total).

4.7.2. Step 3.2: Q-sorting

The researcher electronically uploaded all 37 statements from the Q-set to the
QMethod Software (Lutfallah & Buchanan, 2019). A 37-cell forced-choice distribution grid

74
© University of Pretoria
was automatically generated using QMethod Software (Lutfallah & Buchanan, 2019). The
online Q-sort was undertaken in two steps in which professionals subjectively sorted and
ranked the 37 statements. The first step entailed pre-sorting, where professionals divided the
37 statements into three piles and clicked on three icons to indicate their initial agreement,
disagreement or neutrality toward each statement. An example of the pre-sort is shown in
Figure 4.3.

(agree), (disagree), (neutral)

Figure 4.3: Q-sort (pre-sorting statements) on QMethod Software

The second step was the final sorting, in which the pre-sorted statements were
arranged according to a forced-choice distribution grid, based on a condition of instruction.
This meant that professionals had to rank the statements based on how strongly they agreed
with each statement. As shown in Figure 4.4., statements could be dragged from the top panel
of pre-sorted piles into the extreme right-hand side of the distribution grid to be ranked as
strongly agree (+3). The converse was inferred to statements on the extreme left-hand side,
ranked as strongly disagree (-3). Those statements with no definitive agreement were dragged
and dropped in the middle of the distribution grid as a 0 (neutral) view. This final sorting step

75
© University of Pretoria
aimed to capture each professional’s unique views of the statements and, based on the
premise of subjectivity, there were no right or wrong answers.

Figure 4.4: Q-sort (final sorting of statements) on QMethod Software

The online Q-sort on QMethod Software (Lutfallah & Buchanan, 2019) required from
professionals to drag the pre-sorted statements and drop them into a cell on the forced-choice
distribution grid. This meant that only one statement was permitted in one cell. Consequently,
professionals were free to swop and re-rank the statements based on their unique
subjectivities as many times as they wished before making a submission. This could be time
consuming and perhaps cognitively complex for professionals, which are common limitations
of Q-methodology. All completed Q-sort data was automatically captured on QMethod
Software (Lutfallah & Buchanan, 2019) and at the end of data collection, factor analysis was
performed as is detailed in the next section.

4.8. Factor analysis

In Phase 3A, factor analysis of Q-sort data was facilitated on QMethod Software
(Lutfallah & Buchanan, 2019), which enabled the automatic process of Q-data input,
correlation of factors, factor rotation, and the process of factor extraction. In this study, a

76
© University of Pretoria
factor was defined as a group of professionals whose completed Q-sorts shared a similar
point of view on the components of the PC-AAC package for PwD developed in this study.
Person-by-person factor analysis was employed as a data reduction method (Watts &
Stenner, 2012) to sort data into factors. Factor analysis was based on Q-sort data and a
synopsis of how the final Q-sorts were determined is presented in Table 4.3.

Table 4.3: Synopsis of Q-sort data

Q-sorts Number Percentage
(N=61)
Sent to professionals 61 100%
Not initiated 18 29%
Initiated but not completed 10 10%
Completed and submitted for factor analysis 33 54%
• Average time to complete Q-sort
29 minutes 26 seconds

As shown in Table 4.3, 61 potential professionals were invited to participate in the

online Q-sort. They had to subjectively arrange 37 statements (see Par. 4.7.2) according to an
agreement scale ranging from strongly agree (+3) to strongly disagree (-3). The statements
related to various components of a PC-AAC intervention for PwD (see Chapter 3, Par. 3.2).
Eighteen of the professionals (29%) did not initiate their online Q-sort, while 10 (16%)
initiated the Q-sort, but did not complete the sorting process. A total of 33 professionals
(54%) took an average time of 29 minutes 26 seconds to complete and submit their Q-sort.
The Q-sort data of these 33 professionals was analysed using factor analysis.
Factors were extracted and analysed by utilising the in-built statistical tools on
QMethod Software (Lutfallah & Buchanan, 2019). The analysis of factors in Q-methodology
refers to the application of statistical techniques to determine the patterns of viewpoints and
to explain the maximum variances in professionals’ Q-sort data (Watts & Stenner, 2005).
Factor rotations were performed to maximise the explained variance between Q-sorts.
QMethod Software (Lutfallah & Buchanan, 2019) offered various options as factor rotation
techniques. However, in line with the purpose of this study, Varimax rotation was selected to
statistically explain the maximum range of variances shared by the professionals. This
eliminated human error and contributed to the reliability and ease of visual inspection of
factor analysis results.
Following factor rotation, factor loadings were automatically generated from QMethod
Software (Lutfallah & Buchanan, 2019) and examined for their significance. In this study,

77
© University of Pretoria
p<0.5 was used to determine the significance of a factor loading and was calculated as 0.42,
based on the recommended equation by McKweon (2013):
1
2.58 𝑥 where N represents the number of Q-sorts in the study.
√𝑁
1
Therefore, in this study, the significance level of factor loadings was 2.58 𝑥 =0.42.
√37

4.9. Phase 3B: Case study

The aim of Phase 3B was to assess the feasibility of implementing the electronic life
story co-produced with a person with dementia (see Par. 3.3 and 3.4). The threefold purpose
of Phase 3B (aligned to sub-aim iv of the main study) was as follows:
i. Assess the procedural reliability of implementing the electronic life story
conversational support in a real-life context with a person with moderate to severe
dementia and the researcher as the conversational partner.
ii. Determine the self-reported enjoyment of a participant with dementia in using the
electronic life story conversational support with the researcher (see Par. 3.5.1).
iii. Assess the reliability of scoring the outcome measure, i.e., AACOM-PwD (see Par.
3.6.1).
In order to meet the aims of Phase 3B, a case study design was employed. Yin (2011)
describes a case study as an empirical enquiry that investigates a single case or multiple cases
within a real-life context, using multiple data sources and data analysis methods. Based on
Yin’s (2011) perspective, a single case study was found to be an appropriate method to
achieve the threefold purpose of Phase 3B.
Furthermore, a case study allowed for a detailed description of the process of co-
producing a life story conversational support for a specific person with dementia. This
process was explained in Par. 3.3. In the section that follows, the selection criteria,
recruitment and description of the case study participant are outlined, and ethical
considerations for data collection procedures are detailed.

4.10. Phase 3B: Participants

The recruitment of participants and the sampling method used, the criteria for
selectin participants, details of the participants, as well as the recruitment site are
described next.

78
© University of Pretoria
4.10.1. Recruitment of participants and sampling

A non-probability, purposeful sampling method was used to recruit participants

for the case study in Phase 3B. This sampling technique is time efficient, economical and
assures that the data obtained is representative of the case population under investigation
(McMillan & Schumacher, 2014).
Dementia-specific organisations were sought to retrieve a database list of
dementia care homes (i.e., old age homes, adult day-care facilities and frail care facilities
that cater for persons with dementia) in the Western Cape. This geographical location was
convenient for and accessible to the researcher.
Six organisations were telephoned, but only one expressed any interest in the
study. The Head Nurse of an organisation for the aged was sent an email to request
permission (Appendix N) to recruit participants from their organisation. Permission was
granted to recruit participants from five of their satellite dementia care homes within the
Cape Town Metropole. Following written permission from the organisation for the aged
(Appendix N), the nursing manager at each of the five care homes provided the details of
potential participants’ next of kin or legal guardian within their memory care unit.
Furthermore, a recruitment flyer was placed in the common lounge area of each dementia
care home, for distribution to visiting family members. Nursing staff assisted in reading
an adapted recruitment flyer (Appendix O) to potential participants (i.e., persons with
dementia).
The researcher sent an email with an attached letter of information (Appendix P) to
the legal guardians of 12 potential participants. One legal guardian expressed interest in the
study. A three-step informed consent process was employed to obtain the participation of a
person with dementia in the current research (Murphy, Gray, & Cox, 2007).
i. Firstly, the purpose of the research study was explained in an email to the potential
participant’s legal guardian which contained a letter of information (Appendix P). The
legal guardian granted informed consent by signing and returning the informed consent
reply form (Appendix P). The legal guardian identified a potential familiar
conversational partner to participate in the study and provided the details of an
identified individual for the researcher to contact personally. The researcher made
telephonic contact with the familiar conversational partner before a letter of
information (Appendix Q) and a biographical questionnaire (Appendix U2) were

79
© University of Pretoria
 delivered by hand. The familiar conversational partner agreed to take part in the study
by signing the informed consent form (Appendix Q).
ii. Secondly, the researcher met the potential participant personally at the dementia care
home to explain the purpose of the study by reading an adapted letter of information
(Appendix R) to request informed consent. The researcher supported the informed
consent process by using AAC strategies such as simplified keywords and pointed to
pictures on the letter while reading each sentence (Appendix R). AAC strategies were
employed to aid the potential participants’ understanding of the recruitment process.
Informed consent was provided by one person with dementia who verbally expressed a
‘yes’ in response to three informed consent statements which the researcher recorded
on an informed consent reply form (Appendix R). The participants’ informed consent
was validated under the observation of a nurse at the dementia care home to ensure that
the participant provided voluntary informed consent. An observer-to-informed-consent
form was accordingly signed by the nurse (Appendix S).
iii. Thirdly, the researcher abided by a process of ongoing consent to re-establish the
participants’ willingness to participate in the study at every instance of interaction with
the researcher (Dewing, 2008a; Slaughter, Cole, Jennings, & Reimer, 2007). An
ongoing informed consent form (Appendix T) making use of picture symbols was used
to support the participant in providing ongoing informed consent. If the participant
showed signs of transient fatigue or refusal to participate on any particular day during
the recruitment and subsequent interactions, the recommendations by Slaughter et al.
(2007) were followed:
• The researcher ensured that the participant was made to feel comfortable by
addressing the immediate signs of distress and thereafter requesting their
permission to proceed.
• If consent was declined, the study was suspended for that day.
• Dissent to participate in the study was confirmed if the participant was approached
on a different day and participation was refused.
• Accordingly, the participant’s involvement in the research study would be
terminated.

One participant who was eventually recruited – along with her familiar conversational
partner – to participate in the case study was selected based on the participant selection
criteria outlined below.

80
© University of Pretoria
4.10.2. Participant selection criteria

Table 4.4 outlines the selection criteria of the participant who was recruited for the
case study in Phase 3B.

Table 4.4: Selection criteria for person with dementia as case study participant
Criteria Justification Measure
Older adults (60 Age is a chief risk factor for Biographical questionnaire
years+) developing dementia (American Par. 4.11.2.1, Appendix U
Psychiatric Association [APA], 2013).

Degenerative A medically documented or probable i) Biographical questionnaire

dementia of mild, diagnosis of Alzheimer’s Disease Par. 4.11.2.1, Appendix U
moderate or (AD). Alzheimer’s dementia is the
advanced severity most common degenerative dementia (ii) Montreal Cognitive Assessment
without co- subtype (APA, 2013). Participants (MoCA) (Nasreddine et al., 2005) to
morbid conditions with the same dementia subtype will describe the clinical stage/severity of
ensure homogeneity in behavioural dementia
and communication characteristics. Par. 4.11.2.2.

Dementia-related Specific language and communication Dementia Communication Difficulties

communication difficulties are apparent in persons Scale (Murphy & Gray, 2007)
difficulties with mild and moderate dementia. Par. 4.11.2.4.

English as first The co-produced electronic life story Biographical questionnaire

language conversation will be conducted in Par. 4.11.2.1, Appendix U
English.
Functional hand Participants need to navigate the AAC Manual Ability Classification System
mobility device. (MACS) (Eliasson et al., 2006)
Par. 4.11.2.3.

Functional vision Participants must be able to see the (i) Biographical questionnaire
picture symbols on the AAC device. Par. 4.11.2.1, Appendix U

(ii) Participants will be requested to

wear their prescription reading glasses
if applicable.

The familiar conversational partner was recruited for the purpose of assisting the
participant to select photographs for Component 1 of the PC-AAC intervention package
(see Par. 3.4.1.1). The selection criteria for this partner are shown in Table 4.5.

81
© University of Pretoria
Table 4.5: Selection criteria for familiar conversational partner
Criteria Justification Measure
Adult, who is An adult (spouse, adult child, family Biographical questionnaire
familiar with the member, friend, professional carer) who has Par. 4.11.2.1, Appendix V
participant with background knowledge of the participant’s
dementia life history to assist the participant in
selecting materials for the co-produced
electronic life story conversational support
(Kindell et al., 2014).
Proficiency in All procedures relating to the co-production Biographical questionnaire
English of the electronic life story conversational Par. 4.11.2.1, Appendix V
support will be performed in in English.
4.10.3. Participant description
4.10.3.1. Participant with dementia

The participant with dementia was a 77-year-old female, with probable Alzheimer’s
Disease. She had been a highly skilled speech and drama specialist who studied music, film
and theatre professionally at an international institution. The participant was recently
widowed as her husband had died soon before she moved into the dementia care home. She
shared a close bond with her only daughter, who was also her legal guardian. There were no
reported concerns about her hearing. The participant had recently visited an optometrist for a
new pair of prescription spectacles. Medically, the participant was on chronic diabetic
medication for Type 1 Diabetes. She obtained a total score of 7/30 on the MoCA (Nasreddine
et al., 2005), which described her as a case of moderate to severe dementia (Chang et al.,
2012). She scored 19/39 on the Dementia Communication Difficulties Scale (DCDS)
(Murphy & Gray, 2007) which rates her communication difficulties in the upper range of
moderate stage dementia.
The participant experienced difficulty with understanding information presented to her
verbally, struggled to maintain a conversational topic and always substituted content words
with empty words (e.g., thing). The participant’s ability to handle objects in daily activities
was categorised as Level II on the MACS (Eliasson et al., 2006). She was left-handed and
was observed to have a slight weakness of her wrist due to possible age-related osteoarthritis.
This appeared to reduce the speed and accuracy of her hand movements and she was
observed to require assistance in independently performing certain actions that required a
firm hand grip. She was able to walk independently but preferred to be pushed in a
wheelchair inside the care home due to fatigue induced from walking a distance.

82
© University of Pretoria
4.10.3.2. Familiar conversational partner

The familiar conversational partner had a longstanding relationship with the participant
and her family. She had worked at the family’s residence for five years. Since the participant
moved into the dementia care home, the familiar conversational partner was employed
privately by the participant’s family. She spent eight hours a day with the participant at the
dementia care home for six days a week. The familiar conversational partner was also
responsible for attending to the participant’s personal care needs (e.g., driving her to doctor’s
appointments). The familiar conversational partner was a highly proficient second language
English speaker.

4.10.4. Description of recruitment site

The case study was undertaken at a dementia care home which is part of a multilocation
organisation that specialises in care residences for the elderly. This care home, which had a
designated memory care unit in which persons with dementia were residents, was
geographically located in an urban area in the Cape Town metropole. The majority of the
population residing in this area are typically professionals and are classified as high-income
households.
Within the dementia care home, all residents had their own private rooms. All research
procedures involving the participant were undertaken in her private room. There was a
common area in which all residents’ meals were served meals and where they participated in
joint activities such as music therapy and art and crafts. The dementia care home was a large,
multistorey building with a garden, and a coffee shop in which residents often met their
visitors if preferred.

4.11. Materials

The material related to the permission and informed consent letters, screening materials
and data collection are described below.

4.11.1. Materials used for permission and informed consent

4.11.1.1. Recruitment flyer

The recruitment flyer was used to invite potential legal guardians and potential
participants (Appendix O). The recruitment flyers contained an invitation for potential
participants, together with the contact details of the researcher.

83
© University of Pretoria
4.11.1.2. Permission letter to the dementia care home

The permission letter (Appendix N) was addressed to the manager of the dementia care
home and contained information on the purpose of the study, a request to distribute
recruitment flyers at the dementia care home, and permission to use the premises as a
research site to recruit participants. The permission letter detailed how data was to be
collected from participants and how it would be securely stored. It also confirmed that details
of the dementia care home would remain confidential. A signed permission reply slip
(Appendix N) was requested from the dementia care home to indicate their granting or
refusal of permission to conduct the study at the site.

4.11.1.3. Letter of information and informed consent: Legal guardian or next of kin

This letter contained the purpose of the study and requested the legal guardian’s
consent to approach the potential participant to invite their participation in the study. The
letter also requested consent to utilise the participants’ personal photographs in the electronic
life story conversational support, to identify a familiar conversational partner to assist the
potential participant in selecting photographs, and consent to have all interactions between
the researcher and participant video-recorded. The letter stated that all information would be
kept confidential and be stored electronically in password-protected files at the Centre for
AAC at the University of Pretoria for 15 years. Additionally, the letter assured participants’
voluntary participation and confirmed the protection of their identity. The legal guardian
acknowledged understanding of the contents of the letter, and indicated their informed
consent on a reply slip (Appendix P).

4.11.1.4. Letter of information and informed consent: Familiar conversational partner

The letter to the participants’ familiar conversational partner (Appendix Q) stated how
they were identified as a potential participant in the study, the purpose of the study and the
role they would play, should they take part in the study. Details of their voluntary
participation and withdrawal from the study at any point were also stated in the letter. A
signed reply slip indicating their informed consent to take part in the study was essential for
their participation in the study (Appendix Q).

4.11.1.5. Letter of information and informed consent: Participant

Since the letter of information was adapted for the participant, it contained simple
language. The information was presented in written text with picture symbols to support the
participant in understanding the purpose of the study, how data was to be collected and

84
© University of Pretoria
stored, and that all interactions with the researcher were to be video-recorded (Appendix R).
The adapted letter stated that participation in the study was voluntary, the participant could
withdraw from the study without giving a reason and all personal information would remain
confidential and be stored securely at the centre for AAC for 15 years. The adapted reply slip
consisted of written text and pictures (Appendix R).

4.11.1.6. Observer to informed consent

This form included a tick box in which an observer had to tick a yes or no response
related to whether they felt the participant provided voluntary consent without being coerced
into taking part in the study, and understood what was expected of them in the study
(Appendix S). The form included a section in which the observer could include additional
written comments.

4.11.1.7. Ongoing informed consent form

For ease of administration of the ongoing informed consent procedures, the form was
adapted into a booklet (Appendix T) that included picture symbols and options for the
participant to provide verbal, or gestural responses by pointing to symbols to indicate ‘yes’ or
‘no’. The ongoing informed consent form contained questions that reminded the participant
of the purpose of the study, that they could take a break if they felt tired and that they could
stop at any time. It also asked whether they were still happy to continue with the research.

4.11.2. Materials used for screening procedures

4.11.2.1. Biographical questionnaires

The participant’s biographical questionnaire contained information describing the

participant’s personal details, such as age, date of birth, gender, sensory functioning, highest
educational qualification, previous occupation, and marital status (Appendix U). Likewise,
descriptive information from the familiar communication partner was also obtained by means
of a separate biographical questionnaire (Appendix V).

4.11.2.2. Montreal Cognitive Assessment

The Montreal Cognitive Assessment (MoCA) (Nasreddine et al., 2005) was used as a
standardised cognitive screening tool to describe the participant’s dementia severity. The
MoCA is indicated for research in older adults and has demonstrated high reliability (r=0.92),
high concurrent validity (r = 0.87) and high internal consistency (a = 0.83) (Nasreddine et al.,
2005). The MoCA consists of 12 cognitive tasks related to visuospatial/executive

85
© University of Pretoria
functioning, naming, attention, language, abstraction, memory, and orientation. The severity
of dementia was described by calculating a total score range out of 30, with 14-23 indicating
mild dementia and 8-13 indicating moderate dementia (Chang et al., 2012).

4.11.2.3. Manual Ability Classification System

The Manual Ability Classification System (MACS) (Eliasson et al., 2006) was used as
a tool to classify the participant’s ability to use her hands in daily activities. The researcher
asked questions to the familiar conversational partner as well as to care staff at the dementia
care home regarding the participant’s ability to handle objects in everyday activities. This
descriptive information was used to assess the participant’s ability to press the navigational
tabs on the AAC device.

4.11.2.4. Dementia Communication Difficulties Scale

The Dementia Communication Difficulties Scale (DCDS) (Murphy & Gray, 2013) was
used as a simple screening tool to obtain descriptive information about the specific
communication difficulties experienced by the participant. The scale was completed by the
familiar conversational partner, who descriptively rated the participant’s communication
difficulties on a 5-point option scale (never, sometimes, often, always, says too little for me
to know). A score was assigned to 13 statements judged from the familiar conversational
partner’s perspective as follows: 0 (never); 1 point (sometimes); 2 points (often); 3 points
(always) or 4 (says too little for me to judge). The tallied score ranges describe the
communication difficulty in an associated stage of dementia: 0-10.5 (early stage); 11-19.5
(moderate stage); 20-39 (late stage).

4.11.3. Materials used during data collection

4.11.3.1. Procedures Checklist (Baseline Phase)

The Procedures Checklist (Baseline Phase) (Appendix W) consisted of five main steps
related to venue set-up, greetings and introduction to each session, no scaffolding of the life
story interpersonal interaction, closure, and obtaining of the participant’s perspective. A score
of 1 was obtained for a ‘yes’ (step implemented) and 0 for a ‘no’ (step omitted). Each step
had sub-steps which tallied to give a score out of a total of 16 steps. A script for the baseline
phase was embedded in this checklist to ensure the researcher delivered the greetings,
introduction and closure consistently.

86
© University of Pretoria
4.11.3.2. Procedures Checklist (Intervention Phase)

The Procedures Checklist (Intervention Phase) (Appendix X) also comprised five main
steps with specific sub-steps related to equipment and set-up, greetings and introduction,
AAC device familiarisation, application of PC-AAC scaffolding strategies, and obtaining the
participant’s perspective. A semi-structured script for the intervention phase was embedded
in this checklist to guide the researcher’s consistency in the greetings, introduction, and
performing a familiarisation procedure of the AAC device. One point was scored to indicate
‘yes’ (step implemented) or zero for ‘no’ (step omitted). There was a total of 33 steps, 20 of
which were assigned to the researcher’s application of the PC-AAC scaffolding strategies in
the intervention phases. A minimum of five life story sentences were included in each life
story topic. A score was assigned each time the researcher applied the PC-AAC scaffolding
with the use of the electronic life story conversational support.

4.11.3.3. AAC Outcome Measure for Persons with Dementia

The Augmentative and Alternative Communication Outcome Measure for PwD

(AACOM-PwD), which was developed by the researcher (see Par. 3.6.1), consists of four
turn taking domains categorised as Verbal turns (V), Gestural turns (G), Aided turns (A) and
Multimodal turns (MM). Each of the four turn taking domains is divided further into eleven
sub-domains to specify the type (form) of turn taking within the respective domain
(Appendix F2). The AACOM-PwD is used as an observational tool by watching video-
recorded interpersonal interactions between the researcher and participant. One point is
assigned for each of the turn taking responses at a sub-domain level. Scores are then tallied at
a domain level and a total score is tallied for the total turns (Appendix F1). The video-
recordings were utilised by the researcher and the independent rater, for obtaining intra-rater
and inter-rater scores respectively. Further information of the video-recordings and the
interpretation of video-recorded data are found in Par. 4.13.2 and Par. 4.13.3 respectively.

4.11.3.4. Stakeholder Perspective Checklist

The Stakeholder Perspective Checklist (Appendix Y) contained two questions that

helped the researcher to determine the participant’s enjoyment of using the electronic life
story conversational support. The two questions on the form were: Did you enjoy talking
about your life story today? and Did you enjoy talking about the life story a little or a lot?
The questions had corresponding picture symbols on cards to support the participant to
express her responses.

87
© University of Pretoria
4.11.4. Equipment

Video recording equipment included two Canon Legaria HF-R806 digital camcorders
mounted on a tripod stand, back-up SD memory cards and camera batteries. These were used
in the baseline, and intervention phases.

4.12. Phase 3B: Procedures

Data collection commenced once ethical approval (Appendix G1), permission from the
dementia care home (Appendix N) and relevant informed consent from the legal guardian
(Appendix P), participant (Appendix R) and familiar conversational partner (Appendix Q)
had been obtained. An overview of the data collection procedures, and procedural and data
reliability in Phase 3A is schematically depicted in Figure 4.5.

Phase 3B
Ethics approval
Case study participant recruitment

Data collection procedures

Implementation of the co-produced electronic life Reliability:
story conversational support (interpersonal Procedural reliability
interaction between the researcher and • Baseline phase
participant). Use the AACOM-PwD to measure • Intervention phase
turn taking within the interpersonal interaction: Data reliability of scoring
• Baseline phase AACOM-PwD
Without the use of PC-AAC scaffolding • Intra-rater reliability
strategies • Inter-rater reliability
• Intervention Phase
Researcher’s application of PC-AAC
scaffolding strategies
• Stakeholder’s perspective: Participant’s
enjoyment

Figure 4.5: Schematic representation of Phase 3B data collection procedures and

reliability

4.12.1. Data collection procedures

As shown in Figure 4.5, the main purpose of data collection in Phase 3B was to
implement the electronic life story conversational support with the participant in a baseline
phase and later in an intervention phase. In doing so, data collection allowed for the
assessment of procedural reliability of implementing the electronic life story conversational

88
© University of Pretoria
support (Component 2 of the PC-AAC intervention for PwD), as well as for the assessment
of data reliability of scoring turn taking on the AACOM-PwD. Additionally, data on the
participant’s self-reported enjoyment of using the electronic life story conversational support
was obtained. Data collection commenced when the researcher administered the MoCA
(Nasreddine et al., 2005) screening tool with the participant and obtained observational
information using the MACS (Eliasson et al., 2006).
The familiar conversational partner completed her own biographical questionnaire,
assisted in completing the participant’s biographical questionnaire and completed the rating
of the participant’s communication on the DCDS (Murphy & Gray, 2007). Thereafter, the co-
production process of the PC AAC intervention for PwD was conducted according to the
details described in Sections 3.3 and 3.4. All data was collected in the participant’s private
room at the dementia care home. All sessions were video-recorded for data analysis. The
video recording equipment was set up on each day of data collection prior to the participant
entering the room. Two cameras were used so as to obtain footage of front and rear views of
the interpersonal interactions between the participant and the researcher. The participant was
brought into her room by the familiar conversational partner. Upon arrival, the participant
was welcomed and made to feel comfortable before the process of ongoing consent was
followed (see Par. 4.10.1). Data collection followed a specific sequence beginning with three
baseline sessions and was followed by six intervention sessions described in the section
below.

4.12.2. Baseline phase

The baseline phase entailed three consecutive days of data collection before the
intervention began. The researcher and the participant engaged in a 10-minute interpersonal
interaction using her electronic life story conversational support as developed in Par. 3.4. At
baseline, the researcher responded to the participant in a neutral manner (e.g., “Hmm”) and
the use of agreement or disagreement vocabulary was avoided. Furthermore, there was no
familiarisation procedure on the use of the electronic life story conversational support, nor
application of the scaffolded PC-AAC strategies by the researcher. During the baseline phase,
interpersonal interaction was measured and scored with the AACOM-PwD developed in this
study.

89
© University of Pretoria
4.12.3. Intervention with PC-AAC scaffolding strategies

The intervention phase was implemented over six non-consecutive days, with an
interval allowed on every alternative day. In this phase the researcher implemented the
electronic life story conversational support according to the prescribed intervention protocol
outlined in Par. 3.4.2.2. Similar to the baseline phase, interpersonal interaction was measured
and scored using the AACOM-PwD as developed in Par. 3.6.1.

4.12.4. Stakeholder perspective

At the end of each baseline and intervention phase session, the participant’s self-
reported enjoyment of using the electronic life story conversational support was obtained.
The researcher adhered to the Stakeholder Perspective checklist (Appendix Y) and asked the
participant two questions related to her enjoyment of using the electronic life story
conversational support. The researcher supported the conversation by using multimodal AAC
strategies. She pointed to the written text of each question, used pictures to support the
participant’s understanding of the question, and expanded on the participant’s responses that
were supplemented with gestures (e.g., head nodding to indicate yes). The researcher also had
three option cards available with picture symbols indicating ‘yes’, ‘no’ and ‘not sure’ to
which the participant could point to indicate her responses (Appendix Y). These
conversations were video-recorded and transcribed by the researcher to obtain qualitative
data on the participant’s self-reported enjoyment.

4.13. Reliability

Reliability refers to evidence that procedural reliability of implementing the electronic

life story conversational support was maintained in the baseline and intervention phases in
Phase 3B of this study, and that data on the AACOM-PwD was scored reliably by an intra-
rater and inter-rater. A schematic illustration of the reliability of findings in this study was
depicted in Figure 4.4.

4.13.1. Procedural integrity

Procedural integrity was evaluated for the procedures applied in the baseline phase as
well as in the intervention phase. The researcher rated all video recordings for procedural
integrity. An independent speech therapist who served as the inter-rater independently rated
33% (Schlosser, 2003) of sessions in the baseline and intervention phases by completing the
respective procedural checklists (Appendix W and Appendix X).

90
© University of Pretoria
4.13.1.1. Procedural integrity of the baseline phase

To ensure procedural integrity in the baseline phase, the researcher evaluated the
procedures by using a Procedural Checklist (Baseline Phase) (Appendix W). One point was
scored to indicate ‘yes’, the step had been implemented, and a zero was given to indicate the
step had been omitted. There were a total number of 16 steps in each baseline session.
Procedural integrity was expressed as a percentage by using the formula developed by Kuoch
and Mirenda (2003, p. 222):
Number of correct steps executed by the researcher x 100
Total number of possible steps

4.13.1.2. Procedural integrity in the intervention phase

Procedural integrity as adhered to by the researcher was evaluated by marking off the
steps in the Procedural Checklist (Intervention Phase) (Appendix X). This required the
participant to have been familiarised with the AAC device before the life story interpersonal
interaction was implemented and ensured that the PC-AAC scaffolding strategies were
applied by the researcher. Procedural integrity was also calculated by using the above
formula of Kuoch and Mirenda (2003, p. 222).

4.13.2. Data reliability: Scoring the AACOM-PwD

Data reliability for scoring the AACOM-PwD (outlined in Par. 3.6.1) was deemed to be
an important gauge of whether the outcome measure would be a reliable tool to measure turn
taking during interpersonal interaction with a person with dementia. Data reliability was
assessed through the scoring agreement reached at different points in time by the same rater
(intra-rater) and by different raters (inter-raters) (Zapf, Castell, Morawietz, & Karch, 2016).

4.13.3. Intra-rater reliability

Intra-rater reliability was assessed by scoring the AACOM for PwD immediately after
the sessions in the baseline and intervention phases, and again after a one-month interval.
Scoring was performed a month apart to prevent scoring recall and carryover effects. The
researcher watched the same video recordings and scored turn taking in the same way by
using the turn taking coding scheme (Appendix F1) and operational definitions (Appendix
F2) of AACOM for PwD. Raw scores were obtained at both a domain and a sub-domain level
of the AACOM-PwD. The researcher generated two sets of intra-rater reliability raw scores

91
© University of Pretoria
which were compared on a point-by-point rating. Percentage agreement between intra-rating
1 and intra-rating 2 was calculated as follows (Tawney & Gast, 1984):
Number of agreements x 100
Number of agreements + disagreements

A mean percentage of intra-rating agreement was then determined at a domain level

(e.g., verbal turns) and at a sub-domain level (verbal turns with vocalisations).

4.13.4. Inter-rater reliability

An independent rater, a post-doctoral researcher in AAC registered as a Speech-

Language Therapist with the Health Professions Council of South Africa, served as the
independent rater. She had more than fifteen years of clinical experience of working with
children and adults with severe communication disabilities. The independent rater was given
a coding manual with operational definitions of the AACOM-PwD (Appendix F2). The
researcher trained the independent rater on how to perform observational coding procedures
using one video recording from the case study. The independent rater practised the coding
and the researcher answered questions and provided further guidance before final
independent scoring commenced. The rater was provided an electronic version as well as a
hard copy of the AACOM-PwD to perform scoring. She randomly selected 33% of the video
recordings in the baseline phase and 33% of the video recordings in the intervention phase
(i.e., two intervention sessions), to score turn taking with the AACOM-PwD. This met the
recommended criterion of 20-40% for inter-rater reliability set by Schlosser (2003). Inter-
rater reliability was assessed by comparing the scores obtained by the researcher and the
independent rater on the randomly selected video recordings. Percentage agreement was
calculated by using the previously stated calculation by Tawney and Gast (1984) in Par.
4.13.3.

4.14. Data analysis

Raw data generated from procedural integrity and data reliability from scoring the
AACOM-PwD was captured on a Microsoft Excel© 2016 spreadsheet. Although Cohen’s
Kappa was considered to be a suitable statistical measurement of the level of intra-rater and
inter-rater agreement of turn taking data, a case study with one participant did not meet the
minimum sample size of two participants for statistical interpretation of reliability results
(Bujang & Baharum, 2017). Percentage agreement scores were tabulated and described
descriptively. The participant’s self-reported enjoyment elicited from supported

92
© University of Pretoria
conversations with AAC strategies were video-recorded and transcribed by the researcher.
This generated qualitative data on the participant’s self-reported enjoyment which was
described descriptively.

4.15. Summary

This chapter discussed the research methodology employed in this study. The main
aim and sub-aims were presented, followed by an outline of the overall multimethod research
design, and the three research phases. Research Phase 3 (feasibility testing) was next detailed
in two parts in this chapter. Phase 3A detailed the four steps involved in Q-methodology.
Each step was described separately by outlining concourse development with an e-Delphi
panel, selecting a set of statements, the online ranking of the statements, and factor analysis.
Phase 3B described the participant with dementia involved in the case study, the setting in
which in-person data collection took place, the recruitment materials utilised, data collection
procedures followed and reliability in implementing two components of the PC-AAC
intervention for a PwD.

93
© University of Pretoria
 CHAPTER 5: RESULTS

5.1. Introduction

In this chapter, the results of Phase 3 of the main study are presented in two parts,
namely in Phase 3A and Phase 3B. First, a description of terms is presented. Phase 3A reports
on professionals’ viewpoints regarding the four components of the PC-AAC intervention for
PwD as attained from Q-methodology data and analysed using factor analysis (see Figure
5.1). Next, Phase 3B presents the case study results – first for the procedural reliability of
implementing the electronic life story conversational support (component 2 of the PC-AAC
intervention for PwD), and secondly for the data reliability of scoring the AACOM-PwD
(component 4 of the PC-AAC intervention for PwD). Chapter 5 concludes with qualitative
data related to the self-reported enjoyment of a participant with dementia when using the
electronic life story conversational support.

Phase 3: Feasibility testing of the PC-AAC intervention for PwD

Four components of the PC-AAC intervention for PwD:

• Component 1: Person-centred care approach in AAC

• Component 2: Electronic life story conversational support
• Component 3: AAC outcomes for PwD
• Component 4: AAC outcome measurement for PwD

Phase 3A Phase 3B
Q-methodology results Case study results
Professionals’ clinical and educational Implementation of an electronic life story
expertise (viewpoints) obtained conversational support: (sub-aim iv)
qualitatively (sub-aim iii). Reliability
- Procedural reliability
Quantitative data analysis through factor - Data reliability – scoring of AAC
analysis: outcome measure for PwD (AACOM-
• Highest and lowest ranked PwD)
statements • Intra-rater reliability
• Anchor statements • Inter-rater reliability
• Distinguishing statements Stakeholder perspective: Participant with
• Consensus statements dementia
Self-reported enjoyment of interpersonal
interaction with electronic life story
conversational support (sub-aim vi)
Figure 5.1: Schematic illustration of the results chapter

94
© University of Pretoria
5.2. Terms

• Anchor statements: The statements that were ranked as strongly agree (+3) and
strongly disagree (-3) in each factor (Bartlett & DeWesse, 2015).
• By-person factor analysis: Factor analysis condenses the entire set of data into a few
factors (groupings of people) for analysis and interpretation (Valenta & Wigger,
1997).
• Consensus statement: A statement ranked with similar z scores across all factors
(McKeown & Thomas, 2013).
• Distinguishing statement: A statement within a factor that is significantly different
from its rank position in other factors based on its z-score (McKeown & Thomas,
2013). In this study, a distinguished statement is significant at p < .05.
• Eigenvalues: The sum of square factor loadings for each factor assesses the
“statistical strength and explanatory power” of a factor’s variance (Watts & Stenner,
2012, p. 122). Using the Kaiser-Guttman criterion, an eigenvalue greater than 1.00 is
considered to be interpretable.
• Factor: In this study, a factor refers to a group of professionals with shared views
based on their similar rank-order of statements related to the components of the PC-
AAC intervention for PwD developed in this study (Valenta & Wigger, 1997).
• Factor loading: Shows a correlation or relationships between a professional’s
completed Q-sort to the factor onto which they loaded (Newman & Ramlo, 2015).
• Highest- and lowest-ranked statement: The highest-ranked statement is a statement
with the highest z-score, indicating a factor’s strongest agreement (+3). The lowest-
ranked statement is a statement with the lowest z-score, which is ranked as strongly
disagree (-3). The two statements are most useful for interpretation (Newman &
Ramlo, 2015).
• Participant: A person with moderate to severe dementia, who was the participant in
the case study in Phase 3B.
• Professionals: In this study, professionals include AAC researchers, clinicians and
academic teachers who, based on their clinical and/or educational expertise, expressed
their subjectivity on statements related to the components of the PC-AAC intervention
for PwD through the completion of a Q-sort (Phase 3A).
• Q-sorting: Refers to professionals expressing their viewpoints by rank ordering 37
statements (Q-sample) on a forced-choice distribution grid according to a continuum

95
© University of Pretoria
 (i.e., +3 strongly agree, 0 neutral, and -3 strongly disagree) (McKeown & Thomas,
2013). In this study, Q-sorting was performed electronically on QMethod Software
(Lutfallah & Buchanan, 2019), in other words through online Q-sorting.
• Varimax rotation: Refers to the statistical algorithm, which is a statistical procedure
to indicate the maximum variance among the minimum number of factors to enhance
the interpretability of each factor (Newman & Ramlo, 2015).
• Z-scores: Refers to a weighted average score that similar professionals give to a
statement. Z-scores provide information about how strongly each factor engaged with
a statement (Zabala et al., 2018).

5.3. Phase 3A: Q-Methodology results

5.3.1. Factor extraction

The number of factors to be extracted is not prescribed in the literature (Zalaba,

Sandbrook, & Mukherjee, 2018). To employ a systematic and transparent approach towards
determining the number of factors to be extracted, the researcher followed a factor extraction
decision-making protocol as shown in Table 5.1.

Table 5.1: Factor extraction decision-making protocol

Selection criteria Factors
A B C D E F G
1. Default factors generated by QMethod software ✓ ✓ ✓ ✓ ✓ ✓ ✓
2. Eigenvalue >1.0 ✓ ✓ ✓ ✓ ✓ ✓ ✓
3. More than two Q-sorts loaded ✓ ✓ ✓ ✓ ✓ x x
4. Perform factor rotation comparisons
5-factor rotation ✓ ✓ ✓ ✓ x
4-factor rotation ✓ ✓ ✓ x
3-factor rotation ✓ ✓ ✓

As is evident from Table 5.1, four decision-making factor extraction criteria were
applied. Firstly, factor extraction began with seven factors, because QMethod Software
(Lutfallah & Buchanan, 2019) automatically generated seven factors (Factors A-G) as the
default number of factors. These factors represented seven groups of similar viewpoints
derived from 33 completed Q-sorts. The general characteristics of all seven factors were
inspected.
As the second criterion, the eigenvalues of each factor were inspected. Table 5.2 shows
that the eigenvalues for each factor were as follows: Factor A (6.37); Factor B (5.01); Factor
C (4.680); Factor D (3.80); Factor E (2.75); Factor F (2.35); and Factor G (1.58).

96
© University of Pretoria
Table 5.2: Overview of seven default factors initially generated
Initial seven factors
Characteristic A B C D E F G
Eigenvalues 6.37 5.01 4.68 3.80 2.75 2.34 1.58
Number of loading Q-sorts 5 4 3 2 2 1 1
Total explained variance 5 4 3 2 2 1 1

According to the Kaiser-Guttman criterion, an eigenvalue greater than 1.00 is

considered to be interpretable (Watts & Stenner, 2012; Yang, 2016). All seven factors
satisfied this second criterion and were retained. While eigenvalues greater than 1.00 may be
statistically interpretable, they do not provide meaningful information on the significance of
each factor in providing an understanding of professional’s views on the components of the
PC-AAC intervention for PwD (Watts & Stenner, 2012). Therefore, the third criterion was
applied. The number of factors with two or more loadings was retained (Watts & Stenner,
2005). As shown in Table 5.2, Factors F and G were therefore eliminated, leaving five
potential factors to be extracted.

5.3.2. Factor rotations

To determine whether five factors offered the most variance, factor rotations with five,
four and three factors were performed and compared on QMethod Software (Lutfallah &
Buchanan, 2019). This was done iteratively before a decision on the final numbers of factors
was made. Table 5.3 shows a comparison between the 5-factor, 4-factor and 3-factor rotation.

Table 5.3: Factor rotation comparisons

Factor Total Total number Decision

rotation explained of participants
variance loaded
5-factor 58% 19 Removed: 58% variance explained, with the fewest
number of professionals’ views represented.
4-factor 67% 22 Removed: Although 67% of variance was explained, a
smaller number of participants was represented compared
to a 3-factor rotation.
3-factor 79% 26 Retained: The highest percentage of variance, with the
minimum number of factors (Watts & Stenner, 2005).

As shown in Table 5.3, a 3-factor solution yielded the highest explained variance with
the greatest number of professionals loaded on a minimum number of factors (Watts &
Stenner, 2005). Hence, three factors were extracted for the interpretation of the results.

97
© University of Pretoria
5.3.3. Characteristics of three extracted factors

The three final factors are shown in Table 5.4. The following characteristics of each
factor were examined: average reliability co-efficient; number of loading Q-sorts;
eigenvalues; total variance explained; composite reliability; and standard error of z-scores.

Table 5.4: Characteristics of the three extracted factors

Characteristics Factor A Factor B Factor C
Average reliability coefficient 0.8 0.8 0.8
Composite reliability 0.98 0.97 0.97
Standard error of factor scores 0.16 0.16 0.19
Eigenvalues 8.5 7.2 6.8
Number of loading Q-sorts 10 9 7
Total explained variance 10 (38%) 9 (35%) 7 (27%)

The average reliability co-efficient is an estimate of test-retest reliability, which

indicates that professionals completing successive Q-sorts would respond in a similar
manner. In Q-methodology, this score is set at 0.8 and a composite reliability score higher
than 0.8 is an estimate of good factor reliability (Zabala & Pascual, 2016). The composite
reliabilities of 0.98 (Factor A), 0.97 (Factor B) and 0.96 (Factor C) are above 0.8. This
suggests that the professionals’ views defined in each factor are reliable (Brown, 1980).
As shown in Table 5.4, the high composite reliability scores and lower standard error of
factor scores signify the level of consistency between the three factors (Watts & Stenner,
2012). Overall, the three factors represented the viewpoints of 26 professionals (79%) out of
the 33 who had completed their Q-sorts. This means that 26 professionals loaded onto one of
three factors, while seven professionals (21%) did not load onto any factor. Of the 26
professionals, 10 expressed similar viewpoints on the components of the PC-AAC
intervention for PwD and were loaded onto Factor A. The total explained variance of
viewpoints in Factor A was 10 (38%). Likewise, nine professionals loaded onto Factor B
(35% variance), and seven onto Factor C (27% variance).
In each of the three factors, the significant factor loadings were automatically flagged
on QMethod Software (Lutfallah & Buchanan, 2019) and appear in boldface in Table 5.5.

98
© University of Pretoria
Table 5.5: Factor loadings for the three rotated factors
Professional ID Factor A (10) Factor B (9) Factor C (5)
P654 0.745 X 0.389 0.29
P664 0.403 0.395 0.71 X
P665 0.384 0.612 X 0.40
P666 0.432 0.616 X 0.26
P675 0.200 0.754 X 0.22
P688 0.193 0.347 0.77 X
P700 0.727 X 0.084 0.40
P712 0.350 0.096 0.71 X
P718 0.641 X 0.439 0.33
P724 0.774 X 0.343 0.22
P746 0.557 0.608 X 0.20
P773 0.488 0.492 0.34
P816 0.232 -0.021 0.51 X
P846 0.752 X 0.250 0.42
P856 0.180 0.755 X 0.43
P858 0.537 0.342 0.55
P860 0.385 0.551 0.57
P869 0.589 0.509 0.46
P935 0.296 0.631 X 0.47
P002 0.565 X 0.309 0.36
P017 0.567 0.372 0.43
P068 0.249 0.119 0.80 X
P070 0.619 X 0.270 0.38
P121 0.726 X 0.248 0.49
P133 0.119 0.427 0.68 X
P139 0.084 0.802 X -0.14
P166 0.620 X 0.438 0.25
P175 0.383 0.211 0.62 X
P216 0.385 0.463 0.51
P285 0.377 0.720 X 0.23
P296 0.363 0.495 X 0.31
P303 0.826 X 0.153 0.14
P345 0.595 0.602 0.15
Boldface X: Factor loadings greater than 0.42 (significant) showing a defining Q-sort.
Grey: Confounded (significantly loaded on two or more factors greater than 0.42) i.e., significance level p>.05.

Table 5.5 shows the rotated factor matrix in which significant factor loadings (either
positive or negative) were greater than 0.42. A defining factor loading is presented in
boldface and is considered to be most representative of the specific factor onto which it has
been loaded. This means that 10 professionals’ Q-sorts loaded significantly onto Factor A.
For example, the first professional (i.e., P654) had factor loadings of 0.745 on Factor A, 0.39
on Factor B, and 0.29 on Factor C. This professional’s highest loading was on Factor A and it
was therefore a defining Q-sort in Factor A.

99
© University of Pretoria
 In the case of Table 5.5, the Q-sorts of seven professionals (21%) loaded significantly
on more than one factor. For example, although professional P773 (highlighted in grey in
Table 5.5) loaded significantly on two factors, namely 0.488 (Factor A), 0.492 (Factor B),
their expressed views did not define a specific factor and is considered to be confounded.
Therefore, all seven confounded Q-sorts were excluded from the factor interpretation.
In the next section, the results of each factor are described and interpreted individually.
In each factor of the three factors, the professionals who shared similar views points are
described first. Then, the statements in each factor are examined in four ways:
i. The highest and lowest ranked statements by z-scores
ii. Anchor statements, i.e., the statements ranked at extreme ends of the ranking scale (i.e.,
+3 strongly agreed, and -3 strongly disagreed)
iii. Distinguishing statements, i.e., statements that were statistically significant at a p-value
< .05 as a unique view in each factor
iv. Consensus statements, i.e., statements that are ranked in a similar way across all factors
with a non-significant difference in z-scores at a p-value > .05
At the end of the section, a summary of the three factors is provided, and each factor is
given a label based on its most distinguishing characteristics (Zabala et al., 2016).

5.3.4. Factor interpretation

5.3.4.1. Factor A

Ten professionals ranked the statements on their Q-sorts in a similar way and together
comprised Factor A. This factor represents the largest professional grouping with an
explained variance of 10 (38%) (see Table 5.6). The professional (P303) with the highest
factor loading (0.826) in Factor A was a Speech-Language Pathologist with specialisation in
dementia studies, as well as an interest in person-centred care in adults with neurologically
acquired communication disorders.

100
© University of Pretoria
Table 5.6: Factor A – Description of professionals
Professional Factor Profession, Type of work Focus of Area of interest
*#
loading specialisation work ∆
P654Fd 0.745 SLP (AAC) Research Adults PCCv
P700Fe 0.727 Other (AAC) Research, teaching CWDD Interaction
P718Fe 0.565 SLP (Dementia) Research, clinical, Adults Dementia
teaching communication
P724Fc 0.774 SLP (Dementia) Research, clinical, Adults Interaction
teaching
P846Fc 0.752 SLP (AAC) Research, clinical, Adults Interaction
teaching
P002Me 0.565 SLP (AAC) Research, teaching Adults Interaction
P070Fb 0.619 SLP (AAC) Clinical Adults PCCv
P121Fb 0.726 SLP (AAC) Research Both Instrument
development
P166Fd 0.620 SLP (AAC) Other Both PCCv
P303Fe 0.826 SLP (Dementia) Research, teaching Adults PCCv
Participant ID: participants’ non-identifiable code followed by their *gender (F denotes Female. M denotes
Male); #years of experience: a (0-5 years), b (6-10 years), c (11-15 years), d (16-20 years), e (> 21 years).
∆Focus of work: Adults (adult neurologically acquired communication disorders); CWDD (children with
developmental disorders); Both (adults and CWDD).
v
PCC (Person-Centred Care).

As displayed in Table 5.6, all but one of the professionals in Factor A were Speech-
Language Pathologists (SLPs) who mainly specialised in AAC or in dementia studies. One
professional classified their profession as a researcher who specialised in AAC. The
viewpoints expressed in Factor A were principally from professionals involved in research
(2), research and teaching (3), or a combination of research, teaching and clinical work (3).
While one professional was involved only in clinical work, another professional was involved
in neither clinical work nor research, but focused on assistive technology and AAC
consultation for adults and children with communication disorders. The nine females and one
male in Factor A all had a range of years’ (6- 21 years) worth of experience, mainly in adults
with neurologically acquired communication disorders. Professionals in Factor A were
equally interested in person-centred care (4) and interpersonal interaction generally (4),
followed by instrument development (1) and dementia-related communication specifically
(1).

101
© University of Pretoria
Table 5.7: Factor A – Highest- and lowest-ranked statements

Statement Statement
# z-score Ranking
Highest-ranked statements
24 Persons with dementia should be supported in authoring the content 1.7035 +3
of their life story based conversational supports.
28 Successful interaction should be rated from the perspectives of both 1.5494 +3
the person with dementia (patient-reported) and the conversational
partner (proxy-reported).
30 Social connections are important goals of AAC interventions and 1.4888
designed to support interpersonal interaction with a person with +3
dementia.
23 Conversational supports with personalised life story content of a 1.3202 +3
person with dementia is a useful scaffold for reminiscence-based
conversations with a person with dementia.
Lowest-ranked statements
31 Enjoyment and motivation (as engagement outcomes) within an -1.4898 -3
interpersonal interaction are not important to include in an outcome
measure for persons with dementia.
36 It is not necessary to include participation outcomes (frequency of -1.5845 -3
taking part in everyday interactions) in an AAC interaction outcome
measure for persons with dementia.
27 Person-centred care as a philosophy of care (e.g., one that promotes -1.6076 -3
the independence of persons with dementia and values their life
history) in AAC interventions is not clinically achievable.
29 It is not necessary to measure the satisfaction of the person with -1.6605 -3
dementia in using AAC strategies within an interpersonal interaction.
Grey coding indicates the statements ranked extremely high (at +3) and low (at -3) in Factor A.

As shown in Table 5.7, statement 24 was ranked the highest (z-score of 0.7035), which
reflects the views of Factor A on the person-centred care component of the PC-AAC
intervention for PwD. Statement 29 (z-score of -1.6605) was the lowest ranked statement that
reflected Factor A’s views on AAC outcome measurement for PwD. Professionals in Factor
A strongly agreed with statements related to person-centred care components on personalised
content of life-story-based conversational supports (statements 24 and 23), measuring the
perspectives of PwD and their conversational partners (statement 28), and social connections
as AAC outcomes for PwD (statement 30).
Factor A showed strongest disagreement with statements on AAC outcome
measurement (statement 29), person-centred care as a philosophy of care in AAC
interventions for PwD (statement 27), and in two statements concerned with AAC outcomes
for PwD (statements 36 and 31). Notably, the negative z-scores of these statements do not
infer a negative opinion. Instead, as shown in Table 5.7, they denote a negatively worded
statement. The strong disagreement with these negative worded statements indicates that

102
© University of Pretoria
professionals expressed their support to the contrary wording of the lowest-ranked
statements.
Five statements were found to be statistically significant and distinguished in Factor A
(see Table 5.8). The distinguished statements are listed in a way that shows their composite
factor ranking, rather than the numerical order of statements.

Table 5.8: Factor A – Distinguished statements

Statement Statement z-scores and factor
# Factor A rankings
Distinguished statements * A B C
1 Electronic AAC systems (e.g., tablet computers with applications -1 0 0
[apps]) can be useful to support interpersonal interaction between a (-0.498) (0.420) (0.20)
conversational partner and a person with moderate to severe
dementia.
34 The most important domain to measure in an AAC interaction -1 -2 -2
outcome measure for persons with dementia is turn taking (e.g., (-0.247) (-0.993) (-1.39)
frequency of turn taking exchanges).
13 By repeating the utterances of the person with dementia, a 0 -1 -1
conversational partner shows an understanding of what the person (0.118) (-0.803) (-0.53)
with dementia is expressing within an interpersonal interaction.
35 An AAC interaction outcome measure must include the modality of +2 -1 -1
turn taking (e.g., turn taking with gestures). (0.872) (-0.113) (-0.55)
24 Persons with dementia should be supported in authoring the content +3 0 1
of their life-story-based conversational supports. (1.704) (0.317) (0.80)
*Significant at p<.05

As displayed in Table 5.8, the group of professionals comprising Factor A was

distinguished by their prominent views on five statements. Factor A shared a clearly defined
and distinguished agreement that PwD should be supported in authoring the content of their
own life story conversational supports (statement 24). However, they did not agree that the
use of electronic AAC systems is useful for interpersonal interaction in persons with
moderate to severe dementia (statement 1). Furthermore, Factor A was differentiated in their
neutrality towards repeating of utterances as a useful conversational partner strategy to aid
understanding in an interpersonal interaction with a PwD (statement 13). Professionals in
Factor A had two defined views on an AAC outcome measure for PwD. They were of the
view that an AAC outcome measure for PwD must include measurement of the modality of
turn taking (statement 35). This view was validated as they were more inclined to agree that
the frequency of turn taking was the most important measurement domain.

103
© University of Pretoria
 5.3.4.2. Factor B

Nine professionals made up Factor B and are described in Table 5.10. The professional
(P139) with the highest factor loading of 0.802 was a Speech-Language Pathologist with
specialisation in technology, as well as an interest in person-centred care for both adults and
children with disabilities.

Table 5.9: Factor B – Description of professionals

Professional*# Factor Profession, Type of work Focus of Area of

loading specialisation work ∆ interest
P665Fa 0.612 SLP (AAC) Clinical Adults Interaction
P666Fe 0.616 SLP (AAC) Research, clinical, Adults PCCv
teaching
P675Fe 0.754 SLP (AAC) Research, clinical, CWDD Interaction
teaching
P746Fe 0.608 OT (AAC) Research, teaching Both Other
P856Fb 0.755 SLP (AAC) Clinical Adult PCCv
P935Fb 0.631 Psychologist Research Adult Instrument
(AAC) development
P139Fd 0.802 SLP Research, clinical, Both PCCv
(Technology) teaching
P285Fb 0.720 SLP (AAC) Research, teaching CWDD Interaction
P296Fd 0.495 SLP (Other) Clinical Both Interaction
Participant ID: participants’ non-identifiable code followed by their *gender (F denotes Female. M denotes
Male); and #years of experience: a (0-5 years), b (6-10 years), c (11-15 years), d (16-20 years), e (> 21 years).
∆Focus of work: Adults (adult neurologically acquired communication disorders); CWDD (children with
developmental disorders); Both (adults and CWDD).
v
PCC (Person-Centred Care).

As reported in Table 5.9, Factor B represented a multidisciplinary group of female

professionals with seven SLPs, one Occupational Therapist and one Psychologist. Five SLPs
specialised in AAC, one in technology and one described their specialisation as ‘other’ (i.e.,
communication interventions). Four SLPs were interested in interpersonal interaction and
three in person-centred care through their research, teaching and clinical work with either
adults and/or children with disabilities. The Occupational Therapist specialised in AAC
research and teaching focused on the participation of adults and children with communication
disabilities. The Psychologist was an AAC researcher with an interest in instrument
development in adults with neurologically acquired communication disorders.
Factor B was represented by the greatest number of clinicians (3) whose clinical work
focused on adults with neurologically acquired communication disorders (2), or both adults
and children with disabilities (1). Professionals in Factor B had diverse work experience,

104
© University of Pretoria
ranging from one early-career clinician (0-5 years) to three well-experienced professionals
with more than 21 years of research, teaching and clinical experience. The highest and lowest
ranked statement in Factor B based on their z-scores are shown in Table 5.10, along with the
statements ranked at the extremes with strong agreement (+3), and strong disagreement (-3).

Table 5.10: Factor B – Highest- and lowest-ranked statements

Statement Statement
# z-score Ranking
Highest-ranked statements
26 A patronising style of language (e.g., an exaggerated tone of voice 1.6875 +3
typical of talking to an infant) used during interpersonal interaction
with a person with dementia undermines personhood of a person
with dementia.
15 A person with dementia must be supported in decision-making in all 1.4776 +3
aspects of their AAC intervention.
18 AAC interventions to support interpersonal interaction in persons +3
with dementia must be tailored to the strengths of the person with 1.2946
dementia.
16 A person with dementia must be consulted about their personal 1.2124 +3
preferences for interpersonal interaction.
Lowest-ranked statements
29 It is not necessary to measure the satisfaction of the person with - 1.5611 -3
dementia in using AAC strategies within an interpersonal
interaction.
27 Person-centred care as a philosophy of care (e.g., one that promotes - 1.6944 -3
the independence of persons with dementia and values their life
history) in AAC interventions is not clinically achievable.
31 Enjoyment and motivation (as engagement outcomes) within an -1.7593 -3
interpersonal interaction are not important to include in an outcome
measure for persons with dementia.
37 It is not important to include the behaviour displayed by persons -1.7984 -3
with dementia (e.g., their level of frustration, agitation or
avoidance) as an AAC interaction outcome measure for persons
with dementia.
Grey coding indicates the statements ranked extremely high (at +3) and low (at -3) in Factor B.

As shown in Table 5.10, statement 26 was the highest-ranked statement by Factor B

with a z-score of 1.6875. The lowest-ranked statement with a z-score of -1.7984 (statement
37) concerned the lack of importance of including a behaviour domain in an AAC interaction
outcome measure for PwD. Professionals in Factor B expressed their strongest agreement
with statements related to person-centred care components of the PC-AAC intervention
package for PwD (statements 26, 15, 18 and 16) and were ranked as +3 (strongly agree) on
the ranking scale. Factor B expressed their strongest disagreement with negatively worded
statements related to a behavioural domain of an AAC outcome measure for PwD (statement
37), person-centred care in AAC interventions (statement 27) and measurement of

105
© University of Pretoria
engagement in and satisfaction in respect of an AAC interpersonal interaction with a PwD
(statements 31 and 29). Factor B had eight statements that distinguished their views from
those of Factor A and C (see Table 5.11).

Table 5.11: Factor B – Distinguished statements

Statement Statement z-scores and factor

# Factor B rankings
Distinguished statements* A B C
36 It is not necessary to include participation outcomes -1 -2 -2
(frequency of taking part in everyday interactions) in an (-1.584) (-0.990) (-1.69)
AAC interaction outcome measure for persons with
dementia.
5 Non-electronic AAC systems (e.g., paper-based -1 -1 -2
communication books with pictures) are the most beneficial (-0.905) (-0.123) (-0.89)
communication supports for interpersonal interaction with a
person with mild to moderate dementia.
4 Legacy messages (i.e., unique words or phrases used by a 0 0 -1
person with dementia digitally recorded in their own voice (-0.129) (0.447) (-0.24)
in the early stages of dementia) stored as voice output on
electronic devices (e.g., iPad) are useful to facilitate social
connection with a conversational partner as dementia
severity progresses.
30 Social connections are important goals of AAC +3 0 +3
interventions and are designed to support interpersonal (1.489) (0.111) (1.63)
interaction with a person with dementia.
6 Combining picture pointing with spontaneous speech (i.e., 0 1 -1
augmented input) is a useful strategy to support auditory (-0.085) (0.582) (-0.40)
comprehension of spoken language in persons with
dementia.
9 The unique set of communication behaviours of a person -1 +2 0
with severe dementia (e.g., vocalisations, facial (-0.225) (1.161) (0.22)
expressions, gestures) should be used as the basis for
interpersonal interaction with a conversational partner.
16 A person with dementia must be consulted about their +1 +3 0
personal preferences for interpersonal interaction. (0.475) (1.212) (0.27)
26 A patronising style of language (e.g., an exaggerated tone +2 +3 +2
of voice typical of talking to an infant) used during (0.877) (1.688) (0.95)
interpersonal interaction with a person with dementia
undermines the personhood of a person with dementia.
*Significant at p<.05

As shown in Table 5.11, Factor B distinguished their views regarding the use of
specific AAC strategies and techniques in interpersonal interaction with persons with
dementia. Factor B felt that augmented input was a useful AAC strategy to aid
comprehension in persons with dementia (Statement 6). While they slightly disagreed that
non-electronic AAC systems were the most beneficial support for persons with mild to
moderate dementia (Statement 5), they also affirmed their agreement that the unique set of

106
© University of Pretoria
unaided communication behaviours should be used as a basis for interpersonal interaction
with persons with severe dementia (Statement 9). Although neutral in their view on legacy
messages (Statement 4), they statistically distinguished themselves in their collective view on
the possible usefulness of using digitally recorded legacy messages as a voice output in the
own voice of the person with dementia.
Factor B was distinguished by their neutral view towards social connection as a goal in
AAC interventions for persons with dementia (Statement 30). However, they inferred their
agreement to participation outcomes on an AAC outcome measure in persons with dementia
(Statement 36). They also strongly believed that persons with dementia must be consulted
about their personal preferences for interpersonal interaction (Statement 16) and that
personhood is undermined (Statement 26) through a patronising language style during
interpersonal interactions with persons with dementia.

5.3.4.3. Factor C

The viewpoints in Factor C were defined by seven professionals (two males and five
females) and are next described in Table 5.12.

Table 5.12: Factor C – Description of professionals

Professional*# Factor Profession, Type of work Focus of Area of interest
Loading specialisation work ∆
664Fd 0.71 SLP (AAC) Research, Both Interaction
clinical, teaching
688Me 0.77 Other (AAC) Research, Both Interaction
teaching
712Fe 0.71 SLP (Dementia) Research, clinical Adults Dementia
Communication.
816Fe 0.51 SLP (AAC, Research Both Interaction
Dementia &
Technology)
068Me 0.80 Psychologist Research, Adults PCC
(Other) clinical, teaching
133Fe 0.68 SLP (AAC) Clinical CWDD Interaction
175Fe 0.62 SLP (AAC) Research, CWDD Interaction
teaching
Participant ID: Participants’ non-identifiable code followed by their *gender (F denotes Female; M denotes
Male) and #years of experience: a (0-5 years); b (6-10 years); c (11-15 years); d (16-20 years); e (> 21 years).
∆Focus of work: Adults (adult neurological acquired communication disorders); CWDD (children with
developmental disorders); Both (adults and CWDD)
v
PCC (Person-Centred Care)

As displayed in Table 5.12, the professional (P068) with the highest factor loading
(0.80) was a psychologist with expertise in disability studies and an interest in person-centred
care in adults with neurologically acquired communication disorders. Collectively,

107
© University of Pretoria
professionals in Factor C were all well-experienced, with the majority having more than 21
years of work experience individually, and one with 16-20 years’ experience. There were five
SLPs, specialising in AAC, dementia studies or in a combination of AAC, dementia and
technology. These SLPs were primarily interested in interpersonal interaction in general – in
both adults and children with disabilities. Only one of them was a clinician, while the work of
the other six professionals focused on a combination of research, clinical work or teaching.
The highest- and lowest-ranked statements in Factor C obtained a z-score of 1.6284
(Statement 30) and -1.7324 (Statement 29) respectively (see Table 5.13).

Table 5.13: Factor C: Highest- and lowest-ranked statements

Statement Statement
# z-score Rank
Highest-ranked statements
30 Social connections are important goals of AAC interventions designed 1.6284 +3
to support interpersonal interaction with a person with dementia.
8 A conversational partner should never correct off-topic utterances made 1.2984 +3
by a person with dementia.
18 AAC interventions to support interpersonal interaction in persons with 1.1941 +3
dementia must be tailored to the strengths of the person with dementia.
23 Conversational supports with personalised life story content of a person 1.0734 +3
with dementia is a useful scaffold for reminiscence-based conversations
with a person with dementia.
Lowest-ranked statements
31 Enjoyment and motivation (as engagement outcomes) within an -1.4720 -3
interpersonal interaction are not important to include in an outcome
measure for persons with dementia.
27 Person-centred care as a philosophy of care (e.g., one that promotes the -1.6684 -3
independence of persons with dementia and values their life history) in
AAC interventions is not clinically achievable.
36 It is not necessary to include participation outcomes (frequency of -1.6921 -3
taking part in everyday interactions) in an AAC interaction outcome
measure for persons with dementia.
29 It is not necessary to measure the satisfaction of the person with -1.7324 -3
dementia in using AAC strategies within an interpersonal interaction.
Grey coding indicates the statements ranked extremely high (at +3) and low (at -3) in Factor C.

As revealed in Table 5.13, the highest-ranked statement reflected Factor C’s view on
social connection as an important AAC goal in interpersonal interaction (Statement 30). The
lowest ranked statement in Factor C was Statement 29 (z-score of -1.7324). This statement
expressed Factor C’s views on the necessity of AAC outcome measurement relating to the
level of satisfaction of a person with dementia with an interpersonal interaction.

108
© University of Pretoria
 The professionals grouped in Factor C expressed their strongest agreement in respect of
statements aligned with AAC outcome measurement (Statement 30), conversational partner
strategies (Statement 8), personalisation of conversational supports (Statement 23) and
person-centred care in AAC interventions for persons with dementia (Statement 18). They
expressed their strong disagreement with negatively worded statements related to
participation and engagement outcomes of interpersonal interaction with a person with
dementia (Statements 36 and 31 respectively), as well as person-centred care (Statement 27).
Factor C’s viewpoints were defined by four distinguished statements shown in Table 5.14.

Table 5.14: Factor C – Distinguished statements

Statement Factor
# Factor C
Distinguished statements* A B C
20 Adapted participatory methods (e.g., co-production) should be +1 +2 0
employed to include persons with dementia as collaborators in (0.705) (0.967) (0.17)
the co-design of AAC interventions.
21 Working collaboratively with the family of a person with -2 -2 +1
dementia and other stakeholders throughout the AAC (-1.296) (-1.271) (0.91)
intervention process is important, but not necessary.
11 A conversational partner should display empathy by mirroring 0 0 +2
the emotion of the person with dementia to facilitate
interpersonal interaction with a person with dementia. (-0.092) (0.049) (0.99)
8 A conversational partner should never correct off-topic -1 -1 +3
utterances made by a person with dementia. (-0.936) (-0.539) (1.30)
*Significant at p<.05

As displayed in Table 5.14, Factor C had distinguished views on the role of a

conversational partner in supporting interpersonal interaction with a person with dementia.
These professionals felt strongly that a conversational partner should never correct the off-
topic utterances made by a person with dementia (Statement 8) and should display empathy
to mirror the emotion of a person with dementia (Statement 11). Factor C felt that although
collaboration with persons with dementia and other relevant stakeholders was important, it
was not necessary (Statement 21). They further expressed a neutral view towards adapted
participatory methods that involve persons with dementia as co-designers of their own AAC
intervention (Statement 20).
Before proceeding to the consensus statements across all three factors, a summarised
overview of each factor is provided in Table 5.15. The summary presents the composite z-
score of each of the three factors’ highest- and lowest-ranked statement, and statements that
notably distinguished each factor.

109
© University of Pretoria
 Table 5.15: Summary of Factors: Factors A, B and C

Factor description Highest-ranked Lowest-ranked Distinguished statements

statements (+3) statements (-3)
Factor A 24. Persons with 29. It is not • Electronic AAC systems can be useful to • Persons with dementia should be
Person-centred dementia should be necessary to support interpersonal interaction (-1) supported in authoring the content of
authorship with a focus supported in measure the • By repeating the utterances of the person their life story …(+3)
on directly reported authoring the satisfaction of the with dementia, a conversational partner • The most important domain to measure
outcomes and turn content … person with shows an understanding (0) … is turn taking (-1)
taking outcome dementia … • An AAC interaction outcome measure
measures must include ... modality of turn taking
(+2)

Factor B 26. A patronising 37. It is not • Legacy messages … are useful to • The unique set of communication
Person-centred language style of language important to facilitate social connection …(0) behaviours of a person with severe
style with a focus on … undermines include the • Non-electronic AAC systems … are the dementia … should be used … for
AAC strategies and personhood … behaviour most beneficial … (-1) interpersonal interaction (+2)
techniques displayed by • Augmented input is a useful strategy to • A person with dementia must be
persons with support auditory comprehension ...(+1) consulted … personal preferences (+3)
dementia … as an • Social connections are important goals of • It is not necessary to include participation
AAC interaction AAC interventions (0) outcomes…on an AAC outcome
outcome ... measure…(-2)

Factor C 30. Social 29. It is not • A conversational partner should never • Adapted participatory methods … should
AAC outcomes and connections are necessary to correct off-topic utterances …(+3) be employed to include persons with
outcome measures with important goals of measure the • A conversational partner should display dementia … in the co-design of AAC
a focus on empathetic AAC interventions satisfaction of the empathy by mirroring the emotion of the interventions (0)
interpersonal ... person with person with dementia … (+2) • Working collaboratively with the
interactions dementia … family… and other stakeholders ... is
important, but not necessary (+1)
- 3 (strongly disagree), 0 (neutral), +3 (strongly agree)

110
© University of Pretoria
5.3.5. Consensus statements

Consensus statements are statements that were scored similarly in two or more factors
and were non-significant at a p-value <.05. Overall, there were 12 consensus statements (see
Table 5.16), and complete consensus was reached on five of them, which were negatively
worded statements (i.e., Statements 3, 22, 27, 29 and 31). This meant that all three factors
expressed their strongest agreement with the notion that person-centred care as a philosophy
of care for persons with dementia was clinically achievable (Statement 27). They also
strongly agreed on the necessity of measuring the satisfaction of a person with dementia in
using AAC strategies within an interpersonal interaction (Statement 29). All factors indicated
that enjoyment and motivation of persons with dementia within an interpersonal interaction
(Statement 31) are important AAC outcomes to measure. All factors disagreed that the
immediate interactional context was not important to consider when designing AAC
interventions for persons with dementia (Statement 22). Additionally, all factors disagreed
with Statement 3 in that intelligent assistive technologies should be not explored to support
interpersonal interaction in persons with dementia.
Between Factors A and C, there was consensus of agreement on three statements.
These two factors held a strong view that AAC strategies for persons with dementia are
flexible, depending on the stage and type of dementia (Statement 17), and that pausing to
give a person with dementia time to respond is important within an interpersonal interaction
(Statement 14). They also agreed that personalised music and digital photographs on
electronic devices increased enjoyment in interpersonal interaction with a person with
dementia (Statement 25).
Similarly, there was consensus of agreement on three statements between Factors B and
C. These factors disagreed to a lesser extent that it was not necessary to train persons with
dementia in navigating electronic AAC systems before their use in real-time interpersonal
interaction (Statement 2). Factors B and C held a strong view that AAC interventions should
be tailored to the strengths of persons with dementia (Statement 18), but they did not have a
definitive view on whether interpersonal interaction was analysed best through qualitatively
observations (Statement 32).
The last consensus statement was shared by Factors A and B, which revealed their
agreement that involving persons with dementia in co-designing their own interventions
promoted self-determination (Statement 19).

111
© University of Pretoria
Table 5.16: Consensus statements across factors
Statement Consensus statements Factors
# A B C
2 Training persons with dementia on the use of electronic AAC systems (e.g., navigating on tablet computers -2 -1 -1
with apps) before they use them in real-time interaction settings is not necessary when the purpose is (-1.194) (-0.877) (-0.74)
interpersonal interaction.
3 The use of intelligent assistive technologies (e.g., artificial intelligence) to support communication and -2 -2 -2
interpersonal interaction in persons with dementia should not be explored for persons with dementia. (0.0335) (0.223) (0.190)
14 A conversational partner can support interpersonal interaction by pausing to give a person with dementia 2 1 2
time to respond. (0.3685) (-0.084) (-0.452)
17 AAC strategies for a person with dementia are flexible, depending on the stage and type of dementia. 2 1 2
(0.2214) (-0.059) (-0.281)
18 AAC interventions to support interpersonal interaction in persons with dementia must be tailored to the 2 3 3
strengths of the person with dementia. (1.250) (1.295) (1.19)
19 The self-determination of a person with dementia is promoted by facilitating their involvement in co- 1 1 0
designing their own AAC interventions. (0.501) (0.644) (0.42)
22 The immediate interaction context (e.g., the time of day in which an interaction occurs) does not play a -2 -2 -2
vital role in the success of interpersonal interaction with a person with dementia. (-1.036) (-1.036) (-1.13)
25 Life-story-based conversational supports that include personalised music and digital photographs on 1 2 1
electronic devices (e.g., iPad) enhance the enjoyment of interpersonal interaction for persons with (0.865) (0.854) (0.92)
dementia.
27 Person-centred care as a philosophy of care (e.g., one that promotes the independence of persons with -3 -3 -3
dementia and values their life history) in AAC interventions is not clinically achievable. (1.608) (-1.694) (-1.67)
29 It is not necessary to measure the satisfaction of the person with dementia in using AAC strategies within -3 -3 -3
an interpersonal interaction. (-1.661) (-1.561) (-1.73)
31 Enjoyment and motivation (as engagement outcomes) within an interpersonal interaction are not important -3 -3 -3
to include in an outcome measure for persons with dementia. (-1.490) (-1.759) (-1.47)
32 AAC-supported interaction in persons with dementia is best analysed qualitatively from observation of -1 0 0
spontaneous interaction. (-0.163) (0.017) -0.15
Grey coding shows consensus between two factors.

112
© University of Pretoria
5.4. Phase 3B: Case study results

5.4.1. Reliability

In this section, the reliability results are presented first for procedural integrity of
implementing procedures in the baseline and intervention phases, and secondly for reliability
of scoring the AACOM-PwD (intra-rater reliability and intra-rater reliability).

5.4.1.1.1. Procedural reliability

Procedural reliability was evaluated to provide evidence that the procedures in the
baseline and intervention phases were implemented in a reliable and consistent manner as
intended. The results are presented next.

5.4.1.1.2. Procedural integrity in the baseline and intervention phase

Procedural integrity in the baseline phase was scored using the Procedural Checklist:
Baseline Phase (Appendix W). There were 16 procedural steps in each baseline session. One
point was scored for each procedural step implemented and a zero if omitted. The raw score
out of 16 was converted to a percentage and is reflected in Table 5.17.

Table 5.17: Procedural reliability percentages – Baseline and intervention phases

Phase Baseline Intervention
Session 1 2 3 4 5 6 7 8 9
Procedural integrity
Researcher 100% 94% 94% 100% 97% 100% 100% 100% 100%
Mean 98.3%
Independent rater 94% 100% 100%

As shown in Table 5.17, the researcher’s procedural integrity in all three baseline phase
sessions (sessions 1, 2, 3) ranged from 94% to 100%. Procedural integrity of the intervention
phase sessions (sessions 4, 5, 6, 7, 8, 9) was measured using the Procedural Checklist:
Intervention Phase (Appendix X). In total, there were 33 procedural steps in the intervention
phase. The raw scores were converted to a percentage and are shown in Table 5.17.
Procedural integrity in the intervention phase ranged from 97% to 100%. Overall, a high
mean procedural reliability percentage of 98.3% was obtained. This is considered to be good
reliability and indicates that the baseline and intervention procedures were executed with
high consistency and reliability (McMillan & Schumacher, 2014).

113
© University of Pretoria
 The independent rater followed the same scoring instructions as the researcher and
evaluated three randomly selected video recordings for procedural integrity – 33% in the
baseline phase (session 3), and 33% in the intervention phase (sessions 6 and 7).

5.4.2. Data reliability: Scoring of the AAC Outcome Measure for PwD

The Augmentative and Alternative Communication Outcome Measure for PwD

(AACOM-PwD) was developed as Component 4 of the PC-AAC intervention for PwD. The
AACOM-PwD was administered as an observational tool to score the turn taking of the case
study participant during her interpersonal interaction with the electronic life story
conversational support (Component 2) (see Sections 3.4 and 3.6 respectively). Interpersonal
interaction between the researcher and case study participant took place in two phases (base
line and intervention) and were video recorded. The researched watched all video recordings
to assess the reliability of scoring the AACOM-PwD. The video-recorded data used for
reliability of scoring ranged from 5 minutes 24 seconds in the baseline phase to 10 minutes
11 seconds in the intervention phase. An independent rater (described in Par. 4.13.4) watched
33% of the video recordings and scored the turn taking independently.
Cohen’s Kappa is considered a suitable statistical measurement of the level of intra-
rater and inter-rater agreement. However, a case study with one participant did not meet the
minimum sample size of two participants for the statistical interpretation of reliability results
(Bujang & Baharum, 2017).
The researcher and independent rater used electronic formats of the turn taking coding
scheme and score sheets of the AACOM-PwD to score the turn taking of the case study
participant (Appendix F1). In the sections that follow, intra-rater and inter-rater scoring
agreement results are presented for overall number of turns, four turn taking domains (verbal
turns; gestural turns; aided turns; multimodal turns), and 11 turn taking sub-domains (verbal
turns, such as vocalisations or verbal utterances).

5.4.2.1.1. Intra-rater reliability

The researcher scored nine video recordings (i.e., three baseline sessions and six
intervention sessions) twice. The first scoring occurred subsequent to each baseline and
intervention session with the participant. Following a one-month interval, the second scoring
was performed. The researcher assigned a score on the AACOM-PwD contingent to the case
study participant’s turn taking responses in each session. Each turn taking response was
coded and scored in one of four turn taking domains (verbal, gestural, aided or multimodal)

114
© University of Pretoria
and the overall number of turns was counted. The modality of each turn was then coded as a
turn taking sub-domain. The researcher scored the participant’s turn taking in the same
manner each time by utilising the same turn taking coding scheme and electronic formats of
the scoring sheets (Appendix F2). The raw scores of these intra-rater datasets are displayed as
ratings 1 and 2 in Table 5.18.

Table 5.18: Intra-rater reliability of scoring the AACOM-PwD – domain level

Phase Baseline Intervention Mean percentage
Session 1 2 3 4 5 6 7 8 9 agreement
Total turns Rating 1 27 27 29 79 85 89 62 83 85
Rating 2 29 28 31 81 87 93 64 85 83 96%
% agreement 93 96 94 98 98 96 97 98 98
Verbal turns Rating 1 11 13 13 36 35 39 29 32 30
Rating 2 12 12 14 35 36 41 30 32 30 96%
% agreement 92 92 93 97 97 95 97 100 100
Gestural turns Rating 1 3 3 4 4 4 9 7 6 8
Rating 2 3 3 4 3 6 10 7 7 8 91%
% agreement 100 100 100 75 67 90 100 86 100
Aided turns Rating 1 1 1 1 3 4 3 3 4 3
Rating 2 1 1 1 3 4 3 4 4 3 97%
% agreement 100 100 100 100 100 100 75 100 100
Multimodal Rating 1 12 10 11 36 42 38 23 41 44
turns Rating 2 13 12 12 40 41 39 23 43 42 93%
% agreement 92 83 92 90 98 97 100 95 95
All percentage (%) agreement scores were rounded off to the nearest whole.

To report on the reliability of the intra-rater datasets, the raw scores were compared on
a point-by-point rating, which resulted in a percentage agreement between ratings 1 and 2. A
mean percentage of intra-rater agreement was subsequently determined. Intra-rating of the
AACOM-PwD scores is reported on a domain level (Table 5.18) and on a sub-domain level
(Table 5.19).
As reflected in Table 5.18, the total turns on the AACOM-PwD ranged from 93%
(Session 1) to 96% (Session 2) in the baseline intra-ratings. For total turns in the intervention
sessions, intra-rater agreement ranged from 96% (Session 6) to 98% (Sessions 4,5 8 and 9).
The mean percentage agreement for the total turns in all sessions scored was 96%. Intra-rater
agreement scores for verbal turns ranged from to 92% (Sessions 1 and 3) to 100% (Sessions 8
and 9). The mean scoring of total verbal turns was 96%. Agreement for the gestural turns in
each session varied from 67% (Session 5) to 100% (Sessions 7 and 9). The mean intra-rater
agreement for gestural turns was 96%.

115
© University of Pretoria
 All intra-rating agreements for aided turns were 100%, except for a 75% agreement in
Session 7. Overall, this consistently high agreement resulted in a mean intra-rater agreement
of 97% for total aided turns. The agreement for multimodal intra-rater turns ranged from 83%
in Session 2 to 100% in Session 7. A mean percentage of 93% was obtained for all intra-
ratings of all multimodal turns. Collectively, at a domain level on the AACOM-PwD, intra-
ratings for the total turns (96%), verbal turns (96%), gestural turns (91%), aided turns (97%)
and multimodal turns (93%) met the acceptable minimum agreement of 80% (McHugh,
2012).
Similar to the domain level, the reliability of intra-rating data is reported as the
percentage agreement and mean percentage agreement in each sub-domain. At a sub-domain
level, a raw score of zero indicates that the specific modality of the turn was not observed by
the rater (i.e., the sub-domain turn did not occur). This applies to the baseline phase for aided
turns (with assistance), as the researcher did not provide any support with the AAC device
(see baseline procedures found in Appendix W). In this instance, the agreement refers to a
non-occurrence agreement (Ledford, Dane, & Gast, 2018). The intra-rating reliability for the
11 turn taking sub-domains is reported in Table 5.19.

116
© University of Pretoria
Table 5.19: Intra-rating reliability of scoring the AACOM-PwD – sub-domain level
AACOM-PwD AACOM-PwD sub- Baseline phase Intervention phase Mean %
domain domain Session agreement
1 2 3 4 5 6 7 8 9
Verbal turns Vocalisations Researcher rating 1 2 1 1 5 2 6 1 1 1 94%
Researcher rating 2 2 1 1 5 2 6 2 1 1
% agreement 100 100 100 100 100 100 50 100 100
Verbalisations Researcher rating 1 9 12 12 31 33 33 28 31 29 96%
Researcher rating 2 10 11 13 30 34 35 28 31 29
% agreement 90 92 92 97 97 94 100 100 100
Gestural turns Body movement Researcher rating 1 1 1 1 1 2 1 1 1 1 100%
Researcher rating 2 1 1 1 1 2 1 1 1 1
% agreement 100 100 100 100 100 100 100 100 100
Head nod Researcher rating 1 1 1 1 2 1 5 5 4 6 85%
Researcher rating 2 1 1 1 1 3 5 5 5 6
% agreement 100 100 100 50 33 100 100 80 100
Facial expressions Researcher rating 1 1 1 2 1 1 3 1 1 1 97%
Researcher rating 2 1 1 2 1 1 4 1 1 1
% agreement 100 100 100 100 100 75 100 100 100
Aided turns With assistance Researcher rating 1 0 0 0 2 3 2 2 3 2 96%
Researcher rating 2 0 0 0 2 3 2 3 3 2
% agreement 100 100 100 100 100 100 67 100 100
Without assistance Researcher rating 1 1 1 1 1 1 1 1 1 1 100%
Researcher rating 2 1 1 1 1 1 1 1 1 1
% agreement 100 100 100 100 100 100 100 100 100
Multimodal Verbal +Gestural Researcher rating 1 9 9 8 32 39 35 20 23 31 93%
turns Researcher rating 2 10 11 9 36 38 36 20 25 31
% agreement 90 82 89 89 97 97 100 92 100
Verbal + Aided Researcher rating 1 1 1 1 1 1 1 1 13 2 88%
Researcher rating 2 1 1 1 1 1 1 1 13 2
% agreement 100 100 100 100 100 100 100 100 100
Gestural + Aided Researcher rating 1 1 0 1 1 1 1 1 1 1 100%
Researcher rating 2 1 0 1 1 1 1 1 1 1
% agreement 100 100 100 100 100 100 100 100 100
Verbal + Aided + Researcher rating 1 1 0 1 2 1 1 1 4 10 98%
Gestural Researcher rating 2 1 0 1 2 1 1 1 4 8
% agreement 100 100 100 100 100 100 100 100 80

117
© University of Pretoria
 As reported in Table 5.19, all 11 turn taking sub-domains met the acceptable intra-rater
agreement of 80% (McHugh, 2012). The mean percentage of intra-rating is reported on at a
sub-domain level. This was noted in a 100% agreement in three sub-domains, namely
gestural turns with body movements, aided turns without assistance, and multimodal turns
(i.e., a combination of gestural and aided turns). Mean intra-ratings of 94%, 96%, 97% and
93% were obtained for verbal turns with vocalisations, aided turns with assistance, gestural
turns with facial expressions, and multimodal turns with a combination of verbal and gestural
turns respectively. In four other sub-domains, an acceptable level of mean intra-rater
agreement was obtained. For verbalisations (96%), head nods (85%), and the two multimodal
sub-domains (i.e., combined verbal and aided turns; and combined verbal, aided and gestural
turns), agreement was at 88% and 98% respectively.

5.4.2.1.2. Inter-rater reliability

One session in the baseline phase (Session 2) and two in the intervention phase
(Sessions 8 and 9) were randomly selected to assess inter-rater reliability of scoring the
AACOM-PwD. An acceptable level of inter-rater mean agreement above 80% (McHugh,
2012) was obtained for the total turns observed (93%), as well as for turn taking in the
gestural domain (91%) and multimodal domain (83%). A mean percentage of inter-rater
agreement for the verbal domain (79%) and aided domain (72%) did not meet the minimum
acceptable level of 80% (McHugh, 2012). The inter-rater reliability raw scores, percentage
agreement and mean percentage agreement are displayed in Table 5.20.

Table 5.20: Inter-rater reliability of scoring the AACOM-PwD – domain level

Total scores of turns on AACOM-PwD domains
Inter-rated Inter-raters Total Verbal Gestural Aided Multimodal
session turns turns turns turns turns
Session 2 Researcher 27 13 3 1 10
Independent rater 29 12 3 1 13
% agreement 93% 92% 100% 100% 77%

Session 8 Researcher rating 83 32 6 4 41

Independent rating 88 26 7 2 48
% agreement 94% 81% 86% 50% 85%

Session 9 Researcher rating 85 30 8 3 44

Independent rating 78 19 7 2 50
% agreement 92% 63% 88% 67% 88%
Mean % agreement 93% 79% 91% 72% 83%
All percentage (%) agreement scores were rounded off to the nearest whole.

118
© University of Pretoria
 Further inter-rating reliability scoring results for the 11 turn taking sub-domains are
reported in Table 5.21. In this table, the percentage agreement in each sub-domain is
presented horizontally, while mean agreement across three inter-rated video-recorded
sessions is presented vertically. The mean inter-rater agreement for vocalisations (67%) and
verbalisations (76%) were found to be below the accepted minimum level of 80% for inter-
rater agreement (McHugh, 2012).

Table 5.21: Inter-rater reliability of scoring the AACOM-PwD – sub-domain level

AACOM-PwD Phase Baseline Intervention Mean %

agreement
Domain Sub-domain Session 2 8 9
Inter-ratings
Verbal Vocalisations Researcher 1 1 1
turns Independent rater 1 2 2 67%
% agreement 100 50 50
Verbalisations Researcher 12 31 29
Independent rater 11 24 17 76%
% agreement 92 77 59
Gestural Body movement Researcher 1 1 1
turns Independent rater 1 1 1 100%
% agreement 100 100 100
Head nod Researcher 1 4 6
Independent rater 1 5 4 82%
% agreement 100 80 67
Facial Researcher 1 1 1
expressions Independent rater 1 1 2 83%
% agreement 100 100 50
Aided With assistance Researcher 0 3 2
turns Independent rater 0 1 1 61%
% agreement 100 33 50
Without Researcher 1 1 1
assistance Independent rater 1 1 1 100%
% agreement 100 100 100
Multi- Verbal + Researcher 9 23 31
modal Gestural Independent rater 11 35 35 79%
turns % agreement 82 66 89
Verbal + Aided Researcher 1 13 2
Independent rater 2 9 7 49%
% agreement 50 69 28
Gestural + Researcher 0 1 1
Aided Independent rater 0 1 1 100%
% agreement 100 100 100
Verbal + Aided Researcher 0 4 10
+ Gestural Independent rater 0 4 9 97%
% agreement 100 100 90

119
© University of Pretoria
 The mean inter-rater percentage agreement for the gestural sub-domains were 100%
(i.e., body movements), 82 % (i.e., head nods), 83% (i.e., facial expressions). In the aided
sub-domain, inter-rater agreement scores were 100% (aided turns without assistance) and
61% (aided turns with assistance). Two multimodal sub-domains met the minimum
acceptable level of 80% (McHugh, 2012). These sub-domains were 100% (i.e., multimodal
turns with gestural and aided turns) and 97% (i.e., multimodal turns with verbal, aided and
gestural turns).
One sub-domain (i.e., multimodal turns with verbal and gestural turns) obtained an
inter-rater agreement score of 79%, falling slightly below the 80% threshold of acceptable
inter-rater agreement. Multimodal turns with combined verbal and aided turns (49%) was the
sub-domain with the lowest mean inter-rater percentage agreement (McHugh, 2012). In
summary, inter-rater agreement scores in six sub-domains met the minimum acceptable level
of 80% (McHugh, 2012), while five sub-domains fell below this threshold.

5.5. Stakeholder’s perspective: Participant with dementia

The researcher adhered to the Stakeholder Perspective Checklist (Appendix Y) as a

guide to ask the participant two questions related to her enjoyment of using the electronic life
story conversational support. This process unfolded in a conversational manner and was
recorded on video. The researcher transcribed the participant’s verbatim responses, which are
shown in Table 5.22 as qualitative data.

Table 5.22: Stakeholder’s perspectives – Self-reported enjoyment of a PwD

Session Did you enjoy talking about How much did you Other verbal information
your life story today? enjoy talking about your
(Yes, No, Unsure) life story today?
1 Yes “Hmm” “It was good”
2 Yes Off-topic response
3 Ok “A lot, I guess” “I liked it amazingly”
4 Yes “A lot”
5 “Oh, I don’t know” Off-topic response
6 Off-topic response
7 Yes “Yes”
8 Oh, yes “This is lovely…I haven’t done
anything like it, like it before”
(during the video recording)

9 Off-topic responses “That is very difficult”

120
© University of Pretoria
 As shown in Table 5.22, the researcher asked the first question verbally: “Did you enjoy
talking about your life story today?”. The researcher supported each question as well as the
conversation that was elicited from the questions by using AAC strategies (outlines in Par.
4.14). Although the researcher provided the participant three option cards with symbols to
indicate ‘yes, no, unsure’ responses, the participant provided verbal responses to Question 1.
The participant responded verbally and in the affirmative to the first question in sessions 1, 2,
4, 7 and 8. She also produced off-topic responses (Sessions 2 and 5) or did not respond
(Sessions 6 and 9). Although the researcher attempted to elicit meaningful responses in these
instances, she was sensitive to the participant’s body language, which indicated fatigue or
restlessness.
Based on the participant’s response to Question 1, the researcher asked the second
question using the same AAC strategies as applied in the first question. The use of
multimodal AAC strategies supported the participant in expressing her opinion in a natural
conversational manner. For instance, in Session 3, when the researcher expanded on the
participant’s response, the latter made a further contribution to confirm her enjoyment of the
session, as noted in her response, “I liked it amazingly”.
Although the participant did not self-report her enjoyment in Session 9, observational
information from the video recording of the entire session provided anecdotal information.
For instance, it was evident that the participant laughed many times during Session 9 and
used hand movements to express enjoyment whilst listening to music on the voice output.

5.6. Overview of results: Phases 3A and 3B

In Phase 3A, professionals’ viewpoints on the components of the PC-AAC intervention

for PwD were obtained. Professionals in Factor A viewed person-centred authorship and
directly reported outcomes from the perspective of a person with dementia as important
components. Professionals in Factor B, with more clinical expertise, viewed person-centred
language style as a way to uphold the personhood of a person with dementia and believed
that this may be achieved through AAC strategies and techniques such as augmented input.
Professionals in Factor C valued person-centred communication strategies such as displays of
empathy but expressed a neutral view on adapted participatory methods to involve persons
with dementia in developing their AAC interventions.
In Phase 3B, the feasibility of implementing the electronic life story conversational
support and scoring reliability of the AACOM-PwD was tested. The electronic life story

121
© University of Pretoria
conversational support was reliably implemented by the researcher. Acceptable levels of
scoring agreement were noted in the intra-rater and inter-rater reliability of scoring the
AACOM-PwD at a domain level (e.g., overall turns). However, inter-rater reliability was not
at an acceptable level of scoring agreement in the multimodal sub-domain. The participant
with dementia self-reported her enjoyment in using the electronic life story conversational
support with the researcher.

5.7. Summary

The results of the feasibility phase of the main study were presented in this chapter in
two parts. Firstly, Q-methodology data was collected qualitatively and analysed
quantitatively through by-person factor analysis. Three factors (A, B and C) emerged as the
viewpoints of professionals on the components of the PC-AAC intervention for PwD. The
results of ranked statements by z-scores were reported in distinguishing and consensus
statements. Secondly, the case study results were presented for procedural reliability and for
data reliability of the intra-rater and inter-rater scoring of the AAC Outcome Measure for
PwD developed in this study. Finally, the perspective of the participant on her enjoyment of
using the electronic life story conversational supported was reported. Collectively, the results
presented in this chapter provide a variety of findings on the four components of the PC-AAC
intervention for PwD developed in this study. These findings will be discussed in the next
chapter.

122
© University of Pretoria
 CHAPTER 6: DISCUSSION

6.1. Introduction

This chapter discusses the results presented in Chapter 5. The discussion unfolds in
accordance with the three viewpoints that emerged via Q-sort data. As illustrated in Figure
6.1, the viewpoints of professionals are discussed according to the following four
components of the PC-AAC intervention package for PwD:
i. Person-centred care approach within AAC interventions for PwD
ii. Electronic life story conversational support for PwD
iii. AAC outcomes
iv. AAC outcome measure for PwD
The results of the procedural and scoring reliability of an AAC outcome measure for
PwD, developed as part of Component 4, are also discussed. In each section, the findings of
this study are compared to those of previous research studies. Pertinent literature is also
consulted to highlight important implications of this study.

Results

Four main components of a PC-AAC intervention for PwD

1. 2. 3. 4.
Person- Electronic AAC AAC
centred life story outcomes outcome
care approach conversational for PwD measure for
support PwD

Professionals’ viewpoints Reliability of scoring

Self-reported enjoyment
of a person with dementia

Figure 6.1: Schematic illustration of the discussion chapter

123
© University of Pretoria
 In addressing sub-aim iii of the main study (see Figure 6.1), the researcher examined
the patterns of professionals’ clinical and educational viewpoints on the four components of
the PC-AAC intervention package for PwD developed in this study. The professionals were
not informed of this intervention package or of its specific components or sub-components.
During concourse development (Par. 4.5.1.4), professionals answered open-ended questions
and contributed their unique opinions on elements of person-centred care, AAC strategies for
PwD, AAC outcome and AAC outcome measurement. In the collection of Q-sort data,
professionals had full control of how they interpreted and subsequently ranked each
statement, based on their unique opinion.
Based on the study results, three distinct factors (viewpoints) emerged. These three
factors provided an idea of how researchers, clinicians and academic teachers viewed various
components of the PC-AAC intervention package for PwD, based on their subjectivity
towards statements. In accordance with Q-methodology, the statements that distinguished
each factor, and the highest- and lowest-ranked statements of each factor were brought to the
forefront of the discussion (Zabala et al., 2018). Consensus statements across all three factors
were used to highlight significant components.

6.1. A person-centred care approach in AAC interventions for persons with dementia

The concept of person-centred care is often questioned in the literature, with even wider
speculation about the validity of its clinical application (Edgar et al., 2020; Dewing, 2019;
Dewing & McCormack, 2017; Gibson et al., 2019; McCance et al., 2011; McCormack et al.,
2011; Røsvik et al., 2013; Swaffer, 2019). However, in this study there was a commonly held
strong agreement across all factors that person-centred care as a philosophy of care for
persons with dementia is clinically achievable (Table 5.16). This view was further
substantiated from the data that emerged as specific person-centred care statements were the
highest-ranked statements in Factors A and B (Tables 5.7, and 5.10 respectively). In fact, it is
noteworthy that Factor B prioritised all of their strongest agreements with statements on
person-centred care (Table 5.10). Factor B offered a strong collective view that persons with
dementia must not only be spoken to in a manner that upholds their personhood, they must be
supported in decision making and consulted about their preferences (Bourgeois et al., 2016;
Chang, 2015; Chang & Bourgeois, 2020; Fazio et al., 2018; Reitz & Dalemans, 2016).
Tailored AAC interventions for persons with dementia should also be introduced according
to strengths.

124
© University of Pretoria
 Factors A and B represented the majority of professionals (Table 5.5) in the Q-sort
data. Therefore, it can be inferred that the majority of professionals expressed their strongest
agreement with, and indicated a possible validation of person-centred care components and
sub-components of the PC-AAC intervention package for PwD developed in this study.
A fine-grained analysis revealed that professionals with the highest factor loadings and
most defined views in Factor A (Table 5.6), Factor B (Table 5.9) and Factor C (Table 5.12)
indicated a special interest in person-centred care for adults or for both adults and children
with disabilities. Taken together, their views may suggest that professionals have a strong
view of person-centred care as part of AAC interventions for persons with dementia. Further
specific statements related to the concept of person-centred care are integrated and discussed
in the sections that follow.

6.2. PC-AAC scaffolding strategies

This study found that conversational partners play a significant role in positively
affirming the unique identity of a person with dementia (Brooker, 2004; Kitwood, 1997).
According to Webb (2017), person-centredness within interpersonal interactions is mostly
operationalised as a relational concept achieved through the manner and style of a
conversational partner. The professionals making up Factor B strongly agreed that talking to
a person with dementia in an infant-like manner was patronising. This view is important for
two reasons. Firstly, it validates how person-centredness within AAC-supported interpersonal
interactions lies not only with a person with dementia, but extends to the relationship that
develops between conversational partners (Alm et al., 2004; Astell et al., 2010; Davis &
Shenk, 2015; Ellis & Astell, 2017; Hamel et al., 2016; Hydén, 2011; Kindell et al., 2013;
Purves et al., 2015; Webb, 2017; Wong et al., 2009). The current finding endorses the
relational attribute of upholding the personhood of a person with dementia (Kitwood, 1997;
Webb, 2017).
A complementary aspect of the relational focus of person-centred interpersonal
interactions was introduced by Factor C’s very strong agreement that a conversational partner
should never correct off-topic utterances of a person with dementia. However, this view
stands in contrast to previous studies where the effect of non-electronic memory and
communication supports focused on increasing on-topic statements and reducing the number
of off-topic responses in a person with dementia (Aitken, 2015; Andrews-Salvia et al., 2003;
Bourgeois et al., 2001; Chang, 2011; Dynes, 2018; Fried-Oken et al., 2009; Fried-Oken et al.,

125
© University of Pretoria
2012; Gomèz-Taibo et al., 2014; McPherson et al., 2001; Reitz & Dalemans, 2016; Ruiz,
2015). Whilst reducing off-topic responses may be an appropriate outcome of interventions
focused on improving the quality of language competency, it may arguably be of less
importance when the AAC outcome is aimed at building a social connection with a person
with dementia (Alant, 2017; Hung et al., 2019). Furthermore, this result presents an
interesting clinical insight. As professionals primarily interested in interpersonal interaction,
the views of Factor C appeared to pose an important consideration to AAC interventionists in
terms of how researchers are measuring the competency of a person with dementia in using
AAC. It may be inferred that if a conversational partner views communication competence as
a meaningful contribution by a person with dementia within an interpersonal interaction, and
if the goal is social connection (Light, 1988), then off-topic responses may be assumed to be
irrelevant.
Related to other relational aspects of person-centred care, the findings of the current
study indicated that empathy may be an important construct to apply within AAC-supported
interpersonal interactions with persons with dementia (Alant, 2017; Alant et al., 2015;
Broughton et al., 2011; McEvoy & Plant, 2014). This was reflected in the firm agreement by
professionals in Factor C that conversational partners should display empathy by mirroring
the emotion of the person with dementia as a means to facilitate interpersonal interaction.
Furthermore, Factor A expressed agreement that a conversational partner could support
interpersonal interaction by pausing to give a person with dementia time to respond.
These findings support previous research on person-centred communication which
emphasises the importance of a conversational partner to listen, observe, and validate the
responses of a person with dementia (Alant, 2017; Alant et al., 2015; Broughton et al., 2011;
Dynes, 2018; McEvoy & Plant, 2014; Savundranayagam et al., 2020; Savundranayagam &
Moore-Nielsen, 2015; Savundranayagam et al., 2016). Perhaps this finding may further
suggest that person-centred communication strategies may be a vehicle through which
developing friendships and a social connection with persons with dementia could possibly be
achieved in PC-AAC interventions for persons with dementia (de Medeiros et al., 2012;
Perion & Steiner, 2017; Sabat & Lee, 2012; Ward et al., 2012).
Augmented input was evidenced in previous case studies to be a useful multimodal
conversational partner strategy to support auditory comprehension in persons with semantic
dementia (Cress & King, 1999) and thereby to improve conversational turn taking in a person
with frontotemporal dementia (Gibbons et al., 2012). In the current study, professionals in
Factor B expressed their agreement with the use of augmented input as a strategy to support

126
© University of Pretoria
auditory comprehension of spoken language in persons with dementia. Upon closer
inspection of the biographical data, Factor B was represented by three clinicians with work-
related experience in adults with other neurologically acquired communication disorders
(Table 5.9). It may be plausible to infer that this view may suggest the clinical relevance of
augmented input for persons with dementia.
Whilst the effect of augmented input on persons with aphasia was investigated in
previous studies (Dada et al., 2019; Wallace et al., 2012), there are few studies that
systematically explored this strategy in AAC interventions for persons with dementia (May et
al., 2019). Augmented input using high-context visuographic support was found to support
comprehension of narratives in persons with aphasia in a pilot study conducted by Dada et al.
(2019). This finding potentially suggests the clinical relevance of augmented input as a
strategy to support comprehension of life story content in an electronic format on an AAC
device in the AAC intervention developed in this study (Cress & King, 1999; Gibbons et al.,
2012).
Furthermore, Factor B strongly agreed about the importance of supporting the decision
making of persons with dementia. Previous studies found evidence that visual and verbal
support in understanding and reasoning related to the decision-making capacity of persons
with mild to moderate dementia had a positive effect on their end-of-life care (Bourgeois et
al., 2016; Chang, 2015; Chang & Bourgeois, 2020; Murphy & Oliver, 2013; Reitz &
Dalemans, 2016). In summary, the results of the current study support previous studies (Dada
et al., 2019; Dada et al., 2020) that exemplify the potential of aided input as a multimodal
strategy to support language comprehension in persons who require AAC.

6.3. Electronic life story conversational support

The implementation of assistive technology with artificial intelligence capabilities to

support communication in persons with dementia is a timely topic in the field of AAC (Dada
et al., 2020). Congruently, in this study professionals across all factors agreed that intelligent
assistive technology and AAC devices using artificial intelligence should be explored to
support communication and interpersonal interaction in persons with dementia (Table 5.16).
Whilst this is an optimistic finding, the adoption of both assistive technology with artificial
intelligence and electronic AAC devices for persons with dementia is generally influenced by
their design (Waller, 2019).

127
© University of Pretoria
 The literature highlights the importance of supporting persons with dementia to be
included in research by using adapted participatory methods (Hydén et al., 2018; Keady et al,
2018; Kullberg & Odzakovic, 2018). In this regard, Mooney et al. (2018) developed an AAC
application that comprises computerised processing of human language (natural language
processing) and just-in-time principles using personalised vocabularies co-constructed with
persons with semantic dementia. In the current study there was shared agreement among the
three factors that facilitating the involvement of persons with dementia in co-designing their
own AAC interventions promotes their self-determination. This finding is an important one,
as it is aligned with current discourses on participatory design methods that involve persons
who use AAC as collaborators of their own intervention (Dewing, 2008a; Kindell et al.,
2017; Murphy & Oliver, 2013; Murphy et al., 2010; Johnson, 2020; Moulam et al., 2020;
McLaughlin, 2020; Taylor & Balandin, 2020). Perhaps this finding further suggests that
involving persons with dementia as co-producers of their own electronic AAC interventions,
technology abandonment may potentially be reduced and acceptability of the interventions
may be increased (McConnell et al., 2019; Realpe & Wallace, 2010; Waller et al., 2005;
Waller, 2019).
Regardless of the benefits that participatory methods may hold for promoting person-
centred care, it is professionals who ultimately decide whether such methods actually are
employed in research and clinical practice (Giezendanner et al., 2019; Hickey et al., 2018;
Hopper, 2003; Mahendra et al., 2011; Mahendra et al., 2013; Saccasan & Scerri, 2020). The
results of the current study show that there may be a professional hesitancy with regard to
adopting participatory design methods in research with persons with dementia. This finding
may be problematised in the literature by the limited research on person-centred AAC
interventions and the need for professional training on how to involve persons with dementia
in decision-making (Bourgeois et al., 2016; Crete-Nishihata, 2012; Douglas et al., 2018;
Dynes, 2018; Saccasan & Scerri, 2020).
Moreover, this finding was also evident in the distinguished neutral view of
professionals in Factor C towards co-production as a participatory method with persons with
dementia in the design of an AAC intervention (Table 5.14). Furthermore, Factor C also
affirmed their distinct view that while working collaboratively with persons with dementia
and other stakeholders was important, it was not necessary (Table 5.14).
Notably, professionals in Factor C were the most well-experienced of all factors in their
overall work-related experience (i.e., > 21 years) in the areas of AAC, dementia studies and
technology (Table 5.12). As a factor, they collectively had an interest in interpersonal

128
© University of Pretoria
interaction rather than in person-centred care per se. What their neutral view may suggest, is
that the cognitive and communicative challenges of persons with dementia and ethical issues
related to informed consent make participatory design methods difficult to implement.
Furthermore, Johnson (2020) asserts that there is a general lack of guidance to AAC
professionals on how to implement participatory research with persons who use AAC. While
this may explain Factor C’s neutrality on participatory methods, Taylor and Balandin (2020)
are of the opinion that AAC researchers should be obliged to justify their reasons for
excluding persons who require AAC (such as persons with dementia) from the intervention
design process.
The paradigm of participatory research, experiential knowledge and lived experience,
in which persons with dementia are viewed as experience-based experts, is significantly
underscored in the literature (Boivin et al., 2018; Davies et al., 2019; Hubbard et al., 2003;
Harding et al., 2019; Lepore et al., 2017; Mann & Hung, 2019; McLaughlin, 2020). In the
current study, Factor A expressed their strong distinguished agreement that persons with
dementia should be supported in authoring the content of their own life story conversational
support. Notably, this statement obtained the highest z-score (1.704), making it the highest-
ranked statement of all 37 statements in the Q-sort data. Importantly, this finding resonates
with current literature and has two main implications.
i. Firstly, it extends the essence of person-centred care to one of empowerment, control
and autonomy, where persons with dementia are viewed as experts of their own lives
and therefore should be authoring their own life stories from their unique perspective
(Bartlett, 2014; Bartlett et al., 2017; Crete-Nishihata, 2012; Fazio et al., 2018; Hydén &
Antelius, 2017; Keady et al., 2018; Yeandle et al., 2012). These elements possibly lean
towards person-centred care being an evolving concept and potentially becoming
nuanced with a more contemporary understanding of its operationalisation in research
with persons with dementia (Alsawy et al., 2019; Bosco et al., 2019; Hennely &
O’Shea, 2017; Hung & Chaudhury, 2011; Jesus et al., 2019; Terkelsen et al., 2020).
ii. Secondly, when persons with dementia are supported with adapted methods through the
use of AAC support, a new understanding of their own life story may potentially be
introduced into the design of electronic life story conversational supports (Critten &
Kucirkova, 2019; Kindell et al., 2014; Nedlund & Bartlett, 2019; Subramaniam &
Woods, 2016). The current study also supports literature that advocates the creation of
a ‘conversational space’ into which persons with dementia should be invited, and where
adapted methods may be used to enable the sharing of their stories (Antelius et al.,

129
© University of Pretoria
 2018; Hydén et al., 2018; Kindell et al., 2014; Kullberg & Odzakovic, 2018).
Furthermore, it supports previous AAC studies in which researchers utilised adapted
interview methods to co-construct meaning with persons with use AAC to facilitate
their communicative autonomy to convey their own stories in research (Smith et al.,
2018; Dada, Tönsing & Goldbart, 2020). As such, the results of the current study may
possibly point towards pertinent guidance of how AAC research and the design of
electronic conversational supports could be further explored with persons with
dementia.
Electronic AAC systems with multimedia support have been evidenced in previous
studies to promote enjoyment for both the conversational partner and persons with dementia
in an interpersonal interaction (Alm et al., 2004; Astell et al., 2010; Hamel et al., 2016; Hung
et al., 2020; Ferm, Ekström, Larsson, & Samuelsson, 2020; Purves et al., 2015). This finding
was confirmed in the current study. Both Factors A and C shared a distinct agreement that
life-story-based conversational supports with personalised music and digital photographs on
electronic devices (e.g., iPad) enhance the enjoyment of interpersonal interaction for persons
with dementia. However, on closer analysis of the findings, mixed professional views were
noted on whether electronic AAC systems were useful for persons with more severe
dementia.
Factor B tended to slightly disagree that non-electronic AAC systems were the most
beneficial communication supports for a person with mild to moderate dementia.
Furthermore, the viewpoints of the professionals making up Factor B indicated a possible
agreement with the notion that electronic legacy messages (i.e., digitally recorded by the
person with dementia in their natural voice in the early stages of dementia) stored as voice
output on electronic devices (e.g., iPad) were useful to facilitate social connection with a
person with dementia. This finding is in contrast to a previous exploratory pilot study by
Fried-Oken et al. (2012) in which voice output was found to have a negative impact on
communication in persons with dementia.
It may be plausible to suggest that, based on their collective clinical experience, the
professionals in Factor B may value the clinical utility of using electronic AAC systems with
specific web-based applications (Aitken 2015; Alm et al., 2004; Astell et al., 2010; Dynes,
2018; Hamel et al., 2016; Purves et al., 2015). This is supported in a previous study in which
an application, Computer Interactive Reminiscence and Communication University of
Sheffield (CIRCUS), showed to have a positive effect on interpersonal interaction with
persons with dementia (Samuelsson & Ekström, 2019).

130
© University of Pretoria
 Conversely, Factor A notably disagreed that electronic AAC systems such as tablet
computers with applications (apps) may be useful to support interpersonal interaction in a
person with moderate to severe dementia. This view may be due to concerns raised in the
literature that the novelty of technology use and unfamiliarity with electronic AAC systems
within interpersonal interactions may present challenges for persons with dementia as their
cognitive and communication difficulties advance (Fried-Oken et al., 2012; Fried-Oken et al.,
2009; Hung et al., 2020). The views of Factor A were from a majority of professionals who
were researchers and academic teachers employed at educational institutions. It may be
reasonable to infer that their views are due to the limited research evidence that exists of
communication intervention for persons with moderate to severe dementia in general (Swan
et al., 2018). Electronic AAC systems as a conversational support for persons with moderate
to severe dementia in particular is under-researched (Dynes, 2018; Ekström et al., 2017;
Fried-Oken et al., 2012; Hung et al., 2020; Swan et al., 2018). As the evidence base of
electronic AAC systems for persons with moderate to severe dementia expands and new
insights are discovered, professionals’ views in this regard may evolve.
A possibly cautionary reminder was the distinguished strong agreement of
professionals in Factor B that persons with dementia should be consulted about their personal
preferences for interpersonal interaction. This may indicate that any customisation and
personalisation of electronic conversational supports should be in accordance with the
personal preferences of persons with dementia (Bourgeois et al., 2016; Brooker, 2004; Fazio
et al., 2018; Kitwood, 1997). The literature also suggests that consulting persons with
dementia in the design process allows for the personalisation of electronic life story supports
(Hashim et al., 2013; Subramaniam & Woods, 2016).

6.4. AAC outcomes for Persons with Dementia

A decade ago, Bourgeois et al. (2010) cautioned AAC professionals that whilst
advancements in electronic AAC systems would create new interactional possibilities for
persons with dementia, goals that have a human and social impact on persons with dementia
and their conversational partners must remain. Findings in the current study support AAC
literature, as Factor A and C strongly agreed that a social connection was an important goal
of AAC interventions (Alant, 2017; Granlund et al., 2008) for persons with dementia, while
Factor B held a neutral view in this regard. This finding suggests that social connection as a
valued goal and as determined by a person with dementia should possibly be at the forefront

131
© University of Pretoria
of a person-centred AAC intervention (Birt et al., 2020; Haapala et al., 2019; Hancock et al.,
2006; Patterson et al., 2017; Perion & Steiner, 2017; van der Roest et al., 2009).
Findings in the current study indicate that professionals supported the measurement of
intra-personal outcomes such as enjoyment within an interpersonal interaction (Table 5.16).
Electronic life story conversational support, such as the one developed in this study, has the
potential to not only facilitate turn taking during interpersonal interaction, but also to create
opportunities for enjoyment during such interaction. This was substantiated by the direct self-
report from the participant with dementia in the case study (Table 5.22). The self-reported
enjoyment from the case study participant supports previous studies which indicate that
electronic conversational supports with life story content are enjoyable and may serve as a
means to overcome the social isolation experienced by persons with dementia (Astell et al.,
2005; Astell et al., 2010; Ekström et al., 2017; Elfrink et al., 2018; Ferm et al., 2020; Hamel
et al., 2016; Hung et al., 2020; Samuelsson & Ekström, 2019). The anecdotal evidence from
video recordings, showing that the participant displayed behaviours that indicated enjoyment
although she did not express this enjoyment verbally, is an important finding. This may
suggest that supplementing the self-reports of a person with dementia (Crete-Nishihata et al.,
2012; Mooney et al., 2018a; Mooney et al., 2018b), with proxy reports of enjoyment from a
familiar conversational partner may provide a holistic understanding of AAC outcomes in
persons with dementia.

6.5. AAC outcome measurement for Persons with Dementia

Sharing meaning and participating in everyday interactions may arguably be the most
optimal outcomes for persons using AAC systems (Alant, 2017; Fried-Oken & Granlund,
2012; Kindell et al., 2017; Savundranayagam, 2013). Findings in this study have supported
this argument. Professionals in Factor B felt that an AAC outcome measure for persons with
dementia should measure not only the frequency of turn taking exchanges, but also the
modality of turn taking (e.g., turn taking with gestures). It should possibly also include
participation outcomes, in other words how frequently a person with dementia takes part in
everyday interpersonal interactions.
In the current study, professionals across all three factors agreed about the necessity of
measuring the satisfaction of a person with dementia when using AAC strategies within an
interpersonal interaction. Importantly, this possibly suggests that a satisfaction domain should
be included in an AAC outcome measure, but it may also shed light on whose perspective

132
© University of Pretoria
should be sought to measure satisfaction of an AAC intervention (Broomfield et al., 2019;
Cohen & Hula, 2020; Swaffer, 2015). In the current study, the person with dementia self-
reported her enjoyment of using the electronic conversational support within an interpersonal
interaction with the researcher. This was an important finding for further development of
AAC outcome measures for persons with dementia, as it places value on self-reported
satisfaction expressed from their own perspective. The finding further supports previous
literature which highlight that AAC interventions for persons with dementia should have
meaningful real-life outcomes, and their satisfaction in using electronic systems must be self-
reported (Mooney et al., 2018a; Mooney et al., 2018b). The self-reported enjoyment in using
the electronic conversational support from the perspective of the participant with dementia
provides a complementary source of evidence to the reliability of scoring the AACOM-PwD
as discussed in the next section (Cohen & Hula, 2020).

6.6. Reliability

6.6.1. Procedural reliability

Reliability is central to evidence-based practice in AAC as it provides a way to measure

the accuracy of data collection and sheds light on the validity of research findings (Kent-
Walsh & Binger, 2018). Within the AAC literature, Koul, Petroi and Schlosser (2010)
suggest that AAC professionals rely on evidence of electronic AAC systems benefiting
persons who use AAC, before they consider implementing technology in clinical practice.
Since research on electronic AAC systems for persons with dementia is still developing (May
et al., 2019), the procedural reliability results in this study may provide preliminary evidence
to support further research into electronic AAC interventions for persons with dementia.
The high procedural integrity results indicate that the electronic life story
conversational support developed in this study can be reliably implemented by the same
individual and with a different individual (Barton, Meadan-Kaplanksky, & Ledford, 2018).
This infers that AAC clinicians could possibly be trained in implementing the PC-AAC
scaffolding strategies to support turn taking in persons with dementia. As shown in previous
literature (Kent-Walsh & Binger, 2018; Morin et al., 2018), high procedural integrity
increases the confidence that AAC outcomes resulting from the use of electronic AAC
systems may be due to the accuracy and reliability of implementing procedures as intended
(Barton et al., 2018). Given the neurodegenerative nature of dementia (Bourgeois & Hickey,
2018; Mahendra et al., 2011; Mahendra et al., 2013), the preliminary procedural integrity

133
© University of Pretoria
results in this study may provide an avenue to propel further research to demonstrate the
effect of person-centred AAC interventions for persons with dementia.

6.6.2. Reliability of scoring the AAC Outcome Measure for PwD (AACOM-PwD)

According to Enderby (2014) an outcome measure must be used in the same manner on
more than one instance before it is considered to be effective in detecting changes over time.
In this study, intra-rater reliability was scored in the same manner at different time points and
showed acceptable levels of agreement on a domain level (Table 5.18). This acceptable intra-
rater reliability is of clinical importance as it indicates the preliminary consistency of ratings
of the AACOM-PwD across different time points (Enderby, 2014; Ledford, et al., 2018).
Similarly, at a sub-domain level (Table 5.19), there was acceptable intra-rater reliability
on all eleven sub-domains. Given the degeneration of the expressive language, family
members and other conversational partners of persons with dementia might be inclined to
focus only on the loss of communication (Bourgeois & Hickey, 2018; de Medeiros et al.,
2012; Hopper, 2003; McEvoy & Plant, 2014). With a loss of expressive language and
progression of dementia, the AACOM-PwD may have clinical utility to provide
conversational partners with an observed quantified picture of the specific type of turns that
are used by a person with dementia (Cohen & Hula, 2020). This may be specifically
important for the multimodal turn taking domain (Brock et al., 2019). The literature shows
that there is limited evidence of the use of multimodal AAC systems for persons with
progressive cognitive-communicative conditions (Broughton et al., 2011; Cress & King,
1999; Mooney et al., 2018a; Rebstock & Wallace, 2020; Wong et al., 2009).
Therefore, in this study, multimodal turns as captured on the AACOM-PwD show that
the combined use of verbal utterances, vocalisations with head nods, or body movements
within turn may be a valuable indicator of the overall communicative strengths that a person
with dementia possesses (Brock et al., 2019; Broughton et al., 2011; Cress & King, 1999;
Enderby, 2014; Kindell et al., 2013; Trahan et al., 2014; Wong et al., 2009). This finding
supports current research on outcome measurement tools in persons with dementia, which
focus on the preserved strengths of persons with dementia (Hickey et al., 2018; Lanzi, et al.,
2020). The AACOM-PwD may therefore serve as a strength-based measure of the various
types of turn taking that the person with dementia still utilises in interpersonal interaction,
instead of focusing on communication deficit and loss.
Acceptable inter-rater scores were found in respect of the level of agreement with the
total number of turns at a domain level (Table 5.20), which potentially suggests that the turn

134
© University of Pretoria
taking coding scheme may be used to train other raters. Importantly, however, inter-ratings
were not at an acceptable level for verbal turns (e.g., verbalisations), aided turns (e.g., with
assistance) and multimodal turns (e.g., verbal and aided) at a sub-domain level (Table 5.21).
In this vein, it could be inferred that the observation and coding of subtle multimodal turn
taking behaviours are open to interpretation (Ledford et al., 2018). Therefore, further training
of raters in coding and refinement of the existing coding definitions may be an important
consideration for any future studies (Gast & Ledford, 2018).
Reliability of scoring was ensured owing to video analysis that allowed repeated
viewings of turn taking. Previous studies show that the coding of subtle behaviours based on
video analysis is a time-consuming process (Williams et al., 2017). Given that certain sub-
domains, e.g., multimodal turns, did not meet the acceptable agreement level of scoring at an
intra-rater level, it may be inferred that the scoring reliability of multimodal turns could be
enhanced by supplementing coding with conversational analysis procedures (e.g.,
transcriptions and analysis of turns) (Ekström et al., 2017; Hydén, 2011; Kindell et al., 2013;
Spilken & Bethlehem, 2003) or with computer-based microanalytic video coding of turn
taking (Ellis & Astell, 2017; Spriggs, et al., 2018).
While this was an exploratory study into the reliability of scoring the AACOM-PwD, it
is important to consider anecdotal information that may shed light on the possible clinical
utility of this measure (Zaga et al., 2020). Scrutiny of the surface-level turn taking data of raw
scores revealed that the number of total turns increased sharply from the baseline to the first
intervention session when the researcher applied the PC-AAC scaffolding strategies to the
interpersonal interaction. Firstly, this infers that the measure was able to establish a baseline
or stability of turn taking (Gast & Ledford, 2018). Secondly, this provided plausible evidence
to suggest that the AACOM-PwD may be sensitive to measuring a change in turn taking in a
person with dementia (Brock et al., 2019; Cohen & Hula, 2020; Spriggs et al., 2018). This
anecdotal finding – albeit in a small way – indicates that the AACOM-PwD may show
possible improvement in the turn taking of a person with dementia when using an electronic
life story conversational aid and being supported by a conversational partner with AAC
strategies.
This is a vital anecdotal finding of this study and has important clinical implications
that may warrant further research. In sum, the AACOM-PwD could potentially benefit AAC
clinicians and support them in measuring the effect of the AAC intervention – pre-
intervention, during intervention and as a post-intervention maintenance measure with other
conversational partners (Brock et al., 2019; Spriggs et al., 2018).

135
© University of Pretoria
6.7. Summary

This chapter discussed and interpreted the results of the study. Discussions were
aligned to the four components of the PC-AAC intervention package for PwD developed in
this study by highlighting the significance of three distinct professional viewpoints from Q-
methodology data. An understanding of how expert professionals view person-centred AAC
strategies, techniques, AAC outcomes and outcome measures for PwD can provide valuable
insights into current thinking on intervention development. Furthermore, knowing which
specific components professionals ranked as most or least important is a vital indicator of the
validity of the PC-AAC intervention developed in this study. Findings related to procedural
reliability of implementing the electronic life story conversational support and the scoring
reliability of the AACOM-PwD were also discussed.

136
© University of Pretoria
 CHAPTER 7: CONCLUSION, EVALUATION AND RECOMMENDATIONS

7.1. Introduction

This final chapter presents a summary of results and conclusions regarding the
exploration of the PC-AAC intervention package for interpersonal interaction with PwD that
was developed in this study. Next, the clinical implications of the research findings are
outlined, followed by an evaluation of the strengths and limitations of the study. The chapter
concludes with an outline of recommendations for future research.

7.2. Summary of the findings

This study was a scientific endeavour to develop and explore the PC-AAC intervention
package for interpersonal interaction with PwD. The package comprised four main
components, namely the person-centred care approach in AAC interventions, electronic life
story conversational support, AAC outcomes, and AAC outcomes measurement. The study
followed an evidence-based practice triad of inquiry.
i. Firstly, the assembly of the four components was based on current research evidence
that resulted from the researcher’s findings of a research review on AAC interventions
for persons with dementia. These findings were supplemented with the findings from a
second scoping review of intelligent assistive technology devices for persons with
dementia.
ii. Secondly, based on professionals’ clinical and educational expertise, three viewpoints
on the components of the PC-AAC intervention for PwD were obtained. Professionals
viewed person-centred authorship and directly reported outcomes from the perspective
of a person with dementia as important components of the PC-AAC interventions for
PwD. Professionals with more clinical expertise viewed person-centred language style
as a way to uphold the personhood of a person with dementia and believed that this
may be achieved through AAC strategies and techniques such as augmented input. The
results showed that some professionals view adapted participatory methods as a means
to facilitate the inclusion of persons with dementia in the design of their AAC
interventions, whereas others do not have a clear view on this concept.
iii. Thirdly, the researcher conducted a process of co-producing an electronic life story
conversational support with a person with dementia and implemented it in a real-life
context with the same individual. The feasibility of procedural reliability related to the

137
© University of Pretoria
 implementation of the electronic life story conversational support and intra-rater and
inter-rater reliability of scoring the AACOM-PwD was tested. Procedural integrity
results indicated that the electronic life story conversational support was reliably
implemented by the researcher. Acceptable levels of scoring agreement were noted in
the intra-rater and inter-rater reliability of scoring the measure, especially at a domain
level. Inter-rater reliability was not at an acceptable level of scoring agreement in some
sub-domains. Furthermore, a direct stakeholder’s perspectives were obtained from the
self-reported enjoyment of a person with dementia who used the electronic life story
conversational support with the researcher.

To summarise, this study drew evidence from three evidence-based practice

cornerstones. Accordingly, the results suggest that person-centred AAC interventions could
mean more than just integrating the preferences of the person with dementia once-off.
Instead, professional views point towards other integral elements that include persons with
dementia in playing vital roles throughout the process of designing AAC interventions.
Involving persons with dementia as decision makers creates the possibility to promote greater
autonomy, control and authorship of their AAC interventions. It is possible to infer that the
findings of this study herald a renewed perspective on how professionals could move the
field of AAC forward in re-thinking traditional ways of developing AAC interventions with
persons with dementia.

7.3. Clinical implications of the study

Three main clinical implications of this study emerged:

i. Firstly, the results suggest possible clinical guidelines for AAC professionals on the
participatory processes that could be considered when developing AAC interventions
for persons with dementia. They emphasise the involvement of persons with dementia
at various stages of the development and feasibility-testing phases. Accordingly, this
introduces clinical insights that may suggest the different roles persons with dementia
could play as co-producers of their own interventions. With the current rise in dialogue
on involving persons with dementia in AAC intervention research, the results of this
study could possibly serve as a reference point from which AAC researchers and
clinicians may consult for evidence-based decision making for persons with dementia.
Despite the co-production process being challenging, given the cognitive-
communicative difficulties of persons with dementia, AAC is in itself a vital strategy to

138
© University of Pretoria
 support the participation of persons with dementia in this intervention process. Taken
together, the adapted methodologies, processes and ways of facilitating the involvement
of persons with dementia within this study are of significant clinical relevance to AAC
professionals.
ii. Secondly, to the researcher’s knowledge, the AACOM-PwD is the first attempt of an
outcome measure of interpersonal interaction with persons with dementia in the field of
AAC. The AACOM-PwD makes an important clinical contribution in providing initial
psychometric testing on the reliability of scoring. Specifically, this adds to possible
ways of measuring aided conversations in persons, which is a vital aspect of evidence-
based decision making.
iii. Thirdly (and also as far as the researcher knows), this study is the first to bring together
various international AAC professionals with a range of clinical and educational
expertise in expressing their viewpoints on AAC interventions for persons with
dementia. Discourses of this nature are not only an important avenue to charter the way
forward in the field of AAC, they are also an integral aspect of developing the evidence
base for persons with dementia.

7.4. Evaluation of the study

In view of the scientific impact of this research study in contributing novel evidence to
the field of AAC, the most salient strengths and limitations are highlighted next.

7.4.1. Strengths

Methodologically, this study had two main strengths:

i. Q-methodology enabled patterns of professionals’ viewpoints to be obtained
qualitatively based on their own biases. When combined with the statistical
sophistication of quantitative analysis, three unique professional views on the
components of the PC-AAC intervention package for PwD emerged. This would not
have been possible to be elicited with traditional ranking scales. Hybrid research
methodologies such as Q-methodology add to the robustness of the results obtained in
this study.
ii. The inclusion of views from an e-Delphi panel of international AAC professionals
during concourse development added breadth and diversity of professional input to the
Q-sort data. This would not have been likely with an in-person focus group as is typical
of traditional Q-sorting. As far as the researcher currently knows, this study is the first

139
© University of Pretoria
 to focus on AAC interventions for persons with dementia by employing Q-
methodology as a research method. The combination of Q-methodology with an e-
Delphi panel, specifically during concourse development, further contributes to the
methodological innovation applied in the current study.

Conceptually, the study contributed important information to the three cornerstones of

evidence-based practice as applied to AAC. It was the first to review and synthesise the
current literature on AAC interventions in persons with dementia. The research review
conducted in Phase 1 of the study spurred the need for a secondary scoping, which provided a
synthesis of current evidence on electronic AAC systems for persons with dementia. In Phase
2 of the study, the researcher presented a preliminary organising framework of how to
potentially involve persons with dementia in the co-production of their own electronic
conversational support. The attempt was not to be prescriptive and there was no inference to
involve persons with moderate to severe dementia in all roles at all times during the design of
the AAC intervention. Instead, in the absence of documented guidelines on how to promote
the involvement of persons with dementia in co-producing electronic AAC interventions, this
study should contribute to the current dialogue on incrementally increasing involvement of
persons with dementia in AAC research.
Phase 3 added a novel contribution by offering a nuanced examination of the diversity
of viewpoints held by AAC professionals on the PC-AAC intervention developed in this
study. In sum, all three phases of this study offered an innovative contribution to the field of
AAC for persons with dementia. As such, the study has generated valuable insights for
expanding the knowledge base and making progress in the field of AAC for persons with
dementia.

7.4.2. Limitations
i. Firstly, given the exploratory nature of the research aim, Q-methodology extrapolated
viewpoints of professionals from a purposive sample. In an attempt to increase the
variances of views, only 26 Q-sorts were used in the final factor analysis after factor
rotations. Whilst these viewpoints may provide an understanding of the patterns of
different views on the components of the PC-AAC intervention for PwD, they do not
represent all viewpoints on the topic. Therefore, conclusions reached in this study
cannot be generalised. The online administration of the Q-sort may also have been
challenging for some professionals due to their being unfamiliar with Q-methodology
and the unexpected technical difficulties that accompany online data collection. This

140
© University of Pretoria
 may have potentially affected how professionals ranked the statements on the online
platform.
ii. Further reliability testing of scoring the AACOM-PwD with a larger participant sample
was impeded by the COVID-19 pandemic. Scoring reliability was limited to a single
participant with moderate to severe dementia. Consequently, the exploratory results for
the intra-rater and inter-rater reliability of scoring the AACOM-PwD in this study are
not conclusive.
iii. Family members afford unique perspectives to shaping evidence-based person-centred
AAC interventions for persons with dementia. Although the researcher managed to
obtain the perspectives of a person with dementia, family perspectives were distinctly
underexplored in this study.

7.5. Recommendations for future research

The recommendations for future research arising from this study include the following:
i. The effect that the PC-AAC scaffolding strategies for electronic life story
conversational support developed in this study had on interpersonal interaction in a
person with dementia is an avenue for further investigation. This matter could be
investigated further by using a single-case experimental design. Consequently, this
study may assist in expanding treatment efficacy studies related to electronic AAC
systems for persons with dementia.
ii. Further refinement is required of the turn taking codes on the AACOM-PwD and in-
situ training of inter-raters is recommended to improve the inter-rater reliability at a
sub-domain level.
iii. Reliability of scoring of the AACOM-PwD should be further improved through
training a larger number of raters and using a larger sample size of participants. This
may result in greater accuracy of inter-rater scores through more sophisticated
statistical measures (i.e., inter-class coefficients) to establish the psychometric
properties of the AACOM-PwD.
iv. Further investigation into the content validity of the AACOM-PwD, together with a
panel of AAC content domain experts, is needed as an independent validation study.
This will contribute to the refinement and psychometric properties of the AACOM-
PwD.

141
© University of Pretoria
 v. Further refinement is required of a PC-AAC co-production organising framework for
persons with dementia as applied in this study. Importantly, the involvement of family
members in the process of co-producing AAC interventions for persons with dementia
should be explored further in future studies.
vi. Finally, further research on PC-AAC interventions with persons with dementia, based
on their personalised life stories, must involve diverse persons with dementia,
especially from non-Western countries.

7.6. Summary

Chapter 7 summarised the results and presented the conclusions of the study. The significant
contributions of the study results were highlighted, followed by a presentation of the
strengths and limitations of the study. Finally, important recommendations that emerged from
the current research suggested possible avenues for further study.

142
© University of Pretoria
 REFERENCES

Aitken, K. (2015). The effects of a digital memory book on the quality and quantity of
conversations in adults with mild to moderate dementia. Master’s thesis, University
of Canterbury, New Zealand.
Alant, E. (2017). Augmentative and Alternative Communication: Engagement Participation.
San Diego, CA: Plural Publishing.
Alant, E., Samuelson, B. L., & Ogle, L. (2017). Empathetic communication: Bridging
differences in a global context. In S. Halder, & L.C. Assaf (Eds.), Inclusion,
Disability and Culture: An Ethnographic perspective Traversing Abilities and
Challenges (pp. 153-168). Cham, Switzerland: Springer.
Alm, N., Astell, A., Ellis, M., Dye, R., Gowans, G., & Campbell, J. (2004). A cognitive
prosthesis and communication support for people with dementia. Neuropsychological
Rehabilitation, 14(1-2), 117-134. doi:10.1080/09602010343000147
Alsawy, S., Mansell, W., McEvoy, P., & Tai, S. (2017). What is good communication for
people living with dementia? A mixed-methods systematic review. International
Psychogeriatrics, 29(11), 1785-1800. doi:10.1017/S1041610217001429
Alsawy, S., Tai, S., McEvoy, P., & Mansell, W. (2019). ‘It’s nice to think somebody’s
listening to me instead of saying “oh shut up”’. People with dementia reflect on what
makes communication good and meaningful. Journal of Psychiatric and Mental Health
Nursing, 27(2), 151-161. doi:10.1111/jpm.12559
American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental
Disorders (DSM-5) (5th ed.). Washington, DC: Author.
American Speech-Language-Hearing Association (ASHA). (2019). Augmentative and
alternative communication (AAC). [online]. Retrieved from
https://www.asha.org/public/speech/disorders/aac/
Andrews-Salvia, M., Roy, N., & Cameron, R. M. (2003). Evaluating the effects of memory
books for individuals with severe dementia. Journal of Medical Speech-Language
Pathology, 11(1), 51-59. doi:10.1080/14417040510001669131
Antelius, E., Kiwi, M., & Strandroos, L. (2018). Ethnographic methods for understanding
practices around dementia among culturally and linguistically diverse people. In J.
Keady, L-C. Hydén, A. Johnson, & C. Swarbrick (Eds.). Social Research Methods in
Dementia Studies: Inclusion and Innovation (pp. 121-139). London: Routledge.

143
© University of Pretoria
Arksey, H., & O’Malley, L. (2005). Scoping studies: Towards a methodological framework.
International Journal of Social Research Methodology: Theory and Practice, 8(1), 19-
32. doi:10.1080/1364557032000119616
Astell, A. J., Ellis, M. P., Bernardi, L., Alm, N., Dye, R., Gowans, G., & Campbell, J. (2010).
Using a touch screen computer to support relationships between people with dementia
and caregivers. Interacting with Computers, 22 (4), 267-275. doi:
10.1016/j.intcom.2010.03.003
Astell, A. J., Savundranayagam, M. Y., Kelson, E., Purves, B., & Phinney, A. (2018).
Fostering resilience in dementia through narratives: Contributions of multimedia
technologies. In B. Resnick, L. Gwyther, & K. Roberto (Eds.). Resilience in Aging (pp.
245–260). Cham: Springer. doi:10.1007/978-3-030-04555-5_13
Attainment Company. (2012). GoTalk NOW [Mobile application software]. Retrieved
from https://www.attainmentcompany.com/gotalk-now
Avella, J. R. (2016). Delphi panels: Research design, procedures, advantages, and challenges.
International Journal of Doctoral Studies, 11, 305–321. doi:10.28945/3561
Bakhurst, D., & Shanker, S.G. (2001). "Introduction: Bruner's way". In D. Bakhurst, & S.G.
Shanker (Eds.), Jerome Bruner: Language, Culture, Self (pp. 1-18). London: Sage
Publications Ltd. doi: 10.4135/9781446217634
Bartlett, J. E., & DeWeese, B. (2015). Using the Q methodology approach in human resource
development research. Advances in Developing Human Resources, 17(1), 72-87.
doi:10.1177/1523422314559811
Bartlett, R. (2014). Citizenship in action: The lived experiences of citizens with dementia
who campaign for social change. Disability & Society, 29(8), 1291-1304.
doi:10.1080/09687599.2014.924905
Bartlett, R., Windemuth-Wolfson, L., Oliver, K., & Dening, T. (2017). Suffering with
dementia: The other side of “living well”. International Psychogeriatrics, 29(2), 177-
179. doi:10.1017/s104161021600199x
Barton, E., Meadan-Kaplansky, H., & Ledford, J. (2018). Independent variables, fidelity, and
social validity. In J. Ledford, & D. Gast (Eds.), Single Case Research Methodology. (pp.
133-157). London: Routledge.
Baxter, S., Enderby, P., Evans, P., & Judge, S. (2012). Interventions using high-technology
communication devices: A state of the art review. Folia Phoniatrica et Logopaedica,
64(3), 137-144. doi;10.1159/000338250
Behr, D., Meitinger, K., Braun, M., & Kaczmirek, L. (2017). Web probing - implementing

144
© University of Pretoria
 probing techniques from cognitive interviewing in web surveys with the goal to assess
the validity of survey questions (Version 1.0). Mannheim: GESIS - Leibniz-Institut für
Sozialwissenschaften. doi:10.15465/gesissg_en_023
Birt, L., Griffiths, R., Charlesworth, G., Higgs, P., Orrell, M., & Poland, F. (2020).
Maintaining social connections in dementia: A qualitative synthesis. Qualitative Health
Research, 30(1), 23-42. doi:10.1177/1049732319874782
Blackstone, S. W., Williams, M. B., & Wilkins, D. P. (2007). Key principles underlying
research and practice in AAC. Augmentative and Alternative Communication, 23(3),
191-203. doi:10.1080/07434610701553684
Boivin, A., Richards, T., Forsythe, L., Grégoire, A., L’Espérance, A., Abelson, J., & Carman,
K. L. (2018). Evaluating patient and public involvement in research. BMJ (Online), 363
(December), 16-17. doi:10.1136/bmj.k5147
Bosco, A., Schneider, J., Coleston-Shields, D. M., & Orrell, M. (2019). Dementia care
model: Promoting personhood through co-production. Archives of Gerontology and
Geriatrics, 81 (April), 59-73. doi:10.1016/j.archger.2018.11.003
Bourgeois, M. S. (1990). Enhancing conversation skills in patients with Alzheimer's disease
using a prosthetic memory aid. Journal of Applied Behavior Analysis, 23(1), 29-42.
doi:10.1901/jaba.1990.23-29
Bourgeois, M. S. (1993). Effects of memory aids on the dyadic conversations of individuals
with dementia. Journal of Applied Behavior Analysis, 26 (1), 77-87.
doi:10.1901/jaba.1993.26-77
Bourgeois, M. S., Camp, C. J., Antenucci, V., & Fox, K. (2016). VoiceMyChoice™:
Facilitating understanding of preferences of residents with dementia. Advances in
Aging Research, 5(6), 131-141. doi:10.4236/aar.2016.56013
Bourgeois, M. S., Dijkstra, K., Burgio, L., & Allen-Burge, R. (2001). Memory aids as an
augmentative and alternative communication strategy for nursing home residents with
dementia. Augmentative & Alternative Communication, 17(3), 196-210.
doi: 10.1080/714043383
Bourgeois, M., Fried-Oken, M., & Rowland, C. (2010). AAC strategies and tools for persons
with dementia. ASHA Leader, 15(3), 8-11. doi:10.1044/leader.FTR1.15032010.8
Bourgeois, M.S., & Hickey, E.M. (2018). Introduction: history and philosophy of treatment
in dementia. In E.M. Hickey, & M. S. Bourgeois (Eds.), Dementia Person-Centered
Assessment and Intervention. (pp. 1-10). New York, NY: Routledge.

145
© University of Pretoria
Bourgeois, M. S., & Mason, L. (1996). Memory wallet intervention in an adult day-care
setting. Behavioral Interventions, 11(1), 3-18. doi:10.1002/(SICI)1099-
078X(199601)11:1<3:AID-BRT150>3.0.CO;2-0
Brossard Saxell, T., Ingvert, M., & Lethin, C. (2019). Facilitators for person-centred care of
inpatients with dementia: A meta-synthesis of registered nurses’ experiences. Dementia,
0(0), 1-25. doi:10.1177/1471301219871408
Bowie, P., & Mountain, G. (1993). Using direct observation to record the behaviour of long
stay patients with dementia. International Journal of Geriatric Psychiatry, 8(10), 857-
64. doi:10.1002/gps.930081009
Brady, N. C., Fleming, K., Romine, R. S., Holbrook, A., Muller, K., & Kasari, C. (2018).
Concurrent validity and reliability for the communication complexity scale. American
Journal of Speech-Language Pathology, 27(1), 237-246. doi:10.1044/2017_ajslp-17-
0106
Brewer, J.D., & Hunter, A. (2006). Foundations of multimethod research: Synthesizing
styles. Thousand Oaks, CA: Sage Publications.
Brock, K. L., Koul, R., Corwin, M., & Schlosser, R. W. (2019). The psychometric properties
of the communicative competence scale for individuals with Aphasia using speech-
generating devices. Aphasiology, 33(5), 520-543. doi:10.1080/02687038.2018.1561639
Brooker, D. (2004). What is person-centred care in dementia? Reviews in Clinical
Gerontology, 13(3), 215-222. doi:10.1017/S095925980400108X
Brooker, D., & Latham, I. (2015). Person-centred dementia care: Making services better with
the VIPS framework. London: Jessica Kingsley.
Broomfield, K., Harrop, D., Judge, S., Jones, G., & Sage, K. (2019). Appraising the quality
of tools used to record patient-reported outcomes in users of augmentative and
alternative communication (AAC): A systematic review. Quality of Life Research,
28(10), 2669-2683. doi:10.1007/s11136-019-02228-3
Broughton, M., Smith, E. R., Baker, R., Angwin, A. J., Pachana, N. A., Copland, D. A., …
Chenery, H. J. (2011). Evaluation of a caregiver education program to support
memory and communication in dementia: A controlled pretest–posttest study with
nursing home staff. International Journal of Nursing Studies, 48(11), 1436-1444.
doi:10.1016/j.ijnurstu.2011.05.007
Brown, S. R. (1980). Political subjectivity applications of Q-methodology in political
science. New Haven, CT: Yale University Press.
Brownie, S., & Nancarrow, S. (2013). Effects of person-centered care on residents and staff

146
© University of Pretoria
 in aged-care facilities: A systematic review. Clinical Interventions in Aging, 8(1), 1-10.
doi:10.2147/CIA.S38589
Bruner, J. (1987). The role of dialogue in language acquisition. In A. Sinclair, R.J. Jarvella &
W. J. Levell (Eds.).,The child’s conception of language (pp. 241-256). New York,
NY: Springer-Verlag.
Bujang, M. A., & Baharum, N. (2017). Guidelines of the minimum sample size requirements
for Cohen’s Kappa. Epidemiology Biostatistics and Public Health, 14(2), e12267-1-
e12267-10.doi:10.2427/12267
Chang, W. Z. D. (2011). Effects of memory aids on the conversations of elderly Chinese
persons. Master's thesis, The Ohio State University, Ohio: USA.
Chang, W. Z. D. (2015). Effects of visual stimuli on decision-making capacity of people with
dementia for end-of-life care. Doctoral dissertation, The Ohio State University, Ohio:
USA.
Chang, W. Z. D., & Bourgeois, M. S. (2020). Effects of visual aids for end-of-life care on
decisional capacity of people with dementia. American Journal of Speech-Language
Pathology, 29(1), 185-200. doi:10.1044/2019_AJSLP-19-0028
Clark, H.H. (1996). Using language. New York: Cambridge University Press.
Cohen, M. L., & Hula, W. D. (2020). Patient-reported outcomes and evidence-based practice
in speech-language pathology. American Journal of Speech-Language Pathology, 29(1),
357-370. doi:10.1044/2019_AJSLP-19-00076
Cohen-Mansfield, J., Dakheel-Ali, M., & Marx, M. S. (2009). Engagement in persons with
dementia: The concept and its measurement. The American Journal of Geriatric
Psychiatry, 17(4), 299-307. doi:10.1097/jgp.0b013e31818f3a52
Cooney, A., & O’Shea, E. (2018). The impact of life story work on person-centred care for
people with dementia living in long-stay care settings in Ireland. Dementia, 18(7-8),
2731-2746. doi:10.1177/1471301218756123
Costello, J.M. (2016). Message banking, voice banking and legacy messages. Boston
Children's Hospital. Retrieved from https://www.childrenshospital.org/centers-and-
services/programs/a-e/als-augmentative-communication-program/protocol-of-
assessment-considerations/message-banking#
Cress, C. J., & King, J. M. (1999). AAC strategies for people with primary progressive
aphasia without dementia: Two case studies. Augmentative and Alternative
Communication, 15(4), 248-259. doi:10.1080/07434619912331278785

147
© University of Pretoria
Crete-Nishihata, M., Baecker, R.M., Massimi, M., Ptak, D., Campiogotto, R., Kaufman. M.,
... Black., S. (2012). Reconstructing the past: Personal memory technologies are not
just personal and not just for memory. Human-Computer Interaction, 27(1-2), 92-123.
doi: 10.1080/07370024.2012.656062
Critten, V., & Kucirkova, N. (2019). 'It brings it all back, all those good times; it makes me
go close to tears': Creating digital personalised stories with people who have dementia.
Dementia, 18(3), 864-882. doi: 10.1177/1471301217691162
Dada, S., Flores, K., Bastable, K., & Schlosser, R.W. (2020). The effects of augmentative and
alternative communication interventions on the receptive language skills of children
with developmental disabilities: A scoping review. International Journal of Speech-
Language Pathology, 1-11. doi:10.1080/17549507.2020.1797165
Dada, S., Hyman, C., May, A.A., & Murray, J. (2020). Intelligent assistive technology
devices for persons with dementia: A scoping review. (Manuscript submitted for
publication)
Dada, S., Stockley, N., Wallace, S. E., & Koul, R. (2019). The effect of augmented input on
the auditory comprehension of narratives for people with aphasia: A pilot investigation.
AAC: Augmentative and Alternative Communication, 35(2), 148-155.
doi:10.1080/07434618.2019.1576766
Dada, S., Tonsing, K., & Goldbart, J. (2020). Friendship experiences of young adults who use
Augmentative and Alternative Communication. International Journal of Disability,
Development and Education, 00(0), 1-25. doi:10.1080/1034912X.2020.1746246
Davidson, B., Worrall, L., & Hickson, L. (2008). Exploring the interactional dimension of
social communication: A collective case study of older people with aphasia.
Aphasiology, 22(3), 235-257. doi:10.1080/02687030701268024
Davies, N., Hopwood, J., Walker, N., Ross, J., Iliffe, S., Walters, K., & Rait, G. (2019).
Designing and developing a co-produced theoretical and evidence-based online support
for family caregivers of people with dementia at the end of life. BMC Palliative Care,
18(71), 1-17. doi:10.1186/s12904-019-0455-0
Davis, B. H., & Shenk, D. (2015). Beyond Reminiscence : Using generic video to elicit
conversational language. American Journal of Alzheimer’s Disease & Other Dementias,
30(1), 61-68. doi:10.1177/1533317514534759
de Medeiros, K., Saunders, P. A., & Sabat, S. R. (2012). Friendships and the social
environments of people with dementia: Introduction to the Special Issue. Dementia,
11(3), 281-285. doi:10.1177/1471301211421062

148
© University of Pretoria
Dietz, A., Knollman-Porter, K., Toth, K., & Brown, B. (2014). Supported reading
comprehension for people with aphasia : Visual and linguistic supports. Journal of
Medical Speech-Language Pathology, 21(4), 319-331.
DeTure, M. A., & Dickson, D. W. (2019). The neuropathological diagnosis of Alzheimer’s
disease. Molecular Neurodegeneration, 14(1), 1-18. doi:10.1186/s13024-019-0333-5
Dewing, J. (2008a). Process consent and research with older persons living with dementia.
Research Ethics, 4(2), 59-64. doi:10.1177/174701610800400205
Dewing, J. (2008b). Personhood and dementia: Revisiting Tom Kitwood’s ideas.
International Journal of Older People Nursing, 3(1), 3-13. doi:10.1111/j.1748-
3743.2007.00103.x
Dewing, J. (2019). On being a person: Commentary. In D. Brooker (Eds.), Dementia
Reconsidered: The Person Still Comes First (2nd ed., pp. 17-23). London, England:
Open University Press.
Dewing, J., & McCormack, B. (2017). Tell me, how do you define person-centredness?
Journal of Clinical Nursing, 26(17-18), 42-49. doi:10.1111/jocn.13681
Dodd, B. (2007). Evidence-based practice and speech-language pathology: Strengths,
weaknesses, opportunities and threats. Folia Phoniatrica et Logopaedica, 59(3), 118-
129. doi: 10.1159/000101770
Dollaghan C. A. (2007). The Handbook for Evidence-Based Practice in Communication
Disorders. Baltimore, MD: Brookes Publishing.
Dollaghan, C. A. (2004). Evidence-based practice in communication disorders: What do we
know, and when do we know it? Journal of Communication Disorders, 37(5), 391-400.
doi.org/10.1016/j.jcomdis.2004.04.002
Doran, C., Noonan, M., & Doody, O. (2018). Life-story work in long-term care facilities for
older people : An integrative review. Journal of Clinical Nursing, 28(7-8), 1070-1084.
doi.org/10.1111/jocn.14718
Douglas, N. F., Brush, J., Bourgeois, M. (2018). Person-centered, skilled services using a
montessori approach for persons with dementia. Seminars in Speech and Language,
39(3), 223-230. doi:10.1055/s-0038-1660781
Dynes, K. (2018). iPad use to enhance conversations between persons with dementia and
their family caregivers. Master’s thesis, University of Western Ontario: Canada.
Edgar, D., Wilson, V., & Moroney, T. (2020). Which is it, person-centred culture, practice or
care? It matters. International Practice Development Journal, 10(1), 1-17.
doi:10.19043/ipdj.101.008

149
© University of Pretoria
Elfrink, T. R., Zuidema, S. U., Kunz, M., & Westerhof, G. J. (2018). Life story books for
people with dementia: A systematic review. International Psychogeriatrics, 30(12),
1797-1811. doi:10.1017/s1041610218000376
Ellis, M., & Astell, A. (2017). Communicating with people living with dementia who are
nonverbal: The creation of Adaptive Interaction. PLOS ONE, 12(8), e0180395.
doi:10.1371/journal.pone.0180395
Elwyn, G., Edwards, A., Wensing, M., Hood, K., Atwell, C., & Grol, R. (2003). OPTION
Scale. PsycTESTS Dataset. doi:10.1037/t21722-000
Ekström, A., Ferm, U., & Samuelsson, C. (2017). Digital communication support and
Alzheimer’s disease. Dementia, 16(6), 711-731. doi:10.1177/1471301215615456
Ekström, A., Lindholm, C., Majlesi, A. R., & Samuelsson, C. (2017). Communication and
collaboration in dementia. In L-C, Hydén, E. Antelius (Eds.), Living with Dementia. (pp.
93-115). London: Palgrave.
Enderby, P. (2014). Introducing the therapy outcome measure for AAC services in the
context of a review of other measures. Disability and Rehabilitation: Assistive
Technology, 9(1), 33-0. doi;10.3109/17483107.2013.823576
Fager, S. K., Fried-Oken, M., Jakobs, T., & Beukelman, D. (2019). New and emerging access
technologies for adults with complex communication needs and severe motor
impairments: State of the science. Augmentative and Alternative Communication, 35(1),
13-25. doi:10.1080/07434618.2018.1556730
Fazio, S., Pace, D., Flinner, J., & Kallmyer, B. (2018). The fundamentals of person-centered
care for individuals with dementia. Gerontologist, 58(1), S10-S19.
doi:10.1093/geront/gnx122
Fereday, J., & Muir-Cochrane, E. (2006). Demonstrating rigor using thematic analysis: A
hybrid approach of inductive and deductive coding and theme Development.
International Journal of Qualitative Methods, 5(1), 80-92.
doi:10.1177/160940690600500107
Ferm, U., Ekström, A., Larsson, E., & Samuelsson, C. (2020). Tablet computer-supported
conversation between people with dementia and their carers: technology as interactional
focus. Universal Access in the Information Society, July(0), 1-13. doi:10.1007/s10209-
020-00745-4
Fried-Oken, M., Mooney, A., & Peters, B. (2015). Supporting communication for patients
with neurodegenerative disease. NeuroRehabilitation, 37(1), 69-87. doi:10.3233/NRE-
151241

150
© University of Pretoria
Fried-Oken, M., Rowland, C., Baker, G., Dixon, M., Mills, C., Schultz, D., & Oken, B.
(2009). The effect of voice output on AAC-supported conversations of persons with
alzheimer’s disease. ACM Transactions on Accessible Computing, 1(3), 1-11.
doi:10.1145/1497302.1497305
Fried-Oken, M., Rowland, C., Daniels, D., Dixon, M., Fuller, B., Mills, C., ... & Oken, B.
(2012). AAC to support conversation in persons with moderate Alzheimer's disease.
Augmentative & Alternative Communication, 28(4), 219-231.
doi: 10.3109/07434618.2012.732610
Gast, D., & Ledford (2018). Research approaches in applied settings. In J. Ledford, & D.
Gast (Eds.), Single Case Research Methodology. (pp. 1-26). London: Routledge.
Gibson, B. E., Terry, G., Setchell, J., Bright, F. A. S., Cummins, C., & Kayes, N. M. (2019).
The micro-politics of caring: Tinkering with person-centered rehabilitation. Disability
and Rehabilitation, 14(2), 1529-1538. doi:10.1080/09638288.2019.1587793
Giezendanner, S., Monsch, A. U., Kressig, R. W., Mueller, Y., Streit, S., Essig, S., Zeller, A.,
& Bally, K. (2019). General practitioners’ attitudes towards early diagnosis of dementia:
A cross-sectional survey. BMC Family Practice, 20(1), 1-9. doi:10.1186/s12875-019-
0956-1
Gómez Taibo, M. L., Parga Amado, P., Canosa Domínguez, N., Vieiro Iglesias, P., & García
Real, T. (2014). Conversations about self-identity in Alzheimer disease: Augmentative
and alternative Communication memory books as an aid. Revista de Logopedia,
Foniatría y Audiología, 34(2), 60-67. doi:10.1016/j.rlfa.2013.04.008
Graneheim, U. H., & Lundman, B. (2004). Qualitative content analysis in nursing research:
Concepts, procedures and measures to achieve trustworthiness. Nurse Education Today,
24(2), 105-112. doi:10.1016/j.nedt.2003.10.001
Granlund, M., & Blackstone, S. (1999). Outcomes measurement in AAC. In L. F. J. Clibbens,
H. Arvidson & L. Lloyd (Eds.), Augmentative and alternative communication: New
directions in research and practice (pp. 207-228.). London: Whurr.
Granlund, M., & Olsson, C. (1999). Efficacy of communication intervention for presymbolic
communicators. Augmentative and Alternative Communication, 15(1), 25-37.
doi:10.1080/07434619912331278545
Granlund, M., & Wilder, J. (2006). Studying interaction between children who do not use
symbols in interaction and their parents within the family system-methodological
challenges. Disability and Rehabilitation, 28(3), 175-182.
doi:10.1080/09638280500077903

151
© University of Pretoria
Grant, M. J., & Booth, A. (2009). A typology of reviews: An analysis of 14 review types and
associated methodologies. Health Information and Libraries Journal, 26(2), 91-108.
doi:10.1111/j.1471-1842.2009.00848.x
Gridley, K., Birks, Y., & Parker, G. (2020). Exploring good practice in life story work with
people with dementia: The findings of a qualitative study looking at the multiple views
of stakeholders. Dementia, 19(2), 182-194. doi:10.1177/1471301218768921
Gridley, K., Brooks, J., Birks, Y., Baxter, K., & Parker, G. (2016). Improving care for people
with dementia: Development and initial feasibility study for evaluation of life story
work in dementia care. Health Services and Delivery Research, 4(23), 1-298.
doi:10.3310/hsdr04230
Grøndahl, V. A., Persenius, M., Bååth, C., & Helgesen, A. K. (2017). The use of life stories
and its influence on persons with dementia, their relatives and staff - A systematic mixed
studies review. BMC Nursing, 16(28), 1-11.doi:10.1186/s12912-017-0223-5
Haapala, I., Carr, A., & Biggs, S. (2019). What would I want? Dementia perspectives and
priorities among people with dementia, family carers and service professionals.
International Journal of Care and Caring, 3(1), 39-57.
doi:10.1332/239788218X15321005902545
Hall, K., Lund. C., Young, J., Okell, E., & Van Steenbrugge, W. (2018). Familiar
communication partners’ facilitation of topic management in conversations with
individuals with dementia. International Journal of Language & Communication
Disorders, 53(3), 564-575. doi:10.1111/1460-6984.12369
Hamel, A. V., Sims, T. L., Klassen, D., Havey, T., & Gaugler, J. E. (2016). Memory matters:
A mixed-methods feasibility study of a mobile aid to stimulate reminiscence in
individuals with memory loss. Journal of Gerontological Nursing, 42(7), 15-24.
doi:10.3928/00989134-20160201
Harding, A. J. E., Morbey, H., Ahmed, F., Opdebeeck, C., Lasrado, R., Williamson, P. R.,…
Reilly, S. T. (2019). What is important to people living with dementia ?: The ‘ long-list ’
of outcome items in the development of a core outcome set for use in the evaluation of
non- pharmacological community-based health and social care interventions. BMC
Geriatrics, 19(1), 1-12. doi:10.1186/s12877-019-1103-
Hashim,A.H.A., Ismail,A.N., Rias,R.M, & Mohamed, A. (2015). The development of an
individualized digital memory book for Alzheimer’s disease patient: a case study. In
International Symposium Technology Management and Emerging Technologies
(ISTMET), 2015 (pp. 227–232). Langkawi Island, Malaysia: IEEE. doi:

152
© University of Pretoria
 10.1109/ISTMET.2015.7359034.
Hasson, F., Keeney, S., & McKenna, H. (2000). Research guidelines for the delphi survey
technique. Journal of Advanced Nursing, 32(4), 1008-1015. doi:10.1046/j.1365-
2648.2000.t01-1-01567.x
Hendrix, J., Feng, Y., van Otterdijk, M., & Barakova, E. (2019). Adding a context: Will it
influence human-robot interaction of people living with dementia? Social Robotics,
ICSR 2019, Lecture Notes in Computer Science, 11876, 494-504. doi:10.1007/978-3-
030-35888-4
Hennelly, N., & O’Shea, E. (2017). Personhood, dementia policy and the Irish national
dementia strategy. Dementia, 18(5), 1810-1825. doi:10.1177/1471301217729232
Hesse-Biber, S. (2015). Mixed Methods Research. Qualitative Health Research, 25(6), 775-
788. doi:10.1177/1049732315580558
Hickey, E. M. & Bourgeois, M.S. (2018). Cognitive-communicative interventions: Profiling
types of dementia. In M.S. Bourgeois & E.M. Hickey (Eds.). Dementia Person-
Centred Assessment and Intervention (pp. 46-80). New York: Routledge.
Hickey, E. M., Kinder, R., Khayum, B., Douglas, N., & Bourgeois, M.S. (2018). Setting the
stage for person-centered care. Intervention principles and practical considerations. In
M.S. Bourgeois & E.M. Hickey (Eds.). Dementia Person-Centred Assessment and
Intervention (pp. 81-112). New York: Routledge.
Hoerster, L., Hickey, E. M. &, Bourgeois, M. S. (2001). Effects of memory aids on
conversations between nursing home residents with dementia and nursing assistants
and nursing assistants. Neuropsychological Rehabilitation, 11(3), 399-427.
doi: 10.1080/09602010042000051
Hopper, T., Hickey, E. M., & Bourgeois, M. S. (2018). Clinical and pathophysiology profiles
of various dementia etiologies. In E. M. Hickey, & M. S. Bourgeois (Eds.), Dementia
Person-Centered Assessment and Intervention. (pp. 11-41). New York, NY: Routledge.
Hopper, T. L. (2003). “They’re just going to get worse anyway”: Perspectives on
rehabilitation for nursing home residents with dementia. Journal of Communication
Disorders, 36(5), 345-359. doi:10.1016/S0021-9924(03)00050-9
Howlett, O., McKinstry., C, & Lannin. (2018). Using the cognitive interviewing process to
improve design by allied health: A qualitative study. Australian Occupational Therapy
Journal, 65(2), 126-134. doi:10.1111/1440-1630.12445
Hubbard, G., Downs, M.G,. & Tester, S. (2003). Including older people with dementia in
research: Challenges and strategies. Ageing & Mental Health, 7(5), 351-362.

153
© University of Pretoria
 doi:10.1080/1360786031000150685
Hung, L., & Chaudhury, H. (2011). Exploring personhood in dining experiences of residents
with dementia in long-term care facilities. Journal of Aging Studies, 25(1), 1-12.
doi:10.1016/j.jaging.2010.08.007
Hung, L., Chow, B., Shadarevian, J., O’Neill, R., Berndt, A., Wallsworth, C., … Chaudhury,
H. (2020). Using touchscreen tablets to support social connections and reduce
responsive behaviours among people with dementia in care settings: A scoping review.
Dementia, (0)0, 1-20. doi:10.1177/1471301220922745
Hunter, A., & Brewer, J. (2015). Designing multimethod research. In S. N. Hesse-Biber, &
Johnson, R.B. (Eds.), The Oxford Handbook of Multimethod and Mixed Methods
Research Inquiry, (pp. 185-205). Oxford University Press, Oxford.
Hydén, L-C. (2011). Non-verbal vocalizations, dementia and social interaction.
Communication and Medicine, 8(2), 135-144. doi:10.1558/cam.v8i2.135
Hydén, L-C., & Antelius, E. (2017). Introduction: From empty vessels to active agents. In L-
C, Hydén, E. Antelius (Eds.), Living with Dementia. (pp. 1-13). London: Palgrave.
Hydén, L-C., Swarbrick, C., Johnson, A., & Keady, J. (2018). Conclusion: Messages and
futures in social science methods in dementia studies. In J. Keady, L-C. Hydén, A.
Johnson, & C. Swarbrick (Eds.), Social Research Methods in Dementia Studies
Inclusion and Innovation. (pp. 222-224). London: Routledge.
Johnson, H. (2020). A voice for research and a voice for life. Providing participant
communication supports. Commentary on “Social researchers and participants with
intellectual disabilities and complex communication (access) needs. Whose capacity?
Whose competence?”. Research and Practice in Intellectual and Developmental
Disabilities, 7(2), 144-151. doi:10.1080/23297018.2020.1803119
Jones, C., Sung, B., & Moyle, W. (2018). Engagement of a person with dementia scale:
Establishing content validity and psychometric properties. Journal of Advanced
Nursing, 74(9), 2227-2240. doi: 10.1111/jan.13717
Kaushik, V., & Walsh, C. A. (2019). Pragmatism as a research paradigm and its implications
for Social Work research. Social Sciences, 8(9), 1-17. doi:10.3390/socsci8090255
Keady, J., Hydén, L-C., Johnson, A., & Swarbrick, C. (2018). Introduction. In J. Keady, L-C.
Hydén, A. Johnson, & C. Swarbrick (Eds.), Social Research Methods in Dementia
Studies Inclusion and Innovation. (pp. 1-7). London: Routledge.
Kent-Walsh, J., & Binger, C. (2018). Methodological advances, opportunities, and challenges
in AAC research. Augmentative and Alternative Communication, 34(2), 93-103.

154
© University of Pretoria
 doi:10.1080/07434618.2018.1456560
Kent-Walsh, J., Murza, K. A., Malani, M. D., Binger, C., Murza, K. A., Malani, M. D., &
Binger, C. (2015). Effects of communication partner instruction on the communication
of individuals using AAC : A meta-analysis. Augmentative and Alternative
Communication, 31(4), 271-284. doi.10.3109/07434618.2015.1052153
Kickbusch I., & Gleicher D. (2012). Governance for Health in the 21st Century.
Copenhagen: World Health Organization Regional Office for Europe; 2012. Available
from: http://www.euro .who.int/__data/assets/pdf_file/ 0019/171334/RC62BD01-
Governance-for-Health-Web.pdf
Kindell, J., Burrow, S., Wilkinson, R., & Keady, J. D. (2014). Life story resources in
dementia care: A review. Quality in Ageing and Older Adults, 15(3), 151-161.
doi:10.1108/QAOA-02-2014-0003
Kindell, J., Keady, J., Sage, K., & Wilkinson, R. (2017). Everyday conversation in dementia:
A review of the literature to inform research and practice. International Journal of
Language and Communication Disorders, 52(4), 392-406. doi.org/10.1111/1460-
6984.12298
Kindell, J., Sage, K., Keady, J., & Wilkinson, R. (2013). Adapting to conversation with
semantic dementia: Using enactment as a compensatory strategy in everyday social
interaction. International Journal of Language & Communication Disorders, 48(5),
497-507. doi:10.1111/1460-6984.12023
Kitwood, T. (1997). On being a person. In D. Brooker (Ed.), Dementia Reconsidered,
Revisited: The Person still Comes First, 2nd edition, (pp. 6-23). Open University
Press.
Klimova, B., & Kuca, K. (2016). Speech and language impairments in dementia. Journal of
Applied Biomedicine, 14(2), 97-103. doi:10.1016/j.jab.2016.02.002
Kogan, A. C., Wilber, K., & Mosqueda, L. (2016). Person-centered care for older adults with
chronic conditions and functional impairment: A systematic literature review. Journal of
the American Geriatrics Society, 64(1), 1-7. doi:10.1111/jgs.13873
Kullberg, A., & Odzakovic, E. (2018). Walking interviews as a research method with people
living with dementia in their local community. In J. Keady, L-C. Hydén, A. Johnson, &
C. Swarbrick (Eds.), Social Research Methods in Dementia Studies Inclusion and
Innovation. (pp. 23-37). London: Routledge.
Kuoch, H. & Mirenda, P. (2003). Social story interventions for young children with autism
spectrum disorders. Focus on Autism and other Developmental Disabilities, 18(4), 219-

155
© University of Pretoria
 227.
Lanzi, A., Bourgeois, M. S., & Dedrick, R. F. (2020). Development and initial validation of
the functional external memory aid tool. American Journal of Speech-Language
Pathology, 29(May), 1-14. doi:10.1044/2019_ajslp-19-00075
Lanzi, A., Burshnic, V., & Bourgeois, M. S. (2017). Person-centered memory and
communication strategies for adults with dementia. Topics in Language Disorders,
37(4), 361-374. doi:10.1097/TLD.0000000000000136
Ledford, J., Lane, J.D., & Gast, D. (2018). Dependent variables, measurement, and reliability.
In J. Ledford, & D. Gast (Eds.), Single Case Research Methodology. (pp. 97-132).
London: Routledge.
Lepore, M., Shauman, S. B., Wiener, J.M., & Gould, E. (2017). Challenges in involving
people with dementia as study participants in research on care and services. Research
Summit on Dementia Care. Building Evidence for Services and Supports. Retrieved
from https://aspe.hhs.gov/basic-report/challengesinvolving-people-dementia-study-
participants-research-care-and-services.
Light, J. (1988). Interaction involving individuals using augmentative and alternative
communication systems: State of the art and future directions. Augmentative &
Alternative Communication, 4(2) 66-82. doi:10.1080/07434618812331274657.
Light, J., McNaughton, D., Beukelman, D., Fager, S. K., Fried-Oken, M., Jakobs, T., &
Jakobs, E. (2019). Challenges and opportunities in augmentative and alternative
communication: Research and technology development to enhance communication and
participation for individuals with complex communication needs. Augmentative and
Alternative Communication, 35(1),1-12. doi:0.1080/07434618.2018.1556732
Littlechild, R., Tanner, D., & Hall, K. (2015). Co-research with older people: Perspectives on
impact. Qualitative Social Work, 14(1), 18-35. doi:10.1177/1473325014556791
Lutfallah, S., & Buchanan, L. (2019). Quantifying subjective data using online Q-
methodology software. The mental lexicon, 14(3), 415-423. doi:10.1075/ml20002.lut
Mabire, J., Gay, M., Vrignaud, P., Garitte, C., & Vernooij-dassen, M. (2016). Social
interactions between people with dementia:Pilot evaluation of an observational
instrument in a nursing home. International Psychogeriatrics, 28(6), 1005-1015.
doi:10.1017/S1041610215002483
Mahendra, N. (2019). Guest editorial. Speech, Language and Hearing, 22(1), 1-1.
doi: 10.1080/2050571X.2019.1579407
Mahendra, N., Fremont, K., & Dionne, E. (2013). Teaching future providers about dementia:

156
© University of Pretoria
 The impact of service learning. Seminars in Speech and Language, 34(1), 5-17.
doi:10.1055/s-0033-1337390
Mahendra, N., Hickey, E. M., & Bourgeois, M. S. (2018). Cognitive-communicative
characteristics: Profiling types of dementia. In M. S. Bourgeois, & E. M. Hickey
(Eds.). Dementia Person-Centred Assessment and Intervention (pp. 42-80). New
York: Routledge.
Mahendra, N., Scullion, A., & Hamerschlag, C. (2011). Cognitive-linguistic interventions for
persons with dementia. Topics in Geriatric Rehabilitation, 27(4), 278-288.
doi:10.1097/tgr.0b013e31821e5945
Mann, J., & Hung, L. (2019). Co-research with people living with dementia for change.
Action Research, 17(4), 573-590. doi:10.1177/1476750318787005
May, A. A., Dada, S., & Murray, J. (2019). Review of AAC interventions in persons with
dementia. International Journal of Language and Communication Disorders, 54(6),
857-874. doi:10.1111/1460-6984.12491
May, A. A., Dada, S., & Murray, J. (2020). Implementing AAC in persons with dementia. In
Clinical Cases in AAC. (Book chapter submitted for publication)
McCance, T., McCormack, B., & Dewing, J. (2011). An exploration of person-centredness in
practice. Online Journal of Issues in Nursing, 16(2), 1-9.
doi:10.3912/OJIN.Vol16No02Man01
McConnell, T., Sturm, T., Stevenson, M., McCorry, N., Donnelly, M., Taylor, B., & Best, P.
(2019). Coproducing a shared understanding and definition of empowerment with
people with dementia. Research Involvement and Engagement, 5(19), 2-11.
doi:10.1186/s40900-019-0154-2
McCormack, B., Dewing, J., & McCance, T. (2011). Developing person-centred care:
addressing contextual challenges through practice development. Online Journal of
Issues in Nursing, 16(2), 3. doi:10.3912/OJIN.Vol16No02Man03
McCormack, B., & McCance, T. (2017) Person Centred Practice in Nursing and Health Care.
West Sussex: Wiley Blackwell.
McEvoy, P., & Plant, R. (2014). Dementia care: Using empathic curiosity to establish the
common ground that is necessary for meaningful communication. Journal of Psychiatric
and Mental Health Nursing, 21(6), 477-482. doi.org/10.1111/jpm.12148
McHugh, M. L. (2012). Lessons in biostatistics interrater reliability: The kappa statistic.
Biochemica Medica, 22(3), 276-282.
McKeown, J., Clarke, A., Ingleton, C., Ryan, T., & Repper, J. (2010). The use of life story

157
© University of Pretoria
 work with people with dementia to enhance person-centred care. International Journal
of Older People Nursing, 5(2), 148-158. doi:10.1111/j.1748-3743.2010.00219.x
McKeown, J., Clarke, A., & Repper, J. (2006). Life story work in health and social care:
Systematic literature review. Journal of Advanced Nursing, 55(2), 237-247.
doi:10.1111/j.1365-2648.2006.03897.x
McKeown, B., & Thomas, D. (2013). Communication concourses, Q samples, and conditions
of instruction. Sage Research Methods (pp. 17-30). Sage.
doi:10.4135/9781483384412.n2
McLaughlin., H. (2020). Who are the service users? Language, neo-liberalism and social
constructions. In H. McLaughlin, P. Beresford, C. Cameron, H. Casey, & J. Duffy
(Eds.), The Routledge Handbook of Service User Involvement in Human Sciences
Research and Education. (pp. 30-41). London: Routledge. doi: 10.4324/9780429433306
McMillan, J., & Schumacher, S. (2014). Research in education: Evidence-based inquiry (7th
ed.). New York, NY: Pearson.
McNaughton, D., Light, J., Beukelman, D. R., & Klein, C., Nieder, D., & Nazareth, G.
(2019). Building capacity in AAC: A person-centred approach to supporting
participation by people with complex communication needs. Augmentative and
Alternative Communication, 0(0), 1-13. doi:10.1080/07434618.2018.1556731
McPherson, A., Furniss, F. G., Sdogati, C., Cesaroni, F., Tartaglini, B., & Lindesay, J.
(2001). Effects of individualized memory aids on the conversation of persons with
severe dementia: A pilot study. Aging & Mental Health, 5(3), 289-294.
doi:10.1080/13607860120064970
Mooney, A., Beale, N., Fried-oken, M. (2018a). Group communication treatment for
individuals with PPA and their partners. Seminars in Speech and Language, 39(3), 257-
260. doi:10.1055/s-0038-1660784
Mooney, A., Bedrick, S., Noethe, G., Spaulding, S., & Fried-Oken, M. (2018b). Mobile
technology to support lexical retrieval during activity retell in primary progressive
aphasia. Aphasiology, 32(6), 666-692. doi:10.1080/02687038.2018.1447640
Morello, A., Lima, T. M., & Brandão, L. (2017). Language and communication non-
pharmacological interventions in patients with Alzheimer's disease: A systematic
review. Dementia Neuropsychology, 11(3), 227-241. doi:10.1590/1980-57642016dn11-
030004
Morin, K. L., Ganz, J. B., Gregori, E. V, Foster, M. J., Gerow, S. L., Genç-tosun, D., Hong,
E. R. (2018). A systematic quality review of high-tech AAC interventions as an

158
© University of Pretoria
 evidence-based practice. Augmentative and Alternative Communication, 34(2), 104-117.
doi:10.1080/07434618.2018.1458900
Morris, L., Horne, M., McEvoy, P., & Williams,T. (2017). Communication training
interventions for family and professional carers of people living with dementia: A
systematic review of effectiveness, acceptability and conceptual basis. Aging &
Mental Health, 22(7), 863-880. doi: 10.1080/13607863.2017.1399343
Morse, J. M. (2003). Principles of Mixed Methods and Multimethod Research Design. In
A.Tashakkori & C. Teddlie (Eds.),Handbook of Mixed Methods in Social and Behavioral
Research. Thousand Oaks: Sage Publications
Moulam, L., Meredith, S., Whittle, Lynch, Y., & Murray, J. (2020). Augmented
communication: Patient and public involvement in research: Rhetoric and reality. In H.
McLaughlin, P. Beresford, C. Cameron, H. Casey, & J. Duffy (Eds.), The Routledge
Handbook of Service User Involvement in Human Sciences Research and Education.
(pp. 427-439). London: Routledge. doi:10.4324/9780429433306
Moyle, W., Jones, C. J., Murfield, J. E., Thalib, L., Beattie, E. R. A., Shum, D. K. H., …
Draper, B. M. (2017). Use of a robotic seal as a therapeutic tool to improve dementia
symptoms: A cluster-randomized controlled trial. Journal of the American Medical
Directors Association, 18, 766-773. doi:10.1016/j.jamda.2017.03.018
Moyle, W., Jones, C., Sung, B., Bramble, M., O’Dwyer, S., Blumenstein, M., & Estivill-
Castro, V. (2015). What effect does an animal robot called CuDDler have on the
engagement and emotional response of older people with dementia? A pilot feasibility
study. International Journal of Social Robotics, 8, 145-156. doi:10.1007/s12369-015-
0326-7M
Murphy, J, Gray, C. M., van Achterberg, T., Wyke, S., & Cox, S. (2010). The effectiveness
of the talking mats framework in helping people with dementia to express their views on
well-being. Dementia, 9(4), 454-472. doi:10.1177/1471301210381776
Murphy, J., & Oliver, T. (2013). The use of talking mats to support people with dementia and
their carers to make decisions together. Health & Social Care in the Community, 21(2),
171-180. doi: 10.1111/hsc.12005
Nasreddine, Z. S., Phillips, N. A., Bédirian, V., Charbonneau, S., Whitehead, V., Collin, I., ...
& Chertkow, H. (2005). The Montreal Cognitive Assessment, MoCA: A brief
screening tool for mild cognitive impairment. Journal of the American Geriatrics
Society, 53(4), 695-699. doi: 10.1111/j.1532-5415.2005.53221. x.
Nedlund, A-C., & Bartlett, R. (2017). Citizenship-in-and-as practice: A framework for

159
© University of Pretoria
 improving life with dementia. In L-C. Hydén, & E. Antelius (Eds.), Living with
Dementia. (pp. 49-64). London: Palgrave.
Newman, I., & Ramlo, S. (2010). Using Q methodology and Q factor analysis to facilitate
mixed methods research. In A. Tashakkori, & C. Teddie (Eds.), Handbook of Mixed
Methods in Social and Behavioural Research (pp. 505-530). Thousand Oaks:Sage.
Nicol, G. E., Piccirillo, J. F., Mulsant, B. H., & Lenze, E. J. (2020). Action at a distance:
Geriatric research during a pandemic. Journal of the American Geriatrics Society, 68(5),
922-925. doi:10.1111/jgs.16443
O’Neill, T., Light, J., & Pope, L. (2018). Effects of interventions that include aided
augmentative and alternative communication input on the communication of individuals
with complex communication needs: A meta-analysis. Journal of Speech, Language,
and Hearing Research, 61(7), 1743-1765. doi:10.1044/2018_JSLHR-L-17-0132
Passalacqua, S. A., & Harwood, J. (2012). VIPS communication skills training for
paraprofessional dementia caregivers: An intervention to increase person-centered
dementia care. Clinical Gerontologist, 35(5), 425-445.
doi:10.1080/07317115.2012.702655
Patterson, K. M., Clarke, C., Wolverson, E. L., & Moniz-Cook, E. D. (2017). Through the
eyes of others - the social experiences of people with dementia: A systematic literature
review and synthesis. International Psychogeriatrics, 30 (6), 791-805.
doi:10.1017/S1041610216002374
Pearce, L.D. (2015). Thinking outside the Q boxes: Further motivating a mixed research
perspective. In S. Hesse-Bieber & R. B. Johnson (Eds.), Oxford handbook of
multimethod and mixed methods research inquiry (pp. 42-56). New York, NY: Oxford
University Press
Pennington, L., Goldbart, J., & Marshall, J. (2004). Interaction training for conversational
partners of children with cerebral palsy : A systematic review. International Journal of
Language & Communication Disorders, 39(2), 151-170.
doi:10.1080/13682820310001625598
Perion, J., & Steiner, V. (2017). Perceptions of reciprocity in friendship by community
dwelling people with mild to moderate dementia. Dementia, 0(0), 1-15.
doi:10.1177/1471301217742503
Person, M., & Hanssen, I. (2015). Joy, happiness, and humor in dementia care: A Qualitative
Study. Creative Nursing, 21(1), 47-52. doi:10.1891/1078-4535.21.1.47
Pluut, B. (2016). Differences that matter: developing critical insights into discourses of

160
© University of Pretoria
 patient-centeredness. Medicine, Health Care and Philosophy, 19(4), 501-515.
doi:10.1007/s11019-016-9712-7
Purves, B. A., Phinney, A., Hulko, W., Puurveen, G., & Astell, A. J. (2015). Developing
CIRCA-BC and exploring the role of the computer as a third participant in
conversation. American Journal of Alzheimer's Disease and Other Dementias, 30(1),
101-107. doi: 10.1177/1533317514539031
Realpe, A., & Wallace, L.M. (2010). What is Co-production? London: The Health
Foundation. Retrieved from
https://improve.bmj.com/sites/default/files/resources/what_is_co-production.pdf
Rebstock, A. M., & Wallace, S. E. (2020). Effects of a combined semantic feature analysis
and multimodal treatment for primary progressive aphasia: Pilot study. Communication
Disorders Quarterly, 41(2), 71-85. doi:10.1177/1525740118794399
Reitz, C., & Dalemans, R. (2016). The use of “Talking Mats” by persons with alzheimer in
the Netherlands: Increasing shared decision‐making by using a low‐ tech
communication aid. Journal of Social Inclusion, 7(2), 35. doi:10.36251/josi.110
Republic of South Africa. (2020). Disaster management act: regulations: alert level 4 during
Coronavirus COVID-19 lockdown. Retrieved from: www.gov.za/covid-
19/about/disaster-management-act-regulations-alert-level-4-during-coronavirus-covid-
19-lockdown
Rogers, C. R. (1958). The characteristics of a helping relationship. Personnel & Guidance
Journal, 37, 6-16. doi:10.1002/j.2164-4918.1958.tb01147.x
Røsvik, J., Brooker, D., Mjorud, M., & Kirkevold, Ø. (2013). What is person-centred care in
dementia? Clinical reviews into practice: The development of the VIPS practice model.
Reviews in Clinical Gerontology, 23(2), 155-163. doi:10.1017/s0959259813000014
Roulstone, S. (2015). Exploring the relationship between client perspectives, clinical
expertise and research evidence. International Journal of Speech-Language Pathology,
17(3), 211-221. doi:10.3109/17549507.2015.1016112
Rousseau, D. M., & Gunia, B. C. (2016). Evidence-based practice: The psychology of EBP
implementation. Annual Review of Psychology, 67(1), 667-692. doi:10.1146/annurev-
psych-122414-033336
Ruggero, L., Croot, K., & Nickels, L. (2020). How evidence-based practice (E3BP) informs
speech-language pathology for primary progressive aphasia. American Journal of
Alzheimer’s Disease and Other Dementias, 35 (January), 53-65
doi:10.1177/1533317520915365

161
© University of Pretoria
Ruiz, A. (2015). The memory book as an augmentative tool to increase utterances in
Alzheimer and typical elderlies. Master's thesis, Universidad del Turabo, Gurabo: Puerto
Rico.
Sabat, S. (2019). How personhood is undermined: Commentary. In D. Brooker (Eds.),
Dementia Reconsidered: The Person Still Comes First (2nd ed., pp. 59-64). London,
England: Open University Press.
Sabat, S. R., & Harré, R. (1992). The construction and deconstruction of self in alzheimer’s
disease. Ageing and Society, 12, 443-461.
doi: 10.1017/S0144686X00005262
Sabat, S. R., & Lee, J. M. (2012). Relatedness among people diagnosed with dementia:
Social cognition and the possibility of friendship. Dementia, 11(3), 315-327.
doi:10.1177/1471301211421069
Saccasan, N., & Scerri, C. (2020). Dementia knowledge, attitudes and training needs of
speech–language pathology students and practitioners: A countrywide study.
International Journal of Language and Communication Disorders, 55(6), 955-970.
doi:10.1111/1460-6984.12574
Sackett, D. L., Rosenberg, W. M. C., Gray, M. J., Haynes, B. R., & Richardson, S. W.
(1996). Evidence-based medicine: What is it and what it isn’t. British Medical Journal,
312(7023), 71-72.doi:10.1115/1.2899246
Sahu, P. (2020). Closure of universities due to coronavirus cisease 2019 (COVID-19): Impact
on education and mental health of students and academic staff. Cureus, 2019(4), 4-9.
doi:10.7759/cureus.7541
Samuelsson, C., & Hydén, L-C. (2017). Collaboration, trouble and repair in multiparty
interactions involving couples with dementia or aphasia. International Journal of
Speech-Language Pathology, 19(5), 454-464. doi:10.1080/17549507.2016.1221448
Samuelsson, C., & Ekström, A. (2019). Digital communication support in interaction
involving people with dementia. Logopedics Phoniatrics Vocology, 44(1), 41-50.
doi:10.1080/14015439.2019.1554856
Savundranayagam, M. Y. (2012). Person-centered care: Measurement, implementation, and
outcomes. Clinical Gerontologist, 35(5), 357-359. doi:10.1080/07317115.2012.702646
Savundranayagam, M. Y. (2013). Missed opportunities for person-centered communication:
Implications for staff-resident interactions in long-term care. International
Pyschogeriatrics, 26(4), 645-655. doi: 10.1017/S1041610213002093
Savundranayagam, M.Y., & Moore-Nielsen, K. (2015). Language-based communication

162
© University of Pretoria
 strategies that support person-cantered communication with persons with dementia.
International Psychogeriatrics, 27(10), 1707-1718. doi: 10.1017/S1041610215000903
Savundranayagam, M. Y., Sibalija, J., & Scotchmer, E. (2016). Resident reactions to person-
centered communication by long-term care staff. American Journal of Alzheimers
Disease and Other Dementias, 31(6), 530-537. doi:10.1177/1533317515622291
Savundranayagam, M. Y., Basque, S. R., & Johnson, K. (2020). Feasibility of a dementia-
focused person-centered communication intervention for home care workers. Clinical
Gerontologist, 43(2), 181-192. doi;10.1080/07317115.2019.1694116
Schlosser, R. W. (1999). Social validation of interventions in augmentative and alternative
communication. Augmentative and Alternative Communication, 15(4), 234-247.
doi:10.1080/07434619912331278775
Schlosser, R. W. (2003). Determining the treatment integrity of AAC interventions. In R. W.
Schlosser (Ed.), The Efficacy of Augmentative and Alternative Communication:
Towards Evidence-Based Practice (pp. 181-202). San Diego, California: Elsevier
Science.
Schlosser, R. W. (2006). Evidence-based practice for AAC practitioners. Perspectives on
Augmentative and Alternative Communication, 15(3), 8-9. doi:10.1044/aac15.3.8
Schlosser, R. W., & Raghavendra, P. (2004). Evidence-based practice in augmentative and
alternative communication. Augmentative and Alternative Communication, 20(1), 1-21.
doi:10.1080/07434610310001621083
Schlosser, R. W., & Sigafoos, J. (2009). Navigating evidence-based information sources in
augmentative and alternative communication. Augmentative and Alternative
Communication, 25(4), 225-235. doi:0.3109/07434610903360649
Schlosser, R., & Sigafoos, J. (2008). Identifying “evidence-based practice” versus
“empirically supported treatment.” Evidence-Based Communication Assessment and
Intervention, 2(2), 61-62. doi:10.1080/17489530802308924
Sinha, I. P., Smyth, R. L., & Williamson, P. R. (2011). Using the Delphi technique to
determine which outcomes to measure in clinical trials: Recommendations for the future
based on a systematic review of existing studies. PLoS Medicine, 8(1), 1-5.
doi:10.1371/journal.pmed.1000393
Slaughter, S., Cole, D., Jennings., & Reimer, M. A. (2007). Consent and assent to participate
in research people with dementia. Nursing Ethics, 14(1), 27-40.
doi: 10.1177/0969733007071355
Skela-Savič, B., Gotlib, J., Panczyk, M., Patelarou, A. E., Bole, U., Ramos-Morcillo, A. J.,…

163
© University of Pretoria
 Ruzafa-Martínez, M. (2020). Teaching evidence-based practice (EBP) in nursing
curricula in six European countries: A descriptive study. Nurse Education Today, 94
(August), 104561. doi:10.1016/j.nedt.2020.104561
Skinner, S., Bonnet, K., Schlundt, D., & Karlekar, M. (2019). Life story themes : A
qualitative analysis of recordings rrom patients approaching the end of life. American
Journal of Hospice & Palliative Medicine, 36(9), 753-759.
doi:10.1177/1049909119851717
Smith, M. (2016). Evidence for impact and impact of evidence. Augmentative and
Alternative Communication, 32(4), 2270232. doi:10.1080/07434618.2016.1250283
Smith, B. C., & D'Amico. (2019). Sensory-based interventions for adults with dementia and
Alzheimer's discease: A scoping review. Occupational Therapy in Health Care, 34(3),
1102-1106. doi:10.1080/07380577.2019.1608488
Smith, M. M., & Murray, J. (2011). Parachute without a ripcord: The skydive of
communication interaction. Augmentative and Alternative Communication, 27(4), 292-
303. doi.10.3109/07434618.2011.630022
Smith, K., Ostinelli, E., & Cipriani, A. (2020). Covid-19 and mental health: A
transformational opportunity to apply an evidence-based approach to clinical practice
and research. Evidence-Based Mental Health, 23(2), 45-46. doi:10.1136/ebmental-2020-
300155
Smits, D. W., Van Meeteren, K., Klem, M., Alsem, M., & Ketelaar, M. (2020). Designing a
tool to support patient and public involvement in research projects: The involvement
matrix. Research Involvement and Engagement, 6(1), 1-7. doi:10.1186/s40900-020-
00188-4
Spilkin, M.-L., & Bethlehem, D. (2003). A conversation analysis approach to facilitating
communication with memory books. Advances in Speech Language Pathology, 5(2),
105-118. doi:10.1080/14417040510001669131
Spriggs, A. D., Lane, J.D, & Gast, D. (2018).Visual representation of data. In J. Ledford, &
D. Gast (Eds.), Single Case Research Methodology. (pp. 157-179). London: Routledge.
Subramaniam, P., & Woods, B. (2016). Digital life storybooks for people with dementia
living in care homes: An evaluation. Clinical Interventions in Aging, 16(11), 1263-1276.
doi:0.2147/CIA.S111097
Swaffer, K. (2015). Dementia and prescribed disengagement. Dementia: The International
Journal of Social Research and Practice, 14(1), 3-6. doi:10.1177/1471301214548136
Swaffer, K. (2019). Afterword. In D. Brooker (Eds.), Dementia Reconsidered: The Person

164
© University of Pretoria
 Still Comes First (2nd ed., pp. 178-180). London, England: Open University Press.
Swan, K., Hopper, M., Wenke, R., Jackson, C., Till, T., & Conway, E. (2018). Speech-
language pathologist interventions for communication in moderate–severe dementia: A
systematic review. American Journal of Speech-Language Pathology, 27(2), 836-852.
doi:10.1044/2017_ajslp-17-0043
Swarbrick, C. (2015). The quest for a new methodology for dementia care research.
Dementia, 14(6), 713-715. doi:10.1177/1471301215609247
Takayanagi, K., Kirita, T., & Shibata, T. (2014). Comparison of verbal and emotional
responses of elderly people with mild/moderate dementia and those with severe
dementia in responses to seal robot, PARO. Frontiers in Aging Neuroscience, 6, 1-5.
doi:10.3389/fnagi.2014.00257
Tawney, J.W., & Gast, D.L. (1984). Single subject research in special education.
Columbus: Charles. E. Merrill.
Taylor, S., & Balandin, S. (2020). The ethics of inclusion in AAC research of participants
with complex communication needs. Scandinavian Journal of Disability Research,
22(1), 108-115. doi:10.16993/sjdr.637
Trahan, M. A., Donaldson, J. M., McNabney, M. K., & Kahng, S. (2014). Training and
maintenance of a picture-based communication response in older adults with dementia.
Journal of Applied Behavior Analysis, 47(2), 404-409. doi:10.1002/jaba.111
Terkelsen, A. S., Petersen, J. V., & Kristensen, H. K. (2020). Mapping empirical experiences
of Tom Kitwood’s framework of person-centred care for persons with dementia in
institutional settings. A scoping review. Scandinavian Journal of Caring Sciences,
34(1), 6-22. doi:10.1111/scs.12709
Teri, L., Logson., McCurry, S., Pike., K., & McGough, E. (2020). Translating an evidence-
based multicomponent intervention for older adults with dementia and caregivers. The
Gerontoogist, 60(3), 548-557. doi:10.1093/geront/gny122
United Nations. (2019). World Population Ageing 2019: Highlights. Retrieved from:
https://www.un.org/en/development/desa/population/publications/pdf/ageing/WorldPop
ulationAgeing2019-Highlights.pdf
Valenta, A. L., & Wigger, U. (1997). Q-methodology: Definition and application in health
care informatics. Journal of the American Medical Informatics Association, 4(6), 501-
510. doi:0.1136/jamia.1997.0040501
Van Der Roest, H. G., Meiland, F. J. M., Comijs, H. C., Derksen, E., Jansen, A. P. D., Van
Hout, H. P. J., Jonker, C., & Dröes, R. M. (2009). What do community-dwelling people

165
© University of Pretoria
 with dementia need? A survey of those who are known to care and welfare services.
International Psychogeriatrics, 21(5), 949-965. doi:10.1017/S1041610209990147
Vasse, E., Vernooij-Dassen, M., Spijker, A., Rikkert, M. O., & Koopmans, R. (2010). A
systematic review of communication strategies for people with dementia in residential
and nursing homes. International Psychogeriatrics, 22(2), 189-200.
doi:10.1017/s1041610209990615
Villar, F., Serrat, R., & Bravo-Segal, S. (2019). Giving them a voice: Challenges to narrative
agency in people with dementia. Geriatrics, 4(1), 20. doi:10.3390/geriatrics4010020
Veritas Health Innovation.(n.d.) Covidence Systematic Review Software (Melbourne, VIC:
Veritas Health Innovation). Retrieved from: https://www.covidence.org/home
Wallace, S. E., Dietz, A., Hux, K., & Weissling, K. (2012). Augmented input: The effect of
visuographic supports on the auditory comprehension of people with chronic aphasia.
Aphasiology, 26(2), 162-176. doi:10.1080/02687038.2011.628004
Wallace, S. E., Knollman-Porter, K., Brown, J. A., & Hux, K. (2018). Narrative
comprehension by people with aphasia given single versus combined modality
presentation. Aphasiology, 33(6), 731-734. doi:10.1080/02687038.2018.1506088
Wallace, S. J., Worrall, L., Rose, T., & Le Dorze, G. (2016). Core outcomes in aphasia
treatment research: An e-Delphi consensus study of international aphasia researchers.
American Journal of Speech-Language Pathology, 25 (October), 1-15.
doi:10.1044/2016
Waller, A. (2019). Telling tales: Unlocking the potential of AAC technologies. International
Journal of Language & Communication Disorders, 54(2), 159-169. doi:10.1111/1460-
6984.12449
Waller, A., Balandin, S. A., O’Mara, D. A., & Judson, A. D. (2005). Training AAC users in
user-centred design. In L. Gibson, P. Gregor, & D.Sloan (Eds.), Accessible Design '05:
Proceedings of the 2005 International Conference on Accessible Design in the Digitial
Worlds. Swinton, UK: British Computer Society. doi:10.14236/ewic/ad2005.2
Ward, R., Vass, A. A., Aggarwal, N., Garfield, C., & Cybyk, B. (2008). A different story:
Exploring patterns of communication in residential dementia care. Ageing and Society,
28(5), 629-651.doi:10.1017/S0144686X07006927
Watts, S., & Stenner, P. (2005). Doing Q methodology: Theory, method and interpretation.
Qualitative Research in Psychology, 2, 67-91. doi:10.1191/1478088705qp022oa
Watts, S., & Stenner, P. (2012). Doing Q methodological research: Theory, method, and
interpretation. Washington, DC: Sage.

166
© University of Pretoria
Webb, J. (2017). Conversation takes two: Understanding interactions with people with
dementia. Disability and Society, 32(7), 1102-1106.
doi:10.1080/09687599.2017.1321225
Western Cape Government. (2020). DSD welcomes donations and provides additional
support to old age homes. Retrieved from: https://www.westerncape.gov.za/news/dsd-
welcomes-donations-and-provides-additional-support-old-age-homes
Wilder, J. (2008). Video observations of dyadic interaction: Behaviour style of presymbolic
children. Scandinavian Journal of Disability Research, 10(2), 104-124.
doi:10.1080/15017410801900390
Williams, C. L., Newman, D., & Hammar, L. M. (2018). Preliminary study of a
communication intervention for family caregivers and spouses with dementia.
International Journal of Geriatric Psychiatry, 33(2), 343-349. doi: 10.1002/gps.4816
Wlliams, K. N., Perkhounkova, Y., Jao, Y.-L., Bossen, A., Hein, M., Chung, S., Starykowicz,
A., & Turk, M. (2017). Person-centered communication for nursing home residents with
dementia: Four communication analysis methods. Western Journal of Nursing Research,
40(7), 1012-1031. doi:10.1177/0193945917697226
Wong, S. B., Anand, R., Chapman, S. B., Rackley, A., & Zientz, J. (2009). When nouns and
verbs degrade: Facilitating communication in semantic dementia. Aphasiology, 23(2),
286-301. doi.10.1080/02687030801943112
Williams, M. B., Krezman, C., & McNaughton, D. (2008). “Reach for the stars”: Five
principles for the next 25 years of AAC. Augmentative and Alternative Communication,
24(3) 194-206. doi:10.1080/089902208 02387851
World Health Organisation. (2020). Dementia. Retrieved from: https://www.who.int/news-
room/fact-sheets/detail/dementia
World Health Organization. (2017). Dementia: Global action plan 2012-2025. [online].
Retrieved from
https://www.who.int/mental_health/neurology/dementia/action_plan_2017_2025/en/
World Health Organisation. (2015). WHO global strategy on people-centred and integrated
health services: Interim report. Retrieved from:
https://apps.who.int/iris/bitstream/handle/10665/155002/WHO_HIS_SDS_2015.6_en
g.pdf
Wynn, R., & Khayum, B. (2015). Developing personally relevant goals for people with
moderate to severe dementia. Seminars in Speech and Language, 36(3), 199-208.
doi:10.1055/s-0035-1551837

167
© University of Pretoria
Yin, R. K. (2011). Applications of case study research (3rd ed.). Thousand Oaks, CA: Sage.

Yang, Y. (2016). A brief introduction to Q Methodology. International Journal of Adult

Vocational Education and Technology, 7(2), 42-53. doi:10.4018/ijavet.2016040104
Yeandle, S., Kröger, T., & Cass, B. (2012). Voice and choice for users and carers?
Developments in patterns of care for older people in Australia, England and Finland.
Journal of European Social Policy, 22(4), 432-445. doi:10.1177/0958928712449775
Zabala, A., & Pascual, U. (2016). Bootstrapping Q methodology to improve the
understanding of human perspectives. PLoS ONE, 11(2), 1-19.
doi:10.1371/journal.pone.0148087
Zabala, A., Sandbrook, C., & Mukherjee, N. (2018). When and how to use Q methodology to
understand perspectives in conservation research. Conservation Biology, 32(5), 1185-
1194. doi:10.1111/cobi.13123
Zaga, C. J., Cigognini, B., Vogel, A. P., & Berney, S. (2020). Outcome measurement tools
for communication, voice and speech intelligibility in the ICU and their clinimetric
properties: A systematic review. Journal of the Intensive Care Society, 0(0), 1-14.
doi:10.1177/1751143720963757
Zapf, A., Castell, S., Morawietz, L., & Karch, A. (2016). Measuring inter-rater reliability for
nominal data - Which coefficients and confidence intervals are appropriate? BMC
Medical Research Methodology, 16(1), 1-10. doi:10.1186/s12874-016-0200-9

168
© University of Pretoria
 Appendices

169
© University of Pretoria
 Appendix A: Co-Production Process: Observation Form

Multiple Stakeholder Interview and Observations

Guiding Music Therapist Occupational Nursing Staff Familiar
Questions Therapist Conversational
Partner

What does Mrs

Brown enjoy
talking about?

What are Mrs

Browns’
personal
preferences,
things she
enjoys doing?

Are there
phrases Mrs
Brown says
often?

Is there anything
else that could
assist in getting
to know Mrs
Brown?
Observations:

170
© University of Pretoria
 Appendix B: Co-Production Process: Participant’s Information from the Family

Dear family,
With your help, I will write short sentences on *Mrs. Brown’s life story. This will serve as the basis for
our interpersonal interaction in the research study.
Would you kindly provide any (i) family photographs (other than those in her private bedroom)
depicting interesting and happy times in Mrs Brown’s life (incl., her travels, family and friends) and (ii)
information in the table below?
Thank you for your time.

Warm Regards,
Adele May

Note:
▪ All bullet points may not be applicable to Mrs *Brown, they only serve as a guide.
Childhood and family ▪ Where did she grow up? Mrs Brown’s career:
▪ Who were her parents? ▪ first job
▪ Her position in the family ▪ last job
(e.g., youngest daughter) Interesting work experiences
▪ A distinctive family trait
▪ Childhood interests
▪ She was really good at....
▪ She didn’t like....
▪ School stories…
▪ Special early memories
Achievements/things ▪ Special skills Hobbies/pastimes
she was good at ▪ Milestones ▪ Activities of interests (e.g.,
▪ Highlights in personal life or dancing, art, crafts, play
career sport, cooking, gardening)
▪ Who are her best friends,
people she
socialised/socialises with
most of the time?
Personal preferences ▪ Clothing preferences Religious/cultural/spiritual
▪ Fashion and style ▪ Community/associations
▪ Signature perfumes ▪ Cultural celebrations
▪ Nicknames/special names she ▪ Birthdays
likes to tease or calls others ▪ Spiritual beliefs
▪ Personal expressions, ▪ Values
sayings/phrases that’s
associated with her
▪ Favourite colours
▪ Favourite recipes/dishes
▪ Favourite TV shows, actors,
movies
▪ Celebrities
▪ Music genre’s
*Mrs Brown is a pseudonym used for the case study participant

171
© University of Pretoria
 Appendix C: Life Story Framework and Themes

Participant’s life story centred around the

high points and milestones related to three themes:
LIFE EVENTS
Description
Upbringing Places where you lived, growing up (with your family) preparing meals,
helping others, shopping.
Jobs, occupation, career Previous employment, work or educational experiences.
Education Personal grooming and taking care of your health.
Travel Places you have travelled, went on vacation, places of interests
Special events
Hobbies and pastimes Hobbies or fun activities done during in spare time or for relaxation e.g.,
cooking
PEOPLE
Description
Family life Relationships with a partner or spouse, family, special friends,
Immediate family neighbours, carers.
Extended family
Friends
Community

VALUES
Description
Religion and spirituality Religious beliefs, going to specific religious activities.
Affirm identity through the participants core values

Adapted from Skinner, Bonnet, Schlundt, and Karlekar (2019)

172
© University of Pretoria
 Appendix D: Life Story Vignette and Example Script

Life story 1: Family Meals

1. I am Anna Brown and I am Jewish
2. Every Friday night we celebrate Shabbat
3. We always have a lot of food at Shabbat dinner
4. Lala does all the cooking in the home
5. I am not good at cooking or baking
6. Rosh Hoshana is my favourite time of year
7. We dip apple slices in honey to celebrate the Jewish new year

• Life story screen shots from the AAC device are on the next page (Appendix E)
Intervention Phase (semi-structured script)
Cover Page Mrs Brown, these are your stories. You helped me write them.
Page 1 Researcher: Aided modelling: point to symbols on the AAC device:
You have two stories to choose from.
Would you like to talk about family meals, OR
Would you like to talk about music and theatre?
Page 2 I am Anna Brown, and I am Jewish.
Page 3 Mrs Brown, this is a picture of a beautiful synagogue. I am sure this is a special
place for you?
Page 4 Multimodal responses (option for the participant to point to pictures if she prefers)
Page 5 Aided modelling: point to symbols on the AAC device:
Yes, Mrs Brown, you are Jewish. The synagogue is a special place
Page 6 Every Friday night we celebrate shabbat.
Page 7 Mrs Brown, this is a picture of your family. Nelly looks so happy to be at Shabbat
dinner.
Page 8 Multimodal responses (option for the participant to point to pictures if she prefers)
Page 9 Aided modelling: point to symbols on the AAc device
Your family, and Nelly are happy to be at Shabbat dinner.
Page 10 We always have a lot of food at shabbat dinner.
Page 11 Mrs Brown, in this is picture, there is bread, soup, and salad at Shabbat dinner.
The soup looks delicious.
Page 12 Multimodal responses (option for the participant to point to pictures if she prefers)
Page 13 Researcher: Aided modelling: point to symbols on the AAC device
At Shabbat dinner, the soup and salad were delicious.
Page 14 I am not good at cooking or baking.
Page 15 Lala does all the cooking in the home.
Researcher: Aided modelling: point to symbols on the AAC device
Mrs, Brown, so Lala does all the cooking at home?
Page 16 Multimodal responses (option for the participant to point to pictures if she prefers)
Researcher: Aided modelling: point to symbols on the AAC device
Page 17 Rosh Hoshana is my favourite time of year.
Page 18 We dip apple slices in honey to celebrate the Jewish new year.
Page 19 Mrs, Brown the apples slices and honey must be a sweet treat!
Page 20 Multimodal responses (option for the participant to point to pictures if she prefers)
Page 21 Researcher: Aided modelling: point to symbols on the AAC device
Yes, Mrs Brown, you dip apple slices in honey to celebrate the Jewish new year,
and you love it!

173
© University of Pretoria
 Appendix E: Electronic Life Story Conversational Support on AAC device (Example)

All names are pseudonyms. All personal photographs are blanked out.

174
© University of Pretoria
 Family photograph

Family Nelly’s
photograph photograph

Nelly’s
Family
photograph
photograph

175
© University of Pretoria
 176
© University of Pretoria
 177
© University of Pretoria
Images: Google Images and researchers own.

178
© University of Pretoria
 Appendix F1: AAC Outcome Measure for Persons with Dementia (AACOM-PwD)

Augmentative and Alternative Communication Outcome Measure for A Person with Dementia
(AACOM-PwD)

CODING SHEET (example)

Name of coder: Session coded: Date:

Instruction:
1) Watch the 10-minute video-recording to score turn taking responses of the participant only.
2) Turn taking is scored within the boundary of an interaction opportunity (labelled as interaction on the
scoring sheet, e.g., interaction 1). Each participant has a minimum of 5 interaction opportunities for each
10-minute interaction.
3) Begin by familiarising yourself with the coding definitions and coding colours in the coding guide.
4) Stop scoring when the ten-minutes is completed

Turn taking Domains

Interaction 1 Verbal (V) Gestural (G) Aided (A) Multimodal (MM) Turns
Turn taking Sub-domains
a b a b c a b a b c d

Offer Choice ✓
PCC+wait ✓
Model choice ✓ ✓

Subtotal 2 1 1 4
Subtotal 3 1

Interaction 2 Verbal (V) Gestural (G) Aided (A) Multimodal (MM) Turns
a b a b c a b a b c d

Offer Choice ✓
PCC+wait ✓
Model choice ✓ ✓
4
Subtotal 1 1 1 1
Subtotal 2 1 1

Interaction 3 Verbal (V) Gestural (G) Aided (A) Multimodal Turns

(MM)
a b a b c a b a b c d

Offer Choice ✓ ✓ ✓
PCC+wait ✓
Model choice ✓

Subtotal 1 2 1 1 5
Subtotal 3 1 1

179
© University of Pretoria
Interaction 4 Verbal (V) Gestural (G) Aided (A) Multimodal (MM) Turns
a b a b c a b a b c d

Offer Choice ✓ ✓
PCC+wait ✓
✓
Model choice ✓
5
Subtotal 1 1 2 1
Subtotal 1 4

SCORE SHEET: TALLY THE ABOVE SUB-TOTAL SCORES ON THE SHEET BELOW

Total Turns Turn taking Domains Turn taking Sub-domain

count (horizontally) all turns *Total: count all sub-totals Total: count all sub-totals
in each interaction (pink (vertically) in each interaction (vertically) in each interaction 1-8 in
block) 1-8 in their respective colour their respective colour
Verbal Turns (a)Vocalisations
14 4
9 (b) Verbalisation
5
Gestural Turns (a) Body movement

1 (b) Head nod 1

(c) Facial expression

Aided Turns (a)With assistance

1
1
(b)Without assistance

Multi-modal Turns (a)Verbal +Gestural 3

7
(b) Verbal + Aided

(c) Gestural + Aided

(d)Verbal + Aided+
Gestural
*note: an off -topic turn is counted as a turn

180
© University of Pretoria
 Appendix F2: AACOM-PwD Operational Definitions

AACOM-PwD: TURN TAKING OPERATIONAL DEFINITIONS

DOMAIN SUB-DOMAIN
Score: 1 Point for each occurrence within a turn opportunity
Verbal (V) Turn This score is given when a participant uses a:
A verbal turn is a (a) communicative vocalisation
vocalisation (sound) or i.e., vocalisation/sound (e.g., ooo), or a verbal agreement (e.g., uh-huh,
verbalised (utterance) Hmm).
response or contribution (b) verbal utterance:
to an interaction i.e., single-word (e.g., “right” /pause/) or multi-word utterance (“let me
see”../pause/).
* One score is given if there is a verbal repetition of the same utterance (“me
too-me too”).
Gestural (G) Turn A gestural turn is a response to an interaction initiated by the researcher, or (b) after
A gestural turn indicates the 5 sec expectant delay, (c) or 5 seconds after the final modelling by researcher.
that the participant used a This score is given when a participant uses the following gestures:
body movement, head- (a) Body movement (e.g., handwave, shoulder shrug) (e.g., moves shoulders to
nod or part of the body as indicate “perhaps”)
a turn taking response (b) Headnod (e.g., head moving side-to side to indicate “no”)
(c) Facial movements or eyebrow raising (e.g., to indicate shock)
Aided (A) Turn This score is given when participant is:
An aided turn indicates (a) Physically touching or pointing to AAC symbols or using the iPad with
that the participant used assistance within a turn
the AAC device or AAC (b) Physically touching, pointing or using the iPad without assistance within a
symbol selection during a turn.
turn
Multi-Modal (MM) Dual turn:
Turn (a) Verbal + Gestural
A multimodal turn i.e., vocalisation or verbal utterance [single or multi-word] + gesture] (e.g.
indicates that the saying “yes + moving head up and down)
participant used a (b) Verbal + Aided
combination of any two i.e., vocalisation or verbal utterance [single or multi-word] + pointing
turn taking domains touching, body orientated to the AAC device.
during a turn *The pointing/touching body orientation can be done simultaneously with
vocalisation or verbal utterance
(c) Gestural + Aided
i.e., gesture (before, during or followed after) + physically touching,
pointing, body orientated to AAC device or AAC symbols.
*The pointing/touching body orientation can be done simultaneously with
the gesture, OR before OR after the gesture.
Triadic turn:
(d)Verbal + Aided+ Gestural
vocalisation/verbal utterance +gesture + pointing/touching/body orientated
to the AAC device (e.g., “me too-me too+ raise eyebrow+ thereafter
looking at the AAC device)
*The pointing/touching body orientation can be done simultaneously with
vocalisation/verbal utterance or gesture, OR before OR after vocalisation,
verbal utterance or gesture.
* Aided includes AAC symbol selection using the legacy message (voice-output)
with or without assistance from the researcher.
Note: An off-topic turn: is scored as a turn: i.e., the participant asking an unrelated question or making an
inappropriate comment that is not related to the life story content (e.g., says “ I lost my keys” in response to
“which story would you like to talk about”)

181
© University of Pretoria
Appendix G1: Ethics Approval (original study)

182
© University of Pretoria
Appendix G2: Ethics Approval (amended study)

183
© University of Pretoria
Appendix H: Concourse Development: Letter of Information to Participate as an
Expert on an e-Delphi Panel

184
© University of Pretoria
 185
© University of Pretoria
 186
© University of Pretoria
Appendix I: Concourse Development: Online Questionnaire for e-Delphi Panel

187
© University of Pretoria
 188
© University of Pretoria
 Appendix J: Concourse Development: Content Analysis

Themes Code Category Codes* Instances

Theme 1: Personhood Personhood and dignity 8
PwD Characteristics principle
The personalised Uniqueness of the Personal biography 14
characteristics of the PwD PwD Personal preferences 26
must be considered when Personal words 2
designing AAC interventions
Personal goals 3
for PwD: including personal
preferences, use of life Dementia Neuro-cognitive profile 8
history, adapting to changes in diagnosis Dementia-changes 13
dementia severity, consistency Signals of distress/comfort 3
and tailored around Adapted to Strengths-based 9
individualised strengths. abilities Individualised/tailored 16
Consistency 3
Theme 2: Relational (with PwD) 4
Working with the PwD Collaboration Collaborative (involving others) 9
Different ways of working Key relationships (familiar 11
with the PwD and those close communication partners)
to them, including supporting Supporting Decision-making 6
communication partners’ independence Inclusion 9
(CPs) communication goals
and needs. Self-determination 5
Theme 3: Unaided supports Gestures, eye-contact etc 11
Use of Communication Unaided (other Proxemics 5
supports (AAC) features) Touch 3
Related to the various
Prosodic 1
features, and attributes of
communication supports for Non-electronic Visual/pictorial support 23
PwD supports Paper-based books/boards 8
Objects, remnants, keywords, topic 11
setters
Electronic Technology and apps 4
supports
Combined Combination of supports 7
supports
Communication Level of complexity, knowledge of 7
support features use, availability
Specific strategies Augmented input 1
Visual scene 1
Language and Language based support 14
memory support Memory support 10
Cues and prompts 3
Meaningful Specific activities 9
content Music and signing 5

189
© University of Pretoria
Themes Code Category Codes Instances
Theme 4: Co-construction of meaning 9
Empathetic Communication Co-construction CP co-construction characteristics 20
A meaning-making process in (enabling)
which the CP’s uses empathy Attunement 7
within an AAC interaction to Attentiveness (of CP or Dyad) 6
establish, maintain and co- Emotional resonance 1
construct meaning with the Empathy 2
PwD Observant 3
Being Present 2
Interest shown by CP in 3
Listening skills 2
CP co-construction strategies 7
Questions 7
Theme 5 Interaction structure Initiate interaction 2
Temporal-sequential Maintain interaction 3
interaction features Temporal aspects Pauses and delays 3
Relates to the stops, silences, Pacing and time of the interaction 12
pauses and structure of an
Silences 1
interaction that must be
considered when interacting
with a PwD
Theme 6 Interaction context Context (immediate/situated) 14
Interactional context Context (comm support in context) 7
The association between the Purposes of an Interaction of needs and wants, 4
context, PwD and AAC- interaction etiquette and information transfer
communication support, Interaction for social closeness 7
including the changes in the
immediate context and
purpose of the interaction
Theme 7 External Interaction environment 7
Interactional environment environment
Features of the external
environment which hinders or
supports an AAC interaction
with a PwD
Theme 8 Constructs and Engagement 12
Features of an AAC overarching Participation 12
interaction outcome frameworks to be Quality of Life and wellbeing 5
measure for PwD measured ICF/other 3
Relates to all features Types of outcome Spontaneous interaction measure 1
necessary for developing an measures Performance-based measures 2
AAC interaction outcome
measure for PwD, including
the constructs that should be
measured, domains to be
measured and types of
outcome

190
© University of Pretoria
 Theme 8 (continued) Domains to be measured Turn taking 21
Timing 3
Verbal and vocalisations 1
Language 18
Multimodal communication 1
Body Language 2
Gestural domain 2
Silences 1
CP Domain 5
Social closeness 3
Focus and joint attention 2
Behavioural and emotional domain 11
Other domains 1
Analysis of Interaction Video-recorded analysis 2
Interaction analysis types 1
Measure interaction in Use of comm supports in context 15
various contexts
Theme 9 Intrapersonal outcomes Satisfaction of the interaction 8
Social validity of AAC Feelings about the interaction 2
interaction outcomes in PwD
Assessing the importance and Confidence in the interaction 2
relevance of interaction Success of interaction 5
outcomes from various Control and choice 3
stakeholders
Interpersonal outcomes Fun/Enjoyment 1
Stakeholder’s Perspective Proxy and PWD reported 5
Proxy-reported perspective 2
Theme 10 Inhibiting CP Patronisation and disruptions 4
Inhibitors to an AAC characteristics
interaction with a PwD
Specific CP characteristics that
undermine personhood within
an AAC interaction
Theme 11 Person-centred care does PCC does not work 1
Diversity of perspectives on not work
approaches to AAC
interventions with PwD
Opposing expert opinions on
usefulness or applicability of
specific approaches in AAC
interventions with PwD

191
© University of Pretoria
 Appendix K1: Q-set: Statements (pre-pilot)

1. Electronic AAC systems (e.g., mobile technology, tablet computers with applications [apps]) can
be useful to support interpersonal interaction between a conversational partner and person with
dementia.
2. Training persons with dementia on the use of electronic AAC systems (e.g., navigating on tablet
computers with apps) before they use them in real-time interaction settings is not necessary when
the purpose is interpersonal interaction.
3. The use of intelligent assistive technologies (e.g., using artificial intelligence) to support
communication and interpersonal interaction in persons with dementia should not be explored
for persons with dementia.

4. Combined AAC systems (e.g., clarifications with head nods, verbal comments by pointing to
photographs) are most useful communication supports for persons with dementia
5. Combining picture pointing with spontaneous speech (i.e., augmented input) is a useful strategy
to support auditory comprehension of spoken language in persons with dementia.
6. Meaningful high-context photographs (e.g., visual scene displays) support joint reference during
interpersonal interaction with a person with dementia.

7. A conversational partner should never correct off-topic utterances made by a person with
dementia.
8. A conversational partner should assume that all communication attempts made by a person with
dementia is meaningful.
9. A conversational partner can co-construct meaning with a person with dementia by using
language that enhances social emotional connection.
10. A conversational partner should display empathy by mirroring the emotion of the person with
dementia to facilitate interpersonal interaction with a person with dementia.
11. Active listening shown by paying attention conveys respect for the contributions of a person with
dementia.
12. Silences within an interaction promotes moments of companionship with a person with
dementia.
13. A conversational partner can support interpersonal interaction by pausing to give a person with
dementia time to respond.

14. A person with dementia must be supported in decision-making in all aspects of their AAC
intervention.
15. A person with dementia must be consulted about their personal preferences for interpersonal
interaction.
16. AAC strategies for a person with dementia are flexible depending on the stage and type of
dementia.
17. AAC interventions to support interpersonal interaction in persons with dementia must be tailored
to the strengths of the person with dementia.

18. The self-determination of a person with dementia is promoted by facilitating their involvement in
co-designing their own AAC interventions.
19. Adapted participatory methods (e.g., co-production) should be employed to include persons with
dementia as collaborators in the co-design of AAC interventions.
20. Working collaboratively with the family of a person with dementia and other stakeholders
throughout the AAC intervention process is important, but not necessary.
21. The immediate interaction context plays a vital role in the success of interpersonal interaction
with a person with dementia (e.g., time of day in which interaction occurs).

192
© University of Pretoria
22. Conversational supports with personalised life-story content of a person with dementia is a
useful scaffold for reminiscence-based conversations with a person with dementia.
23. Persons with dementia must be supported in authoring the content of their life-story based
conversational supports.
24. Life-story based conversational supports that include personalized music and digital photographs
on electronic devices (e.g., iPad) enhances the enjoyment of interpersonal interaction for persons
with dementia.

25. A patronising style of language (e.g., an exaggerated tone of voice typical of talking to an infant)
used during interpersonal interaction with a person with dementia undermines personhood of a
person with dementia.
26. Person-centered care as a philosophy of care (e.g., one that promotes the independence of
persons with dementia and values their life history) in AAC interventions is not clinically
achievable.

27. Satisfaction of an interaction should be rated directly from the perspective of a person with
dementia.
28. Intra-personal outcomes such as a sense of connection experienced within an interpersonal
interaction must be measured by asking the person with dementia.
29. Successful interaction should be rated from the perspectives of both the person with dementia
(patient-reported) and conversational partner (proxy-reported).

30. Social connections are important goals of AAC interventions designed to support interpersonal
interaction with a person with dementia
31. Enjoyment and motivation (as engagement outcomes) within an interpersonal interaction is
equally important to include in an outcome measure for persons with dementia.

32. AAC-supported interaction in persons with dementia is best analysed qualitatively from
observation of spontaneous interaction
33. AAC-supported interaction in persons with dementia must be analysed using a combination of
qualitative measures and standardised, quantitative outcome measures.

34. The most important domain to measure in an AAC interaction outcome measure for persons with
dementia is turn taking (e.g., frequency of turn taking exchanges).
35. An AAC interaction outcome measure must include the modality of turn taking (e.g., turn taking
with gestures).
36. Participation outcomes (frequency of taking part in everyday interactions) are important to
include in an AAC interaction outcome measure for persons with dementia.
37. It is important to include the behaviour displayed by persons with dementia (e.g., their level of
frustration, agitation or avoidance) as an AAC interaction outcome measure for persons with
dementia.
*blue font denotes statements recommended to be adjusted

193
© University of Pretoria
 Appendix K2: Q-set: Statements (post-pilot)

AAC scaffolding strategies to support interpersonal interaction in person with dementia

Electronic AAC systems
1. Electronic AAC systems (e.g., tablet computers with applications [apps]) can be useful to support
interpersonal interaction between a conversational partner and a person with moderate-severe
dementia.
2. Training persons with dementia on the use of electronic AAC systems (e.g., navigating on tablet
computers with apps) before they use them in real-time interaction settings is not necessary when the
purpose is interpersonal interaction.
3. The use of intelligent assistive technologies (e.g., using artificial intelligence) to support
communication and interpersonal interaction in persons with dementia should not be explored for
persons with dementia.
4. Legacy messages (i.e., unique words or phrases used by a person with dementia digitally recorded in
their own voice in the early stages of dementia) stored as voice output on electronic devices (e.g.,
iPad) are useful to facilitate social connection with a conversational partner as dementia severity
progresses.
AAC techniques and strategies
5. Non-electronic AAC systems (e.g., paper-based communication books with pictures) are the most
beneficial communication supports for interpersonal interaction with a person with mild-moderate
dementia
6. Combining picture pointing with spontaneous speech (i.e., augmented input) is a useful strategy to
support auditory comprehension of spoken language in persons with dementia.
7. Meaningful high-context photographs (e.g., visual scene displays) support joint reference during
interpersonal interaction with a person with dementia.
“Empathetic Communication” strategies
8. A conversational partner should never correct off-topic utterances made by a person with dementia.
9. The unique set of communication behaviours of a person with severe dementia (e.g., vocalisations,
facial expressions, gestures) should be used as the basis for interpersonal interaction with a
conversational partner.
10. A conversational partner can co-construct meaning with a person with dementia by using language
that enhances social-emotional connection.
11. A conversational partner should display empathy by mirroring the emotion of the person with
dementia to facilitate interpersonal interaction with a person with dementia.
12. Active listening shown by paying attention conveys respect for the contributions of a person with
dementia.
13. By repeating the utterances of the person with dementia, a conversational partner shows an
understanding of what the person with dementia is expressing within an interpersonal interaction.
Silences and expectant delays
14. A conversational partner can support interpersonal interaction by pausing to give a person with
dementia time to respond.
Person-Centred Care elements in AAC interventions that support interpersonal interaction in
persons with dementia
Personalisation and independence
15. A person with dementia must be supported in decision-making in all aspects of their AAC
intervention.
16. A person with dementia must be consulted about their personal preferences for interpersonal
interaction.
17. AAC strategies for a person with dementia are flexible depending on the stage and type of dementia.
18. AAC interventions to support interpersonal interaction in persons with dementia must be tailored to
the strengths of the person with dementia.
Person-centered methods and ways of working
19. The self-determination of a person with dementia is promoted by facilitating their involvement in co-
designing their own AAC interventions.

194
© University of Pretoria
20. Adapted participatory methods (e.g., co-production) should be employed to include persons with
dementia as collaborators in the co-design of AAC interventions.
21. Working collaboratively with the family of a person with dementia and other stakeholders throughout
the AAC intervention process is important, but not necessary.
22. The immediate interaction context (e.g., the time of day in which an interaction occurs) does not play
a vital role in the success of interpersonal interaction with a person with dementia.
Life story
23. Conversational supports with personalised life-story content of a person with dementia is a useful
scaffold for reminiscence-based conversations with a person with dementia.
24. Persons with dementia should be supported in authoring the content of their life-story based
conversational supports.
25. Life-story based conversational supports that include personalized music and digital photographs on
electronic devices (e.g., iPad) enhances the enjoyment of interpersonal interaction for persons with
dementia.
Person-centred care as a practice/other
26. A patronising style of language (e.g., an exaggerated tone of voice typical of talking to an infant)
used during interpersonal interaction with a person with dementia undermines personhood of a person
with dementia.
27. Person-centered care as a philosophy of care (e.g., one that promotes the independence of persons
with dementia and values their life history) in AAC interventions is not clinically achievable.
Outcome measurement for AAC-supported interpersonal interaction with persons with dementia
Intrapersonal outcomes
28. Successful interaction should be rated from the perspectives of both the person with dementia
(patient-reported) and conversational partner (proxy-reported).
29. It is not necessary to measure the satisfaction of the person with dementia in using AAC strategies
within an interpersonal interaction.
Interpersonal outcomes
30. Social connections are important goals of AAC interventions designed to support interpersonal
interaction with a person with dementia.
31. Enjoyment and motivation (as engagement outcomes) within an interpersonal interaction are not
important to include in an outcome measure for persons with dementia.
Interaction analysis
32. AAC-supported interaction in persons with dementia is best analysed qualitatively from observation
of spontaneous interaction.
33. AAC-supported interaction in persons with dementia must be analysed using a combination of
qualitative measures and standardised, quantitative outcome measures.
AAC outcome measure domains
34. The most important domain to measure in an AAC interaction outcome measure for persons with
dementia is turn taking (e.g., frequency of turn taking exchanges).
35. An AAC interaction outcome measure must include the modality of turn taking (e.g., turn taking with
gestures).
36. It is not necessary to include participation outcomes (frequency of taking part in everyday
interactions) in an AAC interaction outcome measure for persons with dementia.
37. It is not important to include the behaviour displayed by persons with dementia (e.g., their level of
frustration, agitation or avoidance) as an AAC interaction outcome measure for persons with
dementia.

195
© University of Pretoria
Appendix L: Q: Sort: Letter of Information to Experts

196
© University of Pretoria
 197
© University of Pretoria
 Appendix M: Online Q-sort

Online instructions on QMethod Software (Lutfallah, & Buchanan, 2019)

198
© University of Pretoria
 199
© University of Pretoria
Appendix N: Permission Letter to Dementia Care Home

200
© University of Pretoria
 201
© University of Pretoria
 202
© University of Pretoria
 203
© University of Pretoria
Appendix O: Recruitment Flyer (for persons with dementia)

204
© University of Pretoria
Appendix P: Case study: Letter of Consent to Legal Guardian of the Participant with
Dementia

205
© University of Pretoria
 206
© University of Pretoria
 207
© University of Pretoria
 208
© University of Pretoria
Appendix Q: Case study: Letter of Consent to a Familiar Conversational Partner

209
© University of Pretoria
 210
© University of Pretoria
Appendix R: Case study: Letter of Consent to Participant with Dementia (adapted)

211
© University of Pretoria
 212
© University of Pretoria
 213
© University of Pretoria
Informed Consent Reply Slip (Adapted)

214
© University of Pretoria
 Appendix S: Observer to Informed Consent Form

Observer to informed consent

Office use: participant nr

Name of observer:

Relationship to participant:
___________________________________________________________________________

I have observed the research project being explained to: (insert name of person with
dementia)
___________________________________________________________________________

Please tick  the options that apply:

Yes No
I feel that he/she understood what was required of him/her for
participation

I feel that he/she was happy to participate in the study without being
coerced by the researcher.

Additional comments on the informed consent observed.

___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________

Signature of Observer: _____________________________

Date: __________________________________________

Place: __________________________________________

Researcher: Adele May

Signature: ___________________________________

Date: ______________________________________

215
© University of Pretoria
 Appendix T: On-going Informed Consent Form and Picture Booklet

Process of on-going consent form (participant with dementia)

Office use: participant nr

Date: _______________________________

Research site: _________________________

I know what the research

study is about

I know that I can take a

break if I feel tired.

I know I can stop at any I want to

time if I don’t want to be in
the study any longer.

I am happy to continue with

the research study today
with Adele May.

Participants name: __________________________________

Participants signature: _______________________________
Adele May (researcher) Signature: ____________________
Date: ____________________________________________
Place: ____________________________________________

216
© University of Pretoria
Process of on-going consent: picture booklet

217
© University of Pretoria
 Appendix U: Biographical Questionnaire: Participant with Dementia

Biographical Information Form for the participant

Office use: participant nr

Please complete this questionnaire by writing the information in the spaces provided or
ticking the boxes where appropriate. The legal guardian, next of kin or the researcher, may
assist the person with dementia to complete this form.
1. What is your date of birth? _______/________/______
2. What is your age? _____________________________

Please tick  the appropriate box.

3. What is your gender?

Male

Female

4. Do you speak English as your first language?

Yes

No

5. What other languages do you speak?

English isiZulu
Afrikaans isiXhosa
Other

Specify _______________

6. What was your highest level of education?

Primary School Diploma Matric

High School Degree Post-graduate

7. What was your last occupation?

______________________________________________________________________
8. What is your marital status?

Single

218
© University of Pretoria
 Married
Divorced
Widow/widower
Other:
9. Do you have any problems with your vision?

Yes
No

10. Do you wear spectacles?

Yes
No

11. Do you have any problems with your hearing?

Yes
No

12. Do you wear hearing aids?

Yes

No

13. Which is your dominant hand?

Left

Right

14. Do you have any problems with your dominant hand?

Yes

No

If yes, please elaborate: ______________________

Completed by:

Thank you for taking the time to complete this form.

219
© University of Pretoria
 Appendix V: Biographical Questionnaire: Familiar Conversational Partner

Biographical information form for familiar communication partner

Office use: participant nr

Please complete this questionnaire by writing the information in the spaces provided or ticking the
boxes where appropriate.

1. What is your date of birth? _______/________/______

2. What is your age? _____________________________

Please tick  the appropriate box.

3. What is your relationship to the participant (person with dementia)?

Spouse
Child
Family member
Professional Caregiver
Other: Specify

4. What is your gender?

Male

Female

5. Do you speak English as your first language?

Yes

No

6. What other languages do you speak?

English isiZulu
Afrikaans isiXhosa
Other

Specify _______________

7. What was your highest level of education?

Primary School Diploma Matric
High School Degree Post-graduate

8. Do you have any problems with your vision?

Yes

No

9. Do you wear spectacles?

220
© University of Pretoria
 Yes
No

10. Do you have any problems with your hearing?

Yes
No

11. How many days in a week do you interact with the person with dementia?
1 day a week
2-3 days per week
3 days or more
Everyday
12. How much time is spent interacting with the person with dementia
15 minutes or less
15-30 minutes
30-45 mins
60 mins
13. Do you struggle to find things to talk about with the person with dementia?
Yes
No

If yes, please elaborate:

_______________________________________________________________________________
_______________________________________________________________________________
__________________________________________________________

14. List the three things that the person with dementia enjoys talking about?
1:___________________________
2: __________________________
3: __________________________

Thank you for taking the time to complete this form.

221
© University of Pretoria
 Appendix W: Procedural Checklist: Baseline Phase

Procedural Checklist: Baseline Phase

Procedural integrity

Initials of Observer: Date: - - 2020 Session: Baseline

Instructions to the observer:
▪ Insert a ✓ to indicate a ‘yes’ -the step was followed
▪ Insert a X to indicate a ‘no’ -the step was not followed
Procedure Yes No
Equipment and set-up
Video-camera switched on and set-up prior to participant entering the room
AAC device placed on the table with adjustable stand in front of the participant
Researcher and participant are seated next to each other facing the AAC device
Life story content is loaded onto the AAC device (visible on video recording)
Audio for voice-output plays during life story interaction
Greetings and introduction
Greet participant and orientate her to the task at hand:
“Good day Mrs…, My name is Adele and I am a speech-therapist. I doing a doctoral
study…”
Before the start of the life story, the researcher reiterates that the lifestory belongs to
the participant
“Mrs…these are your stories. You helped me write these stories for you….”
In each interaction opportunity with the electronic life story conversational support: The
researcher does the following…
Begins the interaction without demonstrating use of AAC device
Reads a pre-selected (researcher selected) life story topic (no choice offered to the
participant)
Reads life story on AAC device (without pointing to AAC symbols)
Provides general prompts “tell me about this...”
Responds to the participant neutrally e.g. “Oh”, “Hmm” (no elaborations, no
comments)
Speaks naturally (no prosodic emphasis)
Refers to the participant generally (during life story interaction there is no specific or
intentional use of participants name)
Closure
End the interaction (no pointing to AAC symbols)
“Thank you…., we have come to the end of the story today.”
Stakeholder Perspective
Obtain stakeholder perspective immediately after interaction

Total steps: 16
Total steps researcher adhered to: ____

222
© University of Pretoria
 Appendix X: Procedural Checklist: Intervention Phase

Procedural Checklist: Intervention Phase

Procedural integrity
Initials of Observer: Date: - - Intervention Session:
Instructions to the observer:
▪ Insert a ✓ to indicate a ‘yes’ -the step was followed
▪ Insert a X to indicate a ‘no’ -the step was not followed
Procedure Yes No
Equipment and set-up
Video-camera switched on and set-up prior to participant entering
AAC device placed on table with adjustable stand in front of participant
Researcher and participant are seated next to each other facing the AAC device
The content of the life story is loaded onto the AAC device (visible on video)
Greetings and introduction
Greet participant and orientate to the task at hand and says:
“Good day Mrs…, My name is Adele and I am a speech-therapist. I doing a doctoral
study…”
Before the start of the life story, the researcher reiterates that the life story belongs to the
participant
“Mrs …these are your stories. You helped me write these stories for you….”
Your stories are on a type of computer called an iPad.
Let me show you how the iPad works”
AAC device familiarisation: Does the researcher
Your stories are on this computer. I will show you how the computer works.
Demonstrate: light, single touch with pad of finger on the AAC device
Demonstrate: Press Home Button
Demonstrate: Press GoTalk app icon
Provide at least 3 opportunities for the participant to navigate AAC device (with or with
assistance)
General: Audio plays for voice-output during life story interaction
PC-AAC Scaffolding Strategies: Did the research apply the following (consistently for a minimum of
5 times in the life story)
1. Aided Modelling: Researcher points to AAC symbols on AAC device while
simultaneously reading life story (introduces story topic)
2. Person-centred comment or question with Aided Modelling: Researcher uses (person
recognition, person elaboration, empathetic responses) while pointing to AAC symbols
on AAC device
3. Expectant delay: Researcher turns toward participants, wait 5 seconds
4. Respond with Aided Modelling: responds by repeating the participants responses by
pointing to AAC symbols on the AAC device
If no response= no verbal, no non-verbal or inappropriate:
Repeat using aided modelling and wait 5 secs before moving on.
Responds to participant initiation, elaboration, repairs contingently with aided modelling before moving on.
Sub-total: PC-AAC Scaffolding strategies /20
Ending: Say” End of story” to signal end of the life story by pointing to AAC symbols
“Thank you Mrs …., we have come to the end of the story today.”
Stakeholder Perspective
Obtain stakeholder perspective immediately after interaction
Total score out of 33 /33

223
© University of Pretoria
 Appendix Y: Stakeholder Perspective Checklist

Question 1: Mrs, Brown, did you enjoy talking about your life story today?

not sure

Based on the participants response, ask the second question.

Question 2: Did you enjoy talking about your life story?

a little?

Or
a lot?

To support this conversation with the participant, the researcher should use multimodal AAC
strategies:
-Point to symbols while reading each question verbally
-Wait for the participant to respond
-Use natural gestures (hand gestures) to enhance meaning of the question
-Repeat or expand on participants responses
-Point to symbols to show an understanding of what the participant responds

224
© University of Pretoria
Appendix Z: Declaration of Originality

225
© University of Pretoria
Appendix AA: Declaration from Language Editor

226
© University of Pretoria