Educational Psychology Review

https://doi.org/10.1007/s10648-019-09461-9

REVIEW ARTICLE

Cognitive Load Theory and Human Movement: Towards

an Integrated Model of Working Memory

Stoo Sepp 1 & Steven J. Howard 1 & Sharon Tindall-Ford 1 & Shirley Agostinho 1 &
Fred Paas 1,2

# Springer Science+Business Media, LLC, part of Springer Nature 2019

Abstract
Cognitive load theory (CLT) applies what is known about human cognitive architecture to the
study of learning and instruction, to generate insights into the characteristics and conditions of
effective instruction and learning. Recent developments in CLT suggest that the human motor
system plays an important role in cognition and learning; however, it is unclear whether
models of working memory (WM) that are typically espoused by CLT researchers can
reconcile these novel findings. For instance, often-cited WM models envision separate infor-
mation processing systems—such as Baddeley and Hitch’s (1974) multicomponent model of
WM—as a means to interpret modality-specific findings, although possible interactions with
the human motor system remain under-explained. In this article, we examine the viability of
these models to theoretically integrate recent research findings regarding the human motor
system, as well as their ability to explain established CLT effects and other findings. We argue,
it is important to explore alternate models of WM that focus on a single and integrated control
of attention system that is applied to visual, phonological, embodied, and other sensory and
nonsensory information. An integrated model such as this may better account for individual
differences in experience and expertise and, parsimoniously, explain both recent and historical
CLT findings across domains. To advance this aim, we propose an integrated model of WM
that envisions a common and finite attentional resource that can be distributed across multiple
modalities. How attention is mobilized and distributed across domains is interdependent, co-
reinforcing, and ever-changing based on learners’ prior experience and their immediate
cognitive demands. As a consequence, the distribution of attentional focus and WM resources
will vary across individuals and tasks, depending on the nature of the specific task being
performed; the neurological, developmental, and experiential abilities of the individual; and
the current availability of internal and external cognitive resources.

Keywords Cognitive load theory . Working memory model . Attention . Human movement .
Gesturing . Learning

* Stoo Sepp
[email protected]

Extended author information available on the last page of the article

 Educational Psychology Review

Introduction

Working memory (WM) theories and models, the measurement of WM capacity, and the
neurological structures involved in WM have been investigated from a variety of
perspectives, including experimental psychology, educational psychology, linguistics,
neurology, and developmental psychology, spanning nearly 50 years of research. Within
the field of education, WM insights have informed educational practice through research
into attentional factors, schema acquisition and construction, spatial awareness, visual
representation, learning material design, and more. Cognitive load theory (CLT), in
particular, is responsible for a number of these advances through its consideration of
the implications of human cognitive architecture for the characteristics and conditions for
effective learning and instruction. Research in CLT has already identified a number of
empirically supported effects that have led to the identification of several instructional
principles that can inform teaching practice and the design of learning materials. This
includes findings that information is acquired more effectively when it is integrated
rather than distributed (split-attention effect; Chandler and Sweller 1992), when it is
augmented multimodally (modality effect; Tindall-Ford et al. 1997), and when this
information is complementary rather than redundant (redundancy effect; Kalyuga et al.
1999). It has also been found that many of these effects tend to reverse at higher levels of
expertise (expertise reversal effect; Kalyuga, Ayers, Chandler, and Sweller, 2003). More
recently, a number of studies investigating the human motor system’s role in the learning
process have suggested that gestures and other human movements can also be beneficial
in educational settings. As just one example, there is now emerging evidence that
physically enacting the concepts to be learned may support the acquisition of information
better than simple auditory presentation (Agostinho et al. 2015; Hu et al. 2014; Mavilidi
et al. 2015). These studies are typically interpreted in relation to widely accepted WM
models, such as those by Baddeley and Hitch (1974), Baddeley (1986), Baddeley 2000,
Baddeley 2003), and Cowan (1988, 1995, 2001).
While these WM models have previously served as a strong framework for CLT findings
(Sweller 1988; Sweller and Chandler 1991; Sweller et al. 1998) and related areas, it is unclear
how these theories can account for the cognitive effects of human movement deriving from
CLT. For instance, within Baddeley and Hitch’s (1974) multicomponent model of WM, it is
unclear how movement can be integrated into its auditory, visual-spatial, and executive control
systems.
In an attempt to reconcile emerging CLT insights with the principles of existing WM
models, this paper will first provide an overview of central concepts related to WM. Based on
findings related to WM depletion, attentional inclinations, and attentional distribution across
modalities, the role of attention in WM theory is examined, with a focus on how attentional
control plays an important role in cognition and learning. To account for these attentional
factors, a distributed attention model of WM is proposed—one which builds upon existing
models and posits a more dynamic and multidimensional approach to understanding WM
processes. A closer examination of the model follows, including how it may explain well-
established CLT effects. Recent research into the human motor system in educational contexts
is then critically discussed in terms of how gestures may be integrated into Baddeley and
Hitch’s model, with the suggestion that when contrasted with the proposed model, gestures and
human movements may constitute an additional modality. Finally, limitations of the proposed
model are discussed, along with potential areas of future inquiry.
Educational Psychology Review

Models of Working Memory

WM is a theoretical construct for describing the processes and systems related to the task-
relevant activation (bringing to mind), maintenance (holding in mind), and processing of
mental information during the performance of a task (Baddeley and Hitch 1974). Miller
(1956) originally proposed that unaided individuals’ capacity to receive, remember, and
process information was limited to seven units of information, plus or minus two (a figure
that has since been debated as being as low as four, or even one, unit; e.g., Cowan 2010). This
idea of a limited active memory capacity has served as the basic assumption for many years.
After Miller (1956), the next major advancement in WM theorizing was Atkinson and
Shiffrin’s (1968) model, which suggested a separation of the human memory system into three
subsystems: a (1) sensory register that processes substantial sensory information for a very
brief amount of time, with only the most pertinent information being temporarily activated in
the (2) short-term store (STS; now analogous to short-term memory (STM)), which, if
processed effectively, would be encoded into a (3) long-term store (LTS). This model is
important for two reasons. First, it separated a passive receipt of sensory input (in the sensory
register) from post-sensory processing (in STS), thereby positing a cognitive process that
influences what becomes activated in the STS—or, more accurately, within the focus of
attention—for processing. Second, it differentiated between different modalities, such that
visual linguistic inputs from the sensory register (e.g., a written word) may be encoded in the
STS in a corresponding visual-linguistic form.
Building upon this theorizing of modalities within the human memory system, Baddeley
and Hitch’s (1974) resultant multicomponent model of working memory (Baddeley 1983,
1986, 2000) sought to account for findings that extended beyond temporary activation of
information in the STS, to also include processing of the information through “reasoning,
comprehension and learning” (Baddeley and Hitch 1974, p. 201). While positioning their
model as describing a WM system, they retained a separation of visual and auditory modal-
ities. This multicomponent model reconceptualized STM as more than just a temporary storage
mechanism, but rather as a system capable of processing information across multiple sensory
inputs (a system more deserving of its description as working memory). As much of the
research at that time involved participants completing visual and auditory retention span tasks,
Baddeley and Hitch’s model postulated one WM system for each, supported by a central
processing system called the central executive (see Fig. 1). One of these modality-specific
“slave” systems was the phonological loop, in which sound- and speech-based information is
processed (Baddeley 1992). It can also register written information when processed as self-
speak. The other, the visuospatial sketchpad, is responsible for processing visual and spatial
information to support motor control (Baddeley and Lieberman 1980). Integrating and pro-
cessing information within the slave systems is the central executive, which also serves to
manage the distribution of attention.
Baddeley (2000) later added a third slave system called the “episodic buffer” (see Fig. 2), as
the existing model could not account for the formation of unitary representations comprised of
integrated information that spanned visual and auditory modalities. The episodic buffer was
thus described as a system that integrates visual, spatial, and verbal information, while also
acting as a bridge to long-term memory, facilitating retrieval from this system for integration
with available task-relevant information in other WM systems (Baddeley 2000).
Cowan (1988) found that existing explanations of how information is processed immediately
post-stimulus in the “pipeline” of human memory (e.g., Broadbent, 1958, Broadbent 1982) may not
 Educational Psychology Review

Central Executive
Visuospatial Phonological
Sketchpad Loop

Fig. 1 Diagram based on Baddeley and Hitch’s working memory model (Baddeley and Hitch 1974)

be well suited to account for the timing and sequence of mental processing across all contexts. In an
attempt to reconcile these findings, he aimed to describe processes underlying the selection of certain
stimuli and the dismissal of others via the application and focus of attention. Cowan’s theorizing
about WM included a number of departures from previous theories. For one, whereas Baddeley
described different WM systems for dealing with verbal and visual-spatial information, Cowan’s
work posited a single central attentional resource that was responsible for dealing with all forms of
information to be processed. Cowan’s research also included a quantification of WM capacity, for
which he posited that storage capacity (the amount of information that could be maintained in mind
without rehearsal or chunking strategies) and control of attention (processes that enable the
processing, combination, and reconciliation of this mentally represented information) needed to be
considered separately. As such, his estimate of WM capacity as three to five items reflects the
number of unrehearsed and uncombined pieces of information that can be concurrently activated in
WM (Cowan, 2000, Cowan 2010).

Fig. 2 Diagram based on Baddeley and Hitch’s working memory model (Baddeley 2000)
Educational Psychology Review

In this section, an overview or historical conceptions of WM have been discussed,

including the well-established models of Baddeley and Hitch (1974) and of Cowan (1998),
1999). Despite the differences between these models, the basic theoretical assumption of a
limited WM capacity has persisted since Miller (1956) first reported his findings. Yet, while
both theories retain their proponents and evidence base, the focus on a single attentional
resource as described by Cowan has been expanded upon by a number of WM theorists. The
next section discusses how the relationship between prior experience, expertise, gesturing, and
attentional distribution may provide a robust foundation for reconceptualizing WM processes.

The Distribution of Attention Within a Single Integrated Working

Memory System

Cowan’s (1995), 1999) WM model differs significantly from Baddeley and Hitch’s in that, like
earlier proposals of Pascual-Leone (1970) and later formulations by Engle (2002) and Heitz
and Engle (2007), there is no emphasis on distinguishing between processing different
modalities. Instead, words, images, gestures, or other modalities are not relegated to a specific
subsystem; rather, all information is processed through a single system with a limited capacity.
According to Cowan, WM involves more than just the mental activation and maintenance of
information but also, if nomenclature is taken as intended, working with information imme-
diately following exposure to a stimulus through effortless (e.g., recoiling your hand upon
touching a hot surface) and effortful attentional control processes (e.g., activating schemas
after being presented with a formula on a whiteboard). A number of studies have shown that
scope and control of attention are intrinsically linked in WM processes (Conway et al. 2005;
Engle 2002; Kane et al. 2001), although Cowan asserts they are separable and individually
constrained. For this reason, Cowan’s (2000) review asserted that Miller’s (1956) original “7
plus or minus 2” capacity limit was merely an estimate, and that its true limit may be around
three to five items if chunking and rehearsal are restricted. Cowan’s model, though important
for highlighting the concept of attentional focus as a means to discuss WM processes, does not
directly speak to how attention can play a role in information processing during problem
solving and learning—a core focus for CLT.
To explore the relationship between attention and the processing of information during
learning, an alternate conception of WM may provide a more meaningful framework for CLT
and educational contexts. For instance, Engle (2002) stands in contrast to Cowan by proposing
that WM capacity is not defined by individual differences in storage capacity alone, but also as
a function of the ability to control attention for goal-orientated means (Engle 2002). Unsworth
and Engle (2007) assert that WM capacity can actually be thought of as a product of two
concurrent processes: attentional focus on task-relevant information and ignoring task-
irrelevant information (suppression), and simultaneously using cues (e.g., stimuli) to reconcile
information from the environment and long-term memory for task completion. Further, in
contrast to Baddeley and Hitch’s focus on modality-specific WM stores, this theory concep-
tualizes attention as a domain-free resource for information processing (which underpins WM
capacity), and it diverges from Cowan in its inextricable inclusion of the processing of
(working with) information within the focus of WM.
When considering CLT’s redundancy effect though this lens, for example, as a learner
focuses on two or more sources of redundant information, limited attentional resources must,
regardless of the modality of information, be devoted to the reconciling and disregarding of
 Educational Psychology Review

any extraneous information, and the processing of relevant information. Successful task
completion in this case would depend on the ability of each learner to use attentional control
to reduce interference between redundant sources of information. Given that CLT effects such
as split attention, modality, and redundancy effects often require participants to use attentional
control in an attempt to establish the relationships between different sources of information,
Engle’s conceptualization of WM processing aligns well with individual differences in the
ability to control attentional resources. If we then consider WM capacity as dependent on the
ability to control a domain-free attentional resource, individual differences between learners
may be explained by experiential factors such as prior learning experiences or exposure to
different modalities. Differences in expertise may thus explain differential performance across
visual-spatial and auditory domains, as well as their differential susceptibility to interference.
For example, a musician might have more experience and thus more robust schemas and
strategies (e.g., chunking, understanding viable and less plausible connections) to deal with
auditory pattern information. Rather than separate WM systems, expert strategies for mobi-
lizing and applying attention to select, process, and reconcile information in WM may explain
their advantage in this domain. Indeed, de Groot’s (1965) seminal study found that expertise in
playing chess was determined by continual exposure, practice, and rehearsal, further suggest-
ing that these players have more experience to deal with visual-spatial pattern information in
this context (rather than a superior WM capacity, or even a generally more-advanced visual-
spatial sketchpad).
When also taking into account findings that demonstrate the benefit of iconic gesturing for
hearing participants, in contrast to the effects demonstrated with deaf learners, experience-
related biases in the application of attention (and competence in this application across
domains) may exist for all learners but merely are weighted differently. This distribution of
attention would be based on an individual’s previous experiences, expertise, available WM
resources, as well as the nature of the immediate task at hand. For example, learners with more
experience in reading may have an attentional inclination towards visuospatial information
(e.g., seek, be more familiar with, and have better strategies to interpret and integrate), while
learners with more experience in using their hands may have an attentional inclination towards
an embodied modality. This does not mean that attentional focus is placed on a single modality
at any given time, or that there are separate WM systems devoted to each, but rather an
attentional distribution that constantly shifts and adjusts across modalities in a manner most
efficient for the individual given the current task.
It is also important to emphasize that schemas activated, maintained, and processed across
these modalities do not require attentional focus to rapidly shift between systems or modalities
to be processed centrally. Instead, information across modalities can be activated, consolidated,
and processed within the focus of effortful mental attention (see Pascual-Leone, 1970). This
aligns with Halford et al.’ (1998) work on the nature of complexity, in that complexity—
regardless of the modalities from which it derives—is defined by the processing of relation-
ships between elements (i.e., items, elements, memory traces, schemas) activated concurrently
in WM plus the serial (in-sequence) processes required to reconcile them. In these formula-
tions, there is no need to define visuospatial, auditory, or embodied (or any additional) WM
subsystems. In fact, the presence of an as-yet-unknown quantity of modalities should be
assumed, so as to account for future findings.
For those with less competence processing information in a particular form (e.g., written
information), supportive strategies may reduce the cognitive load placed upon these learners.
In line with suggestions that gestures may temporarily “offload” mentally represented
Educational Psychology Review

information, Chu et al. (2013) found that participants with a lower visuospatial WM capacity
gestured more frequently than those with higher capacities. Brucker et al. (2015) similarly
found that participants with lower visuospatial abilities benefited from observing aligned
gestures when learning about fish locomotion patterns. Pouw et al. (2016) found that gesturing
reduced saccadic eye movements (a measure of cognitive processing) in problem-solving tasks
when participants had lower visual WM capacity. This suggests that gesturing may serve as a
general support mechanism for WM and attentional resources—especially if WM resources
devoted to that information is cognitively taxing—without the need for reference to distinct
WM systems. Indeed, Pouw et al. (2014a), in a review of the cognitive function of gestures,
assert that gestures provide an external support mechanism for internal cognitive processes
provided the gestures do not create a significant cognitive burden to interfere with the current
task at hand (see also Pouw et al. 2014b). Schmalenbach et al. (2017) extend this argument to
suggest gestures act as a supportive and compensatory mechanism when cognitive load is
high, when perceptual and cognitive abilities are low, or both.
This suggests that gestures can support WM processing in two ways. First, gestures may
temporarily offload mental information (Cook et al. 2012; Goldin-Meadow et al. 2001; Ping
and Goldin-Meadow 2010; for a review, see Risko and Gilbert 2016), to postpone the temporal
decay of a memory trace (i.e., “hold it in your hands”) and remove its demand from WM. With
the gesture physically maintaining this information, attentional resources normally devoted to
the internal maintenance of that information are offloaded, freeing up WM resources and
permitting attentional focus to be directed where it is most needed. Second, gestures can
support processing in WM through the physical embodiment of a process, concept, or object.
This can reframe a problem to be solved in an alternate modality, essentially externalizing the
problem itself. Rather than freeing WM resources for allocation elsewhere, this serves to focus
attention towards an alternative, aligned modality as a means to increase efficiency of problem
solving. For example, Chu and Kita (2008) found that when participants were asked to solve
mental rotation tasks, they used their hands as proxy objects for the item to be rotated. Further,
Pouw et al. (2014a) assert that for the high WM demands in novel tasks, gestures can serve as a
cognitive support mechanism (see Fig. 3). In contrast, distracting or redundant gestures may
have the opposite effect, given both forms of information would need to be reconciled and
integrated; indeed, De Koning and Tabbers (2013) found that making pointing gestures
simultaneously with an animated arrow did not provide any learning benefits, a result that
has implications for modality and human movement effects within CLT. By making or
observing unrelated or unnecessary gestures, this may impair the processing of activated
schemas and divert attentional focus from information essential for learning.
A recent development in CLT exploring the concept of WM depletion as a construct (Chen
et al. 2017; Healy, Hasher, & Danilova, 2011; Schmeichel 2007) may also have implications
for both attentional distribution and the use of gestures. Well-established results on the spacing
effect demonstrate that learning episodes over time result in increased retention when com-
pared to a single “massed” learning episode (Delaney et al. 2010; Ebbinghaus 1885/1964;
Gluckman et al. 2014; Kapler et al. 2015). This depletion effect asserts that WM resources are
not fixed and deplete over time based on varying cognitive demands—like a car’s fuel gauge
decreasing at different rates during a long drive across plains and through mountain passes.
While this topic continues to be debated, it nevertheless supports a single attentional resource
applied to the various foci of learning and suggests that gesturing may be able to forestall
depletion of WM resources. It does so not by refilling the tank, so to speak, but by forestalling
depletion by offloading some of the demand to a secondary resource (like a battery activating
 Educational Psychology Review

Fig. 3 Attentional redistribution

based on individual attentional
inclination and cognitive demands

in a hybrid car to reduce the car’s fuel consumption). Within the context of CLT, a model that
attempts to reconcile across these various findings and effects has not previously been
undertaken. As such, the next section will present an integrated model of WM, the one that,
envisioning a single attentional resource for the activation, maintenance, and processing of
information in WM, explains gestures, prior experience, expertise, and the distribution of
attentional focus towards learning and problem solving.

A Distributed Attention Model of Working Memory

The proposed integration of the variety of aforementioned WM principles and applications to

the breadth of CLT effects is presented in Fig. 4 as a visual representation that illustrates how
multiple modalities may combine within an integrated WM system in which a common
attentional resource activates all forms of information to facilitate processing. On the outside
are external environmental stimuli that learners may encounter, including instruction, learning
materials, and other aspects of the classroom environment or learning situation that may catch
their attention. Additionally, the external environment can also include externalized supports as
a means to redistribute cognitive resources in a manner more efficient for problem solving
(e.g., gesturing and reading aloud).
Educational Psychology Review

Fig. 4 The proposed distributed attention model of working memory including three traditionally investigated
foci of attention

Moving inward, the large circle—encompassing the periphery of attention and the focus of
attention—represents everything in WM, although the degrees of activation may vary.
Following the model of Pascual-Leone (1970), the outer ring reflects information that is less
activated within the periphery of attention (e.g., automatically or effortlessly), such as over-
learned, novel, salient, or misleading stimuli that nevertheless capture at least some of the
attentional resource. In the center circle is information within the effortful and intentional
focus of attention—those things currently and intentionally being processed in mind. Fol-
lowing the proponents of a domain-free attentional resource that underpins WM, the focus of
attention is comprised of information derived from any number of domains; visuospatial,
auditory, and embodied foci are some examples. It is important to also note that foci can be
both derived from the environment or searched for and activated in long-term memory
(LTM). The term “foci” is chosen specifically to both include modalities and build upon
them, as well as to include innumerable additional foci (e.g., sociocultural knowledge,
emotional state, language proficiency, cognitive differences, with any as-yet-unidentified
foci represented by blank circles). This model has included separated foci to allow for a
specific exploration of each to investigate whether learning effects may apply uniquely or
manifest differently across them. Foci provide opportunities to also explore how different
combinations of information across them may affect reinforcement and interference and, by
extension, learning.
 Educational Psychology Review

At the center of this model, which builds upon the work of Engle (2002) and many others
(Kane et al. 2001; Pascual-Leone and Baillargeon 1994; Pascual-Leone and Smith 1969;
Turner and Engle 1989; Unsworth and Engle, 2007), is the effortful and intentional focus of a
unitary and capacity-constrained attentional control mechanism. The limits of this system are
the reason we have a WM capacity limit. Our application and control of this attentional
resource—through the competitive process of our schemas and environmental cues vying for
our attention—determines which foci are selected for processing and which foci are dismissed,
based on the task at hand. This includes those involuntary external stimuli that successfully
capture our attention (e.g., an animated or novel stimulus), as well as the effortful focus of
attention on aspects of the situation we deem to be important. This information derived from
the environment is also reconciled, via attentional control processes, with information activat-
ed from LTM (which is also brought into the focus of attention for recombination, integration,
or reconciliation, as needed). Individuals’ attentional inclinations also influence the distribu-
tion of attention focus. For example, when faced with solving a math problem in the
classroom, some learners will focus attention on the voice of their instructor. Other learners
may focus solely on the whiteboard, while ignoring the voice of the instructor. Still others may
focus their attention on their textbook to derive the necessary information, returning their
attention to the instructor only at key junctures. Each of these can be seen as the consequence
of a competitive process between individual self-conceptions (i.e., beliefs about how one
learns most effectively), situational cues (e.g., instructions and emphases of the facilitator), and
individual schemas (e.g., learned strategies for which behaviors and processes are needed to
achieve one’s goals in this context). Therefore, attentional inclination towards specific modal-
ities is based on an individual’s experience and expertise, combined with the ability to control
the focus of attention across those foci, which will result in a diversity of learning strategies
across all students in a single classroom.
Attentional foci in different modalities are not processed independently, as articulated in
dual-processing models but interdependently in the focus of attention, connecting relevant
foci. The types of lines between each circle (dashed, solid, and gray) represent the different
ways in which attentional control can combine foci, which may result in either interference or
reinforcement. For example, when a learner engages with audiovisual learning materials in the
classroom and the modality effect is observed, visuospatial and auditory foci are combined to
reinforce concepts across sounds heard and diagrams presented. On the other hand, if auditory
information conflicts with visual information, additional attentional resources are devoted to
noticing and reconciling this information, resulting in interference. It should also be noted that
attentional focus across these foci imply a spatial element within each, as visual, embodied,
and auditory inputs can each provide spatial information to aid in situating an individual in a
specific location (Rhodes 1987; Smith et al. 2009).
In the classroom example above, if the instructor walks across the room to help a student,
this may redistribute WM resources for a moment through automatic and effortless shifts in
attentional focus (i.e., distraction). The speed and weight of this redistribution would depend
on an individual’s expectancies, needs, and previous experiences. To account for this, the
model assumes that attentional distribution across foci fluctuates over time, based on the
attentional inclination, expertise, previous experience, and current cognitive demands of the
learner. An individual’s experience with particular modalities may differ, so their attentional
inclination for, and shift in attentional focus towards, other modalities may also differ. This
model is designed to account for this so as to present the state of a learner in multiple ways.
Figure 4 presents a visualization that assumes equal attentional distribution across foci, while
Educational Psychology Review

Fig. 5a, b provides potential examples of how the model may conceivably display attentional
inclination towards specific foci over others (with or without combination effects). Another
possible static version of the model may visualize attentional distributions over time, capturing
key points over the course of task completion (Fig. 6), conveying the changes in attentional
distribution with foci growing and shrinking over time.
The proposed model thus integrates gestures as an additional modality within the focus of
attention, which can provide a means of presenting the details of reinforcement and interfer-
ence across foci during a learning experience, all while capturing the nuances of attentional
inclination and redistribution.
Though useful as an abstract visualization, this distributed attention model must also
account for CLT effects if it is to be useful for exploring cognition within the context of
education. The next section will discuss specific alignment with established cognitive load
effects and how the proposed model can explain these effects.

Distributed Attention and Cognitive Load Theory

The basic assumption of a limited WM capacity has led to research into the implications of
WM capacity constraints for learners and learning. Specifically, CLT was developed (Sweller
1988; Sweller and Chandler 1991; Sweller et al. 1998) as a branch of educational psychology
that seeks to inform educational practice through empirically supported instructional interven-
tions that account for the WM demands (load) of teaching, learning, and instruction. Building
upon Miller’s limited WM capacity of “7 plus or minus 2” pieces of information, CLT recast
WM terminology to describe units of information as “elements” of information that interact
with each other to create schematic links and support WM processing (Halford et al. 1998;
Sweller et al. 2011). For example, solving a simple math equation requires the cognitive
“juggling” of multiple elements maintained and processed in WM, such as numbers and their
meaning, rules governing how the numbers are to be processed, and the interpretation of
symbols required to solve a presented equation. During information processing (e.g., learning),
as the number of interacting elements being processed increases, cognitive load will also

Visuospatial Auditory Visuospatial Auditory

Foci
Foci Foci Foci

Embodied
Embodied
Foci
Foci

a b
Fig. 5 a, b Examples of individual differences in attentional inclination
 Educational Psychology Review

Fig. 6 Examples of key moments during learning episode

increase until WM resources reach their limit. A cognitive load at this level is not considered
negative, although when WM capacity is exceeded cognitive overload occurs. Cognitive
overload results in the loss of one or more elements from within WM and thus an inability
to process the meaning of, and relationships between, active elements with those that were lost.
In most learning contexts, this loss is an impediment to successful task completion and has a
negative impact on learning.
The work by Engle (2002) and colleagues (Kane et al. 2001) and others (Pascual-Leone and
Baillargeon 1994; Pascual-Leone and Smith 1969), which emphasizes the role that attentional
control plays in allocating resources within a limited-capacity WM system, served as a basis
for the proposed integrated WM model. One of CLT’s foundational assumptions is that
intrinsic and extraneous cognitive loads are determined by the number of multiple interacting
elements (items activated) within WM. This distributed attention model aligns with this
assumption, in that an increase in the number of foci processed within the focus of attention
increases the attentional (WM) resources required to reconcile and integrate these foci.
Importantly, this model accounts for a broad range of CLT effects, including emerging findings
exploring the human motor system’s role in cognition, which will be discussed later.
The first identified effect within CLT is the goal-free effect, which describes an increase in
learning outcomes observed when a problem is phrased without an end goal described.
Generally speaking, novices solving problems in subjects such as math and physics do not
know the solution of the problem they are working on. As a result, learners engage in a means-
ends analysis, which requires they hold in mind the goal, the known information presented in
the problem, the differences between this known information and the goal, and the strategies
and rules required to solve the problem. This results in an increased cognitive load due to the
required simultaneous activation, maintenance, and processing of a variety of different
schemas related to the problem. In a series of studies exploring the effects of this means-
ends analysis problem-solving process, Sweller and colleagues (Ayres 1993; Ayres and
Sweller 1990; Sweller and Levine 1982; Sweller 1988) found that when learners were
provided intermediate steps to solve without an end goal—as opposed to a whole problem
with a specific goal—learners were better able to find solutions through the exploration of
these intermediary. Within a CLT framework, this effect is explained in terms of multiple
elements interacting during the means-ends analysis process. When intermediate steps are
introduced and the goal removed, element interactivity and therefore cognitive load are
decreased, allowing WM resources to be freed to focus more on the steps involved in how
to solve the problem (germane load). In terms of the proposed model, this effect can be
explained through a reduction in the number of foci activated and the associated attentional
Educational Psychology Review

resources required to engage in means-ends analysis within the effortful focus of attention
(inner circle of the model). For example, to solve a problem, a learner would need to activate,
maintain, and process multiple foci related to the known elements of the problem, the goal, the
current state of their solution in relation to the goal, and the rules and strategies needed to reach
that goal. When the goal is removed and intermediate steps are the only stage the learner is
required to solve, limited attentional resources that were previously devoted to processing the
goal, its relationship to the problem, and its current state of completion are removed. This frees
limited resources to focus attention on exploring the nature of the problem itself, including
processing relationships between elements and how they fit into each stage of the problem,
thus fostering more robust schema creation and integration with existing expertise.
The split-attention effect is related to situations in which learners are exposed to instruc-
tional diagrams with mutually referring text and images separated spatially. This forces the
learner to split their attention to look back and forth between the diagram and explanatory text
(Chandler and Sweller 1992; Sweller and Chandler 1991). The mental integration of both
information sources needed to understand the learning materials imposes a high demand on
WM and can impact negatively on learning. In contrast, research shows that the associated
high cognitive load can be reduced and learning better facilitated by spatially integrating the
text into the diagrams (Chandler and Sweller 1992; Sweller and Chandler 1991) or by
replacing the visual text with spoken text (the modality effect; Mousavi et al. 1995; Tindall-
Ford et al. 1997). Mayer’s (2005) cognitive theory of multimedia learning (CTML) posits a
similar, yet more generally applicable concept called the “spatial contiguity principle” which
asserts that when related text and imagery are displayed near to each other, learners are able to
engage more deeply with the materials (for a meta-analysis of both principles, see Schroeder
and Cenkci 2018). The distributed attention model explains the split-attention effect in terms of
the cognitive cost of visual search between different foci. When materials are not integrated,
activated schema for each foci must be effortfully maintained within the focus of attention
(inner circle of the model) while the learner engages in visual search to seek out related
information (Schmidt-Weigand et al. 2010), thus decreasing available attentional resources and
increasing cognitive load. One study demonstrated a decrease in gaze shifts when learners
were exposed to non-integrated materials (Bauhoff et al. 2012). These reduced gaze shifts
suggest a longer duration of visual search, during which attentional resources must be devoted
to maintaining foci related to the task. Florax and Ploetzner (2010) further suggest that learning
can be facilitated through a well-organized integrated format intended to reduce visual search.
Other studies which use eye tracking to explore spatial contiguity suggest that integrated
materials result in increased eye movements (gaze shifts) between related text and images due
to a reduced requirement for visual search. This provides learners more opportunities to
establish relationships between the two foci (Holsanova et al. 2009; Johnson and Mayer
2012). By integrating text and images, the attentional resources required for schema mainte-
nance when visual search is being undertaken are reduced along with extraneous cognitive
load, allowing WM resources to better focus on establishing relationships in the support of
learning (germane load).
Derived from the split-attention effect is the modality effect, which occurs in mixed mode
instruction when visual learning materials (such as diagrams) are supplemented with comple-
mentary auditory information (e.g., a verbal statement in place of written text). This is more
effective for learning than single-modality instruction, because the combination of modalities
have been shown to make more effective use of WM resources by reinforcing each other
(Mousavi et al. 1995; Tindall-Ford et al. 1997). One explanation for the modality effect
 Educational Psychology Review

suggests that learning materials presented in both written and pictorial formats induce split
attention, which results in interference within the visual modality, making it challenging to
integrate this information (Low and Sweller 2005). By replacing the written statements with
spoken statements, split attention is reduced or negated due to the benefit of paired auditory
statements with pictorial information. This benefit is often explained through the lens of
separate slave systems defined in Baddeley and Hitch’s multicomponent model of WM or
Penney’s (1989) separate streams hypothesis whereby the WM system is comprised of
dedicated systems for processing visual and auditory information independent of one another,
offloading information from one channel to both channels and allowing for concurrent
processing and a reduction of cognitive load (Ginns 2005). Mayer’s cognitive theory of
multimedia learning (Mayer 1997, 2002, 2009; Mayer and Anderson 1992; Mayer and
Moreno 1998, 2003) refers to this explanation as the visuospatial load hypothesis (Mayer
2001). Rummer et al. (2010) argue this explanation is not compatible with Baddeley and
Hitch’s model due to the phonological loop being responsible for both written and verbal
(auditory) information retention and processing, thus negating the benefit. Another explanation
for the modality effect which is also used to describe the split-attention effect is the contiguity
assumption (Ginns 2006; Moreno & Mayer, 1998; Rummer et al. 2011) which refers to the
temporal delay between activated information in memory that occurs after the immediate
perception of separated text and pictures. In the case of the modality effect, the simultaneous
exposure to auditory and visual information reduces or eliminates this delay, while more
efficiently allocating WM resources for processing and learning. Further, Rummer et al. (2010)
suggest a third explanation based on the auditory recency effect (Penney 1989), which
describes when auditory information is more readily retained through sensory perception than
visual information, limited in scope to the most recently heard item. It is argued that this
internal verbal echo reduces limited WM resources required to integrate auditory and visual
information as the auditory information is retained in a more salient manner.
The split-attention and modality effects continue to be explored, discussed, and debated
through the lens of differing explanations for these effects rooted in perception and WM
processing. The distributed attention model presented in this paper explains the modality effect
by reframing established assumptions. Instead of simultaneous processing in separate subsys-
tems dedicated to each modality, the distributed model frames these explanations through the
allocation of attentional resources combined and distributed across modalities within a single,
limited-capacity focus of attention system. When a learner is exposed to mutually referential
auditory and visual information such as a diagram with a supporting audio statement, cognitive
load is reduced in two ways. First, activated foci combined across modalities within the
effortful focus of attention (inner circle of the model) do not need to be reconciled as they
are not in the same modality; thus, visuospatial load hypothesis still applies in the proposed
model, retaining the assumption that interference can occur between two foci of in the same
modality (see Fig. 6). Further, activation of foci in a connected and distributed manner allows
for more efficient processing of relationships and creation of unitary representations within
WM. Second, when learners are exposed to multimodal learning materials, automatic and
effortless activation of foci across modalities occurs within the periphery of attention (outer
ring of the model), simultaneously, and inclusive of a more robust activation of immediate
auditory information as described by the auditory recency effect. As there is little to no delay
between activations of these foci, attentional resources that would be devoted to maintaining
activated foci and searching for other related foci within the same modality are no longer
required; thus, the contiguity assumption still holds. When considering these differing
Educational Psychology Review

explanations for the modality effect, the distributed attention model is able to account for each
through the lens of attentional focus and distribution, which succinctly describes the activation,
combination, and integration of attentional foci across modalities, resulting in a more efficient
allocation of attentional resources, which supports learning and schema creation.
As a further extension of the split-attention effect, the redundancy effect (Chandler &
Sweller, 1991; Kalyuga et al. 1999) shows that when information is duplicated–either in the
same modality or in a different one–learning can be negatively impacted. The redundancy
effect and expertise reversal effects refer to a nullification of cognitive efficiencies through the
presentation of redundant information or existing expertise, respectively (Kalyuga et al. 2003;
Kalyuga et al. 1999). A study by Kalyuga et al. (1999) demonstrated that when learners were
presented with multimedia instructions related to mechanical engineering, learners who were
presented with auditory statements corresponding to visual diagrams outperformed those who
were given just written statements or combined written and auditory statements. According to
Baddeley and Hitch’s (1974) model, these compatible sources of information would be
activated and processed in separate slave systems, but this model does not appear to account
for why an impairment in learning should occur in this manner. Alternatively, Cowan’s (1988,
1995) theorizing of a single WM system can explain this redundancy in terms of two or more
sources of duplicated information requiring reconciliation and integration within the scope and
control of attention, all of which accumulates to tax available WM resources. In terms of the
distributed attention model, the explanation is similar to Cowan’s in that the need to reconcile
redundant information across two different foci wastes WM resources in the pursuit of
reconciling what has already been presented. With regard to expertise reversal, if students
are presented with information they have already mastered, existing schemas related to this
expertise are automatically retrieved from LTM and activated within the effortful focus of
attention (inner circle of the model). This results in an increase in attentional resources devoted
to reconciling existing knowledge with unnecessarily presented foci (extraneous load) and may
reduce learning outcomes.
For each described CLT effect, interpretation does not require allusion to a multidomain
WM system or the separation of storage from processing. Instead, the processing of infor-
mation (any information) places similar demand on attentional resources for working with
information in WM as processing of that information. As a final example, consider simple
multiplication; before this process is automated, mentally calculating the product of two
digits is more cognitively demanding than simply holding those digits in mind. According to
this model, this is because the number and complexity of elements (including the complexity
of mental manipulations to be performed on these elements), and their interactivity, all draw
upon a unitary but limited attentional resource. Differences in performance across domains
are explained by our different levels of expertise within those domains, just as expertise
playing an instrument does not automatically transfer to expertise with another. In these
propositions, CLT effects can be explained both comprehensively and simply, providing a
model that does not need to be revised with each new domain or effect uncovered by CLT
researchers.
In this section, we have discussed how the proposed distributed attention model may
explain well-established effects and instructional principles identified over many years of
CLT research. While the models of Baddeley and Hitch (1974), Cowan (1988), and Mayer
(2005) are able to explain specific CLT effects in terms of perception, processing, and storage
where appropriate, these models are individually unable to account for every CLT effect in a
flexible and robust manner. This issue is further exacerbated by recent CLT findings regarding
 Educational Psychology Review

the human motor system, which these WM models have not yet explicitly attempted to
reconcile. In the following sections, an overview of research into the human motor system
and its effect on learning is discussed, including how the proposed model integrates these
findings in contrast to established models, while also positioning gestures and human move-
ments as an additional modality.

How the Human Motor System Supports Learning and Cognition

When considering the reinforcement or interference between sources of information across

different modalities, educational and psychological research has historically focused on
isolating auditory and visual information through passive exposure to these sources. However,
a question remains about whether and how other modalities may also play a role in learning
and cognition. For instance, most learners are able to involve their own motor system in the
learning process, by using either fine motor movements, such as hand and finger gestures (e.g.,
finger counting, tracing, and pointing), or gross motor movements, such as whole body
movements (e.g., mimicking others and enacting concepts). It is thus plausible that human
movement may constitute one of many additional modalities that have yet to be comprehen-
sively considered in relation to existing WM models.
A human movement effect was recently identified within CLT, suggesting that learning
procedural motor tasks from dynamic visualizations, such as animations, can be more effective
than learning from static visualizations when the transient animated information incorporates
human movement (Ayres et al. 2009; Castro-Alonso et al. 2014; Paas and Sweller 2012; Wong
et al. 2009; see also Höffler and Leutner 2007). For instance, Wong et al. (2009) found that
when learners observed animations of paper folding from a first-person perspective, they were
able to replicate the folding task with increased proficiency when compared with those who
viewed static images of each step. While the animations did not show hands, they presented
paper folding as if hands were present, suggesting a potential role for gestures in WM
processes.
The field of embodied cognition (Foglia and Wilson 2013) is a separate area of research that
is viewed as a meaningful theoretical companion to CLT (Sweller et al. 2011). Embodied
cognition asserts that cognitive processes, including information processing and learning, are
inextricably linked with all forms of sensory input (not just sight and sound), including
physical and environmental experiences of an individual. In a basic sense, embodied cognition
frames all cognition as being intrinsically linked with sensory and motor functions within the
environment, including gestures and other human movements (Barsalou 1999). Although not
originally rooted in educational contexts, embodied cognition presents a meaningful anchor for
existing research in CLT focusing on human movement and gestures, allowing a theoretical
link to be made (Choi et al. 2014; Paas and Sweller 2012). Framed in this way, a learner’s
motor functions, including gestures and tactile experiences, play a similar role to (and
introduce similar effects as) visual and auditory information in learning. In one study of note,
tangible physical user interfaces in a multimedia learning context were investigated for
learning anatomy, with results demonstrating increased learning outcomes when compared
to a traditional mouse and computer display experience (Skulmowski et al. 2016). This, along
with studies on the use of cognitive load measurement for embodied learning (for a review, see
Skulmowski and Rey 2017), has begun to position physical experiences as an established area
of inquiry along with cognitive load.
Educational Psychology Review

Within a CLT framework, the use of iconic hand gestures (that is, those that meaningfully
represent objects, visuospatial traits, or actions) has been shown to benefit learning of a foreign
language (Macedonia and Klimesch 2014; Mavilidi et al. 2015). When preschool-aged
children produced iconic gestures representing an action that matches its foreign language
word/phrase, such as acting out the word for “swim” while learning the word in Italian
(nuotare), children’s learning outcomes were enhanced and when full-body movement was
compared with arm and hand gesturing while sitting, the learners engaged in full-body
movement benefitted further (Mavilidi et al. 2015). These same types of gestures also seem
capable of supporting mimicry. When observing an adult demonstrates how a toy worked
through an iconic gesture (pretending to act on an object nearby without touching it), an
infant’s ability to successfully operate the toy increased (Novack et al. 2015). These gesturing
effects are also supported by findings in neurology around the mirror neuron system (MNS).
The MNS is a neurological system located in the premotor cortex that activates when humans
and other nonhuman primates observe an action performed by another individual. This
observation is thought to cognitively prime the observer to perform the same action (for a
review, see Rizzolatti and Craighero 2004), a finding that serves as a partial explanation for the
human movement effect (Van Gog et al. 2009).
In contrast to research in more-controlled environments such as research labs, a body of
literature has grown around exploring the use of gestures and their effects in more traditional
learning environments, such as classrooms and preschools. Research by Goldin-Meadow and
others (Cook et al. 2013; Cook et al. 2008; Goldin-Meadow 2009; Goldin-Meadow et al.
2009; Novack and Goldin-Meadow 2015), exploring the role of pointing, iconic gesture, and
metaphoric gestures (i.e., gestures that represent abstract ideas) in math and language learning,
has demonstrated that these gestures provide a support mechanism that improves learning
outcomes. In explaining these findings, it has been suggested that gestures could be used for
the “cognitive offloading” of information during problem solving, leading to a reduction in
cognitive load (Cook et al. 2012; Goldin-Meadow et al. 2001; Ping and Goldin-Meadow 2010;
Risko and Gilbert 2016). These findings provide further evidence for the positive role that
gestures may play in classroom-based learning, positioning gestures as an additional contrib-
utor to learning alongside more traditional auditory and visual inputs. How gesturing effects
are explained by existing and emerging theories of WM, however, requires a discussion of
gestures from a psychological perspective.
Efforts have begun to explain these motor-related findings through the construct of WM.
Research exploring how human movement relates to WM (Engelkamp 1995; Engelkamp et al.
2005; Engelkamp et al. 1994) has focused specifically on the memory traces of action
performance. In one study, participants were exposed to both ordinary and “bizarre” action
phrases (actions that are novel or surreal, such as “plant the hammer”), with half of participants
learning the phrases with only verbal prompts and the other half performing a motor task
aligned with the phrase. When participants were given a list of phrases and asked to identify
which of them they had learned, those who performed the actions during the learning phase
demonstrated increased recognition (Engelkamp et al. 1994). While Engelkamp explained
these findings through an extension of dual coding theory (DCT), it is nevertheless informative
for research focusing on WM. DCT posits that mental representations can be visual and verbal
in nature, processed through two distinct and co-reinforcing channels (Paivio and Okovita
1971; see also Clark and Paivio 1991). However, Wilson (2001) observes that although
memory coding for human movement is empirically supported, “recent models have shown
a consistent trend away from sensorimotor representations” (p. 44). This raises questions about
 Educational Psychology Review

the appropriateness of CLT referencing WM models of Baddeley and Hitch (1974) and Cowan
(1988) to explain the effects of human movement on learning. As such, in the next section,
Baddeley and Hitch’s WM model will be discussed in terms of whether gestures and
movement might be reconcilable with the multicomponent model.

Gestures as an Additional Modality

In contrast to Baddeley and Hitch’s focus on verbal and visual-spatial information, there is a
growing body of evidence which suggests that motor information may constitute an additional
modality that can also occupy WM’s limited resources. For instance, Wilson and Emmorey’s
(1997) work with speakers of American Sign Language (ASL) found that when participants’
rehearsal of signs included interference (performing a nonsense sign), their recall of actual
signed concepts was reduced. Paralleling CLT findings of a negative impact of extraneous
information, the nonsense signs reduced overall performance (that is, they placed demands on
WM, leading to cognitive overload and loss of information). Based on these results, the
researchers noted, “the working memory system of a deaf ASL signer contains a rehearsal
loop that possesses many of the structural properties of the phonological loop for speech”
(Wilson and Emmorey 1997, p. 319). It is thus important to consider that while the phono-
logical loop is traditionally tied to spoken language, this is not the only form of communication
available. In a review by Rudner et al. (2009), the authors concluded that sign language and
spoken language, from both a WM and neurological perspective, are similar. However, it was
found that deaf individuals who signed from a young age exhibit an inclination towards spatial
organization of information, whereas hearing participants often prefer temporal organization
(Cumming and Rodda 1985). For example, spoken languages such as English are linear in
nature with a sequenced ordering of sounds to represent ideas, whereas signed languages such
as ASL rely on the spatial referents and semantic structures of sequenced gestures, body
movements, and facial expressions to convey meaning. This means that experience in a
particular language can play a role in WM processing through an inclination towards the
modality of that language (spoken or signed). Wilson and Emmorey (2003) also found that
when hearing and deaf participants were asked to recall concepts presented in their first
language, only the deaf participants were sensitive to interference inputs presented in ASL.
This mirrors similar findings for hearing participants presented with written interference tasks
(e.g., reading while engaged in a word-span task; Turner and Engle 1989).
These findings suggest that signed languages use similar cognitive processes to spoken and
written languages and are similarly susceptible to interference and reinforcement. Further,
proficiency in a particular language influences WM processing in favor of that language’s
modality (e.g., visuospatial, auditory, gesture), which, by extension, can influence the appli-
cation and allocation of WM resources. These findings are important for conceptions of
separate WM systems for different modalities: first, because they contribute to previous work
in psychology through the investigation of modality isolation and cognitive interference, and
second, because it indicates that gestures and human movement may be considered both
visuospatial and verbal, so relegating them to the phonological loop or visuospatial sketchpad
alone may not be viable.
Given that signed languages and spoken languages can share similar WM processes, the
distinction between Baddeley and Hitch’s phonological loop and visuospatial sketchpad in
terms of where gestures fit is unclear. Baddeley (2012) himself even raised questions about
Educational Psychology Review

how other physical experiences such as tactile and nonspeech kinesthetic experiences could be
integrated within his existing model, and suggested how slave systems could possibly exist for
even more senses and even more types of stimuli input (see Fig. 7). However, how these might
all interact, and how this would account for reinforcement and interference effects across
different combinations of modalities, remains unclear.
Instead of investigating and revising an ever-expanding list of WM systems, their functions,
and their interactions, it is contended that distributed attention model presented in this paper
can better account for the inclusion of multiple sensory and processing domains, through a
single and integrated attentional resource that is involved in the activation of visual, phono-
logical, and embodied information (within the scope of attention), while also accounting for
individual differences in experience and expertise (through the control of attention).
The distributed attention model thus provides a framework for integrating gestures and
human movement as a modality which, much like auditory and visual information, provide
learning gains through reinforcement or learning deficits through the interference between
multiple sources of information within the focus of attention. The human movement effect is
therefore framed as the freeing of WM resources for the creation of schema when learning is
supported by movement (reinforcement).
In this section, recent findings in embodied cognition and the human movement effect have
been presented, including how the human motor system can support learning. A discussion of
how these findings may be integrated into established models of WM presents a unique
challenge while also speaking to the affordances of the proposed model for the integration
of gestures and human movement as an additional modality. In the next section, limitations of
the proposed model are discussed including its scope, potential for future application, and
opportunities for validation.

Limitations of the Proposed Model

Studies investigating cognitive processes across multiple sensory modalities such as auditory
and visuospatial have traditionally explored these areas from the perspective of cognitive
interference and reinforcement. Isolating specific modalities (i.e., attentional foci) has provided

Fig. 7 Diagram based on Baddeley’s speculative view of the flow of information from perception to working
memory (Baddeley 2012)
 Educational Psychology Review

many theoretical and practical advances both in psychology and learning sciences. The
primary limitation of the distributed attention model presented in this paper is that it is not
formulated to explore modalities in isolation, but instead, the interaction between them. While
it may not be possible to explore isolated modalities through the lens of this model, it does
afford the opportunity to investigate a learners’ ability to control the focus and distribution of
attention across modalities. The dynamic nature of this model serves primarily as a theoretical
framework for representing individual differences and changes in WM resource allocation
over time, which itself may be presently challenging to quantify. In addition, the proposed
model does not explicitly account for initial perceptual limitations but focuses more on post-
stimulus processing within WM. The intentional dynamic nature of the model could, however,
be used to indicate initial attentional inclination and attentional focus immediately post-
stimulus to explore the relationships between modalities in future research.
To investigate further, future studies may choose to build upon measures of attentional focus,
including eye tracking with an emphasis on tagging or categorizing objects of visual focus over
time. Recent advances in motion tracking technology may also allow for capturing of gross full
body movements, as well as fine hand and finger movements. As the ability to capture informa-
tion related to the human motor system advances, attentional focus on auditory information may
still prove challenging to quantify in educational settings. Collecting subjective learner reflections
on external or internal differences in attentional focus, including distractions, cognitive or
emotional challenges, and previous cultural experiences may also provide a more nuanced picture
of cognitive processes during task completion. It may also prove beneficial to investigate
distribution of attention through a combination of these measures along with subjective learner
perception of their own inclination towards specific attentional foci.

Conclusion

Given uncertainty around how recent gesture effects could be reconciled within the WM
frameworks that CLT researchers normally defer to, we attempted a reconciliation and also cast
our attention more widely. In doing so, an integrated distributed attention model of WM was
proposed, which accounts for the breadth of CLT findings in a comprehensive but parsimo-
nious way. This model integrates WM principles and insights from theorists including Cowan,
Engle, Pascual-Leone, and others. By presenting a unitary attentional (WM) resource that can
integrate information from multiple sources and modalities and can adjust the distribution of
attentional focus during processing, this model may assist in explaining, clarifying, recasting,
and supporting CLT findings now and in the future.

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

References

Agostinho, S., Tindall-Ford, S., Ginns, P., Howard, S. J., Leahy, W., & Paas, F. (2015). Giving learning a helping
hand: finger tracing of temperature graphs on an iPad. Educational Psychology Review, 27(3), 427–443.
https://doi.org/10.1007/s10648-015-9315-5.
Atkinson, R. C., & Shiffrin, R. M. (1968). Human memory: a proposed system and its control processes. In K. W.
Spence & J. T. Spence (Eds.), The psychology of learning and motivation (Vol. 2, pp. 89–195). London:
Academic.
Educational Psychology Review

Ayres, P. L. (1993). Why goal-free problems can facilitate learning. Contemporary Educational Psychology,
18(3), 376–381. https://doi.org/10.1006/ceps.1993.1027.
Ayres, P., Marcus, N., Chan, C., & Qian, N. (2009). Learning hand manipulative tasks: when instructional
animations are superior to equivalent static representations. Computers in Human Behavior, 25(2), 348–353.
https://doi.org/10.1016/j.chb.2008.12.013.
Ayres, P., & Sweller, J. (1990). Locus of difficulty in multi-stage mathematics problems (Ph.D. thesis). The
American Journal of Psychology, 103(2), 167–193.
Baddeley, A. D. (1983). Working memory. Philosophical Transactions of the Royal Society B: Biological
Sciences, 302(1110), 311–324. https://doi.org/10.1098/rstb.1983.0057.
Baddeley, A. D. (1986). Working memory (p. 1986). Oxford: Oxford University Press.
Baddeley, A. D. (1992). Working memory. Science, 255(5044), 556–559. https://doi.org/10.1126
/science.1736359.
Baddeley, A. D. (2000). The episodic buffer: a new component of working memory? Trends in Cognitive
Sciences, 4(11), 417–423. https://doi.org/10.1016/S1364-6613(00)01538-2.
Baddeley, A. D. (2003). Working memory: looking back and looking forward. Nature Reviews Neuroscience,
4(10), 829–839. https://doi.org/10.1038/nrn1201.
Baddeley, A. D. (2012). Working memory: theories, models, and controversies. Annual Review of Psychology,
63(1), 1–29. https://doi.org/10.1146/annurev-psych-120710-100422.
Baddeley, A. D., & Hitch, G. (1974). Working memory. In G. H. Bower (Ed.), The psychology of learning and
motivation: advances in research and theory (Vol. 8, pp. 47–89). New York: Academic.
Baddeley, A. D., & Lieberman, K. (1980). Spatial working memory. In R. S. Nickerson (Ed.), Attention and
performance VIII (pp. 521–539). Hillsdale: Lawrence Erlbaum Associates, Inc.
Barsalou, L. W. (1999). Perceptions of perceptual symbols. Behavioral and Brain Sciences, 22(04), 637–660.
Bauhoff, V., Huff, M., & Schwan, S. (2012). Distance matters: spatial contiguity effects as trade-off between gaze
switches and memory load. Applied Cognitive Psychology, 26(6), 863–871. https://doi.org/10.1002
/acp.2887.
Broadbent, D. E. (1958). Perception and Communication. New York: Pergamon.
Broadbent, D. E. (1982). Task combination and selective intake of information. Acta Psychologica, 50(3), 253–
290.
Brucker, B., Ehlis, A.-C., Häußinger, F. B., Fallgatter, A. J., & Gerjets, P. (2015). Watching corresponding
gestures facilitates learning with animations by activating human mirror-neurons. An fNIRS study, 36(C), 27–
37. https://doi.org/10.1016/j.learninstruc.2014.11.003.
Castro-Alonso, J. C., Ayres, P., & Paas, F. (2014). Learning from observing hands in static and animated versions
of non-manipulative tasks. Learning and Instruction, 34, 11–22.
Chandler, P., & Sweller, J. (1991). Cognitive load theory and the format of instruction. Cognition and Instruction,
8(4), 293–332. https://doi.org/10.1207/s1532690xci0804_2.
Chandler, P., & Sweller, J. (1992). The split-attention effect as a factor in the design of instruction. British Journal
of Educational Psychology, 62(2), 233–246. https://doi.org/10.1111/j.2044-8279.1992.tb01017.x.
Chen, O., Castro-Alonso, J. C., Paas, F., & Sweller, J. (2017). Extending cognitive load theory to incorporate
working memory resource depletion: evidence from the spacing effect. Educational Psychology Review,
61(2), 1–19. https://doi.org/10.1007/s10648-017-9426-2.
Choi, H. H., Van Merrienboer, J. J. G., & Paas, F. (2014). Effects of the physical environment on cognitive load
and learning: towards a new model of cognitive load. Educational Psychology Review, 26(2), 225–244.
Chu, M., & Kita, S. (2008). Spontaneous gestures during mental rotation tasks: insights into the
microdevelopment of the motor strategy. Journal of Experimental Psychology: General, 137(4), 706–723.
https://doi.org/10.1037/a0013157.
Chu, M., Meyer, A., Foulkes, L., & Kita, S. (2013). Individual differences in frequency and saliency of speech-
accompanying gestures: the role of cognitive abilities and empathy. Journal of Experimental Psychology:
General, 143, 694–709. https://doi.org/10.1037/a0033861.supp.
Clark, J. M., & Paivio, A. (1991). Dual coding theory and education. Educational Psychology Review, 3(3), 149–
210.
Conway, A. R. A., Kane, M. J., Bunting, M. F., Hambrick, D. Z., Wilhelm, O., & Engle, R. W. (2005). Working
memory span tasks: a methodological review and user’s guide. Psychonomic Bulletin & Review, 12(5), 769–
786. https://doi.org/10.3758/BF03196772.
Cook, S. W., Duffy, R. G., & Fenn, K. M. (2013). Consolidation and transfer of learning after observing hand
gesture. Child Development, 84(6), 1863–1871. https://doi.org/10.1111/cdev.12097.
Cook, S. W., Mitchell, Z., & Goldin-Meadow, S. (2008). Gesturing makes learning last. Cognition, 106(2),
1047–1058. https://doi.org/10.1016/j.cognition.2007.04.010.
 Educational Psychology Review

Cook, S. W., Yip, T. K., & Goldin-Meadow, S. (2012). Gestures, but not meaningless movements, lighten
working memory load when explaining math. Language and Cognitive Processes, 27(4), 594–610.
https://doi.org/10.1080/01690965.2011.567074.
Cowan, N. (1988). Evolving conceptions of memory storage, selective attention, and their mutual constraints
within the human information-processing system. Psychological Bulletin, 104(2), 163–191.
Cowan, N. (1995). Attention and memory: an integrated framework. New York: Oxford University Press;
Oxford: Clarendon, 1995.
Cowan, N. (2000). The magical number 4 in short-term memory: a reconsideration of mental storage capacity.
Behavioral and Brain Sciences, 24(1), 87–185.
Cowan, N. (2010). The magical mystery four. Current Directions in Psychological Science, 19(1), 51–57.
https://doi.org/10.1177/0963721409359277.
Cumming, C. E., & Rodda, M. (1985). The effects of auditory deprivation on successive processing. Canadian
Journal of Behavioural Science / Revue Canadienne des Sciences du Comportement, 17(3), 232–245.
Delaney, P. F., Verkoeijen, P. P. J. L., & Spirgel, A. (2010). Spacing and testing effects: a deeply critical, lengthy,
and at times discursive review of the literature. In B. H. Ross (Ed.), The psychology of learning and
motivation: advances in research and theory (Vol. 53, pp. 63–147). New York: Academic. https://doi.
org/10.1016/S0079-7421(10)53003-2.
De Groot, A. (1965). Thought and choice in chess. The Hague: Mouton. Original work published 1946.
De Koning, B., & Tabbers, H. K. (2013). Gestures in instructional animations: a helping hand to understanding
non-human movements? Applied Cognitive Psychology, 27, 683–689.
Ebbinghaus, H. (1885/1964). Memory: a contribution to experimental psychology. Oxford: Dover.
Engelkamp, J. (1995). Visual imagery and enactment of actions in memory. British Journal of Psychology, 86(2),
227–240. https://doi.org/10.1111/j.2044-8295.1995.tb02558.x.
Engelkamp, J., Seiler, K. H., & Zimmer, H. D. (2005). Differential relational encoding of categorical information
in memory for action events. Memory & Cognition, 33(3), 371–389.
Engelkamp, J., Zimmer, H. D., Mohr, G., & Sellen, O. (1994). Memory of self-performed tasks: self-performing
during recognition. Memory & Cognition, 22(1), 34–39. https://doi.org/10.3758/BF03202759.
Engle, R. W. (2002). Working memory capacity as executive attention. Current Directions in Psychological
Science, 11(1), 19–23. https://doi.org/10.1111/1467-8721.00160.
Florax, M., & Ploetzner, R. (2010). What contributes to the split-attention effect? The role of text segmentation,
picture labelling, and spatial proximity. Learning and Instruction, 20(3), 216–224. https://doi.org/10.1016/j.
learninstruc.2009.02.021.
Foglia, L., & Wilson, R. A. (2013). Embodied cognition. Wiley Interdisciplinary Reviews: Cognitive Science,
4(3), 319–325.
Ginns, P. (2005). Meta-analysis of the modality effect. Learning and Instruction, 15(4), 313–331. https://doi.
org/10.1016/j.learninstruc.2005.07.001.
Ginns, P. (2006). Integrating information: a meta-analysis of the spatial contiguity and temporal contiguity
effects. Learning and Instruction, 16(6), 511–525. https://doi.org/10.1016/j.learninstruc.2006.10.001.
Gluckman, M., Vlach, H. A., & Sandhofer, C. M. (2014). Spacing simultaneously promotes multiple forms of
learning in children’s science curriculum. Applied Cognitive Psychology, 28(2), 266–273. https://doi.
org/10.1002/acp.2997.
Goldin-Meadow, S. (2009). How gesture promotes learning throughout childhood. Child Development
Perspectives, 3(2), 106–111.
Goldin-Meadow, S., Cook, S. W., & Mitchell, Z. (2009). Gesturing gives children new ideas about math.
Psychological Science, 20(3), 267–272. https://doi.org/10.1086/659964?ref=search-gateway:4d5c7866182
d0f01d9bef79f563ad121.
Goldin-Meadow, S., Nusbaum, H., Kelly, S. D., & Wagner, S. (2001). Explaining math: gesturing lightens the
load. Psychological Science, 12(6), 516–522.
Healey, M. K., Hasher, L., & Danilova, E. (2011). The stability of working memory: do previous tasks influence
complex span? Journal of Experimental Psychology: General, 140, 573–585. https://doi.org/10.1037
/a0024587.
Heitz, R. P., & Engle, R. W. (2007). Focusing the spotlight: individual differences in visual attention control. Journal
of Experimental Psychology: General, 136(2), 217–240. https://doi.org/10.1037/0096-3445.136.2.217.
Hu, F. T., Ginns, P., & Bobis, J. (2014). Does tracing worked examples enhance geometry learning? Australian
Journal of Educational Developmental Psychology, 14, 45–49.
Kalyuga, S., Ayres, P., Chandler, P. A., & Sweller, J. (2003). The expertise reversal effect. Educational
Psychologist, 38(1), 23–31. https://doi.org/10.1207/S15326985EP3801_4.
Kalyuga, S., Chandler, P., & Sweller, J. (1999). Managing split-attention and redundancy in multimedia
instruction. Applied Cognitive Psychology, 25(S1), 123–S144. https://doi.org/10.1002/acp.1773.
Educational Psychology Review

Kane, M. J., Bleckley, M. K., Conway, A. R. A., & Engle, R. W. (2001). A controlled-attention view of working-
memory capacity. Journal of Experimental Psychology: General, 130(2), 169–183. https://doi.org/10.1037
//0096-3445.130.2.169.
Kapler, I. V., Weston, T., & Wiseheart, M. (2015). Spacing in a simulated undergraduate classroom: long-term
benefits for factual and higher-level learning. Learning and Instruction, 36, 38–45. https://doi.org/10.1016/j.
learninstruc.2014.11.001.
Halford, G. S., Wilson, W. H., & Phillips, S. (1998). Processing capacity defined by relational complexity:
implications for comparative, developmental, and cognitive psychology. Behavioral and Brain Sciences, 21,
803–831.
Höffler, T. N., & Leutner, D. (2007). Instructional animation versus static pictures: a meta-analysis. Learning an
Instruction, 17(6), 722–738.
Holsanova, J., Holmberg, N., & Holmqvist, K. (2009). Reading information graphics: the role of spatial
contiguity and dual attentional guidance. Applied Cognitive Psychology, 23(9), 1215–1226. https://doi.
org/10.1002/acp.1525.
Juan Pascual-Leone, (1970) A mathematical model for the transition rule in Piaget's developmental stages. Acta
Psychologica, 32 301–345
Johnson, C. I., & Mayer, R. E. (2012). An eye movement analysis of the spatial contiguity effect in multimedia
learning. Journal of Experimental Psychology: Applied, 18(2), 178–191. https://doi.org/10.1037/a0026923.
Low, R., & Sweller, J. (2005). The modality principle in multimedia learning. In R. E. Mayer (Ed.), The
Cambridge handbook of multimedia learning (pp. 147–158). New York: Cambridge University Press.
Macedonia, M., & Klimesch, W. (2014). Long-term effects of gestures on memory for foreign language words
trained in the classroom. Mind, Brain, and Education, 8(2), 74–88.
Mavilidi, M. F., Okely, A. D., Chandler, P. A., Cliff, D. P., & Paas, F. (2015). Effects of integrated physical
exercises and gestures on preschool children’s foreign language vocabulary learning. Educational
Psychology Review, 27(3), 413–426. https://doi.org/10.1007/s10648-015-9337-z.
Mayer, R. E. (1997). Multimedia learning: Are we asking the right questions? Educational Psychologist, 32(1),
1–19.
Mayer, R. E. (2001). Multimedia learning. Cambridge: Cambridge University Press.
Mayer, R. E. (2002). Multimedia learning. Psychology of Learning and Motivation, 41, 85–139.
Mayer, R. E. (2005). Cognitive theory of multimedia learning. In R. E. Mayer (Ed.), The Cambridge handbook of
multimedia learning (pp. 31–48). New York: Cambridge University Press.
Mayer, R. E. (2009). Multimedia learning (p. 2009). Cambridge: Cambridge University Press.
Mayer, R. E., & Anderson, R. B. (1992). The instructive animation: helping students build connections between
words and pictures in multimedia learning. Journal of Educational Psychology, 84(4), 444–452. https://doi.
org/10.1037/0022-0663.84.4.444.
Mayer, R. E., & Moreno, R. (1998). A split-attention effect in multimedia learning: evidence for dual processing
systems in working memory. Journal of Educational Psychology, 90(2), 312–320. https://doi.org/10.1037
/0022-0663.90.2.312.
Mayer, R. E., & Moreno, R. (2003). Nine ways to reduce cognitive load in multimedia learning. Educational
Psychologist, 38(1), 43–52. https://doi.org/10.1207/S15326985EP3801_6.
Miller, G. A. (1956). The magical number seven, plus or minus two: some limits on our capacity for processing
information. Psychological Review, 63(2), 81–97. https://doi.org/10.1037/h0043158.
Mousavi, S. Y., Low, R., & Sweller, J. (1995). Reducing cognitive load by mixing auditory and visual
presentation modes. Journal of Educational Psychology, 87(2), 319–334. https://doi.org/10.1037/0022-
0663.87.2.319.
Novack, M., & Goldin-Meadow, S. (2015). Learning from gesture: how our hands change our minds.
Educational Psychology Review, 27(3), 405–412.
Novack, M., Goldin-Meadow, S., & Woodward, A. L. (2015). Learning from gesture: how early does it happen?
Cognition, 142, 138–147.
Paas, F., & Sweller, J. (2012). An evolutionary upgrade of cognitive load theory: using the human motor system and
collaboration to support the learning of complex cognitive tasks. Educational Psychology Review, 24(1), 27–45.
Paivio, A., & Okovita, H. W. (1971). Word imagery modalities and associative learning in blind and sighted
subjects. Journal of Verbal Learning and Verbal Behavior, 10(5), 506–510.
Pascual-Leone, J., & Baillargeon, R. (1994). Developmental measurement of mental attention. International
Journal of Behavioral Development, 17(1), 161–200. https://doi.org/10.1177/016502549401700110.
Pascual-Leone, J., & Smith, J. (1969). The encoding and decoding of symbols by children: a new experimental
paradigm and a neo-Piagetian model. Journal of Experimental Child Psychology, 8(2), 328–355. https://doi.
org/10.1016/0022-0965(69)90107-6.
Penney, C. G. (1989). Modality effects and the structure of short-term verbal memory. Memory & Cognition,
17(4), 398–422. https://doi.org/10.3758/BF03202613.
 Educational Psychology Review

Ping, R., & Goldin-Meadow, S. (2010). Gesturing saves cognitive resources when talking about nonpresent
objects. Cognitive Science, 34(4), 602–619.
Pouw, W., Mavilidi, M.-F., Van Gog, T., & Paas, F. (2016). Gesturing during mental problem solving reduces eye
movements, especially for individuals with lower visual working memory capacity. Cognitive Processing,
17(3), 269–277. https://doi.org/10.1007/s10339-016-0757-6.
Pouw, W. T. J. L., de Nooijer, J. A., Van Gog, T., Zwaan, R. A. & Paas, F. (2014a). Toward a more embedded/
extended perspective on the cognitive function of gestures. Frontiers in Psychology, 5, 1–14. https://doi.
org/10.3389/fpsyg.2014.00359.
Pouw, W., Van Gog, T., & Paas, F. (2014b). An embedded and embodied cognition review of instructional
manipulatives. Educational Psychology Review, 26(1), 51–72.
Risko, E. F., & Gilbert, S. J. (2016). Cognitive offloading. Trends in Cognitive Sciences, 20(9), 676–688.
https://doi.org/10.1016/j.tics.2016.07.002.
Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. Annual Review of Neuroscience, 27(1), 169–
192. https://doi.org/10.1146/annurev.neuro.27.070203.144230.
Rhodes, G. (1987). Auditory attention and the representation of spatial information. Perception & Psychophysics,
42(1), 1–14.
Rudner, M., Andin, J., & Rönnberg, J. (2009). Working memory, deafness and sign language. Scandinavian
Journal of Psychology, 50(5), 495–505. https://doi.org/10.1111/j.1467-9450.2009.00744.x.
Rummer, R., Schweppe, J., Fürstenberg, A., Seufert, T., & Brünken, R. (2010). Working memory interference
during processing texts and pictures: implications for the explanation of the modality effect. Applied
Cognitive Psychology, 24(2), 164–176. https://doi.org/10.1002/acp.1546.
Rummer, R., Schweppe, J., Fürstenberg, A., Scheiter, K., & Zindler, A. (2011). The perceptual basis of the
modality effect in multimedia learning. Journal of Experimental Psychology: Applied, 17(2), 159–173.
https://doi.org/10.1037/a0023588.
Schmalenbach, S. B., Billino, J., Kircher, T., van Kemenade, B. M., & Straube, B. (2017). Links between
gestures and multisensory processing: individual differences suggest a compensation mechanism. Frontiers
in Psychology, 8, 635–638. https://doi.org/10.3389/fpsyg.2017.01828.
Schmeichel, B. J. (2007). Attention control, memory updating, and emotion regulation temporarily reduce the
capacity for executive control. Journal of Experimental Psychology: General, 136(2), 241–255. https://doi.
org/10.1037/0096-3445.136.2.241.
Schmidt-Weigand, F., Kohnert, A., & Glowalla, U. (2010). Explaining the modality and contiguity effects: new
insights from investigating students’ viewing behaviour. Applied Cognitive Psychology, 24(2), 226–237.
https://doi.org/10.1002/acp.1554.
Schroeder, N. L., & Cenkci, A. T. (2018). Spatial contiguity and spatial split-attention effects in multimedia
learning environments: a meta-analysis. Educational Psychology Review, 30(3), 679–701.
Skulmowski, A., Pradel, S., Kühnert, T., Brunnett, G., & Rey, G. D. (2016). Embodied learning using a tangible
user interface: the effects of haptic perception and selective pointing on a spatial learning task. Computers &
Education, 92-93(c), 64–75. https://doi.org/10.1016/j.compedu.2015.10.011.
Skulmowski, A., & Rey, G. D. (2017). Measuring cognitive load in embodied learning settings. Frontiers in
Psychology, 8, 1–6. https://doi.org/10.3389/fpsyg.2017.01191.
Smith, D., Davis, B., Niu, K., Healy, E., Bonilha, L., Fridriksson, J., Morgan, P. S., & Rorden, C. (2009). Spatial
attention evokes similar activation patterns for visual and auditory stimuli. Journal of Cognitive
Neuroscience, 22, 347–361.
Sweller, J. (1988). Cognitive load during problem solving: effects on learning. Cognitive Science, 12(2), 257–
285. https://doi.org/10.1207/s15516709cog1202_4.
Sweller, J., Ayres, P., & Kalyuga, S. (2011). Cognitive load theory. New York: Springer Science & Business Media.
Sweller, J., & Chandler, P. A. (1991). Cognitive load theory and the format of instruction. Cognition and
Instruction, 8(4), 293–332. https://doi.org/10.1207/s1532690xci0804_2.
Sweller, J., Levine, M., 1982. (1982). Effects of goal specificity on means–ends analysis and learning. Journal of
Experimental Psychology, 8(5), 463–474.
Sweller, J., Van Merrienboer, J. J., & Paas, F. (1998). Cognitive architecture and instructional design. Educational
Psychology Review, 10(3), 251–296.
Tindall-Ford, S., Chandler, P., & Sweller, J. (1997). When two sensory modes are better than one. Journal of
Experimental Psychology: Applied, 3, 257–287.
Turner, M. L., & Engle, R. W. (1989). Is working memory capacity task dependent? Journal of Memory and
Language, 28(2), 127–154.
Unsworth, N., & Engle, R. W. (2007). The nature of individual differences in working memory capacity: active
maintenance in primary memory and controlled search from secondary memory. Psychological Bulletin,
114(1), 104–132. https://doi.org/10.1037/0033-295X.114.1.104.
Educational Psychology Review

Van Gog, T., Paas, F., Marcus, N., Ayres, P., & Sweller, J. (2009). The mirror neuron system and observational
learning: implications for the effectiveness of dynamic visualizations. Educational Psychology Review, 21,
21–30.
Wilson, M. (2001). The case for sensorimotor coding in working memory. Psychonomic Bulletin & Review, 8(1),
44–57.
Wilson, M., & Emmorey, K. (1997). A visuospatial “phonological loop” in working memory: evidence from
American Sign Language. Memory & Cognition, 25(3), 313–320. https://doi.org/10.3758/BF03211287.
Wilson, M., & Emmorey, K. (2003). The effect of irrelevant visual input on working memory for sign language.
Journal of Deaf Studies and Deaf Education, 8(2), 97–103.
Wong, A., Marcus, N., Ayres, P., Smith, L., Cooper, G. A., Paas, F., & Sweller, J. (2009). Instructional animations
can be superior to statics when learning human motor skills. Computers in Human Behavior, 25(2), 339–
347. https://doi.org/10.1016/j.chb.2008.12.012.

Affiliations

Stoo Sepp 1 & Steven J. Howard 1 & Sharon Tindall-Ford 1 & Shirley Agostinho 1 & Fred
Paas 1,2

1
Early Start and School of Education, University of Wollongong, Wollongong, NSW 2522, Australia
2
Department of Psychology, Education, and Child Studies, Erasmus University Rotterdam, Rotterdam,
The Netherlands