See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/51236456

Cognitive Reserve Index questionnaire

(CRIq): A new instrument for measuring
cognitive reserve
ARTICLE in AGING - CLINICAL AND EXPERIMENTAL RESEARCH JUNE 2011
Impact Factor: 1.22 DOI: 10.3275/7800 Source: PubMed

CITATIONS

READS

34

821

3 AUTHORS, INCLUDING:
Daniela Mapelli

Sara Mondini

University of Padova

University of Padova

64 PUBLICATIONS 1,147 CITATIONS

52 PUBLICATIONS 556 CITATIONS

SEE PROFILE

SEE PROFILE

Available from: Sara Mondini

Retrieved on: 12 February 2016

Aging Clinical and Experimental Research

Cognitive Reserve Index questionnaire (CRIq):

a new instrument for measuring cognitive reserve
Massimo Nucci1, Daniela Mapelli1,2 and Sara Mondini1,3
1Department of General Psychology, University of Padova, 2Centro Interdipartimentale CIRMANMEC,
University of Padova, 3Casa di Cura Figlie di San Camillo, Cremona, Italy

ABSTRACT. Background and aims: The concept of reserve has been used to explain the difference between individuals in their capacity to cope with or compensate for
pathology. Brain reserve refers to structural aspects of the
brain, such as brain size and synapse count. Cognitive reserve is the ability to optimize and maximize performance through two mechanisms: recruitment of brain
networks, and/or compensation by alternative cognitive
strategies. The aim of the present research was to devise
an instrument for comprehensive assessment and measurement of the quantity of cognitive reserve accumulated
by individuals throughout their lifespan. Methods: A new
approach using the Cognitive Reserve Index questionnaire
(CRIq) was developed and tested in a sample of 588
healthy individuals, from 18 to 102 years old, stratified
by age (Young, Adults, Elderly) and gender. The CRIq includes demographic data and items grouped into three
sections: education, working activity and leisure time,
each of which returns a subscore. The WAIS Vocabulary
test and TIB were also administered. Results: The main
descriptive features and some inferential results are described. Intelligence was only moderately correlated
with cognitive reserve, stressing the distinction between
these two concepts. Age and gender significantly affected CRIq scores, whereas no effect emerged from
their interaction. Adults showed a higher score than
Young and Elderly. Conclusions: This study provides a
new instrument for a standardized measure of the cognitive reserve accumulated by individuals through their
lifespan. The potential use of the CRIq in both experimental research and clinical practice is discussed.
(Aging Clin Exp Res 2012; 24: 218-226)

ability to cope with physiological or pathological cognitive

decline. There is not always a direct relationship between the severity of brain pathologies or brain damage
and the degree of deficit in performance. For many
years, Brain Reserve (BR) was the prevalent construct of
the potential ability of the brain to cope with neuronal
damage. Katzman et al. (1) examined the brains of ten
subjects who had documented post-mortem neuropathology of Alzheimers dementia (AD), even though
they had not expressed any sign of cognitive decline
when alive. The authors attributed the absence of clinical
signs of dementia to the higher-than-average weight of
their brains. Later, BR was defined as the brains resilience: that is, the possibility of the brain itself coping
with increasing brain damage (2). The Brain Reserve
hypothesis is primarily a passive-quantitative model related
to individual differences (e.g., brain size and synapse
count); a greater BR is considered as a protective factor,
and a lower one indicates vulnerability.
The debate on BR and aging introduced and developed
the concept of Cognitive Reserve (CR), a fascinating
concept at the basis of brain plasticity. The Cognitive Reserve hypothesis suggests that the brain actively attempts
to cope with damage by using pre-existing cognitive processes or enlisting compensatory strategies. Thus, people
with a high CR can withstand more age-related changes
and disease-related pathologies by effectively and flexibly
using cognitive paradigms or compensatory brain networks (2-7). However, there is no a clear-cut distinction between BR and CR. As Stern (2) explicitly suggested, there
is neural implementation of CR in terms of efficiency, capacity and flexibility of synaptic reorganization, and in
terms of the relative utilization of specific brain regions.
Similarly, intensive cognitive stimulation may be associated
with increased brain volume in childhood (e.g., 8).
So far, CR has been estimated by extremely heterogeneous methods and more than a few proxies, as shown

s
i
t
ur

K
e
c
i
r
LY
t
N
i
O
d
E
E
S
U
,
L
2
A
1
N
O
S
20
PER
R
O
F

2012,

Editrice Kurtis

INTRODUCTION
The concept of reserve was proposed in the late
1980s to explain the differences among individuals in their

Key words: Brain reserve, cognitive reserve, cognitive reserve questionnaire, dementia.
Correspondence: Massimo Nucci, Department of General Psychology, University of Padova, Via Venezia 8, 35131 Padova, Italy.
E-mail: [email protected]
Received February 18, 2011; accepted in revised form May 24, 2011.
First published ahead of print June 20, 2011 as DOI: 10.3275/7800

Aging Clin Exp Res, Vol. 24, No. 3 218

M. Nucci, D. Mapelli and S. Mondini

Table 1 - Cognitive reserve proxies identified in 24 studies explicitly evaluating CR.

Study

Education

Occupation

Intelligence

Leisure activity

Alexander et al. (9)

numerical scale
(years of education)

no

no

Bialystok et al. (10)

numerical scale
(years of education)

Christensen et al. (11)

3-point scale (less than

10, 10-12, over 12
years of education)
numerical scale
(years of education)
numerical scale
(years of education)
numerical scale
(years of education)
4-point scale (less than 5, 6-9,
10-12, over 12 years of education)
numerical scale (years of education)
numerical scale (years of education)
and 4-point scale (qualification)

5-point scale (Human

Resources and Skills
Development, Canada, 2001)
no

demographics-based IQ
estimation; WRAT
(reading subtest)
no
Spot-the-Word Test

Self-directed
Search Test

AMNART; Ravens
Progressive Matrices Test
no

no

NART

no

Daffner et al. (12)

Garibotto et al. (13)
Garrett et al. (14)
Le Carret et al. (15)
Martino et al. (16)
McDowell et al. (17)

no
6-point scale
(NEST-DD project protocol)
no

Perneczky et al. (19)

Ropacki et al. (20)

3-point scale (less than 5, 6-8,

over 8 years of education)
numerical scale (years of education)
no

no

s
i
t
ur

no

no

no
6-point scale
(Statistics Canadas standard
occupational codes, 1981)
no

WAIS-R (Vocabulary subtest)

no

no
no

no

no

K
e
c
i
r
LY
t
N
i
O
d
E
E
S
U
,
L
2
A
1
N
O
S
20
PER

Ngandu et al. (18)

no

no

no
no
7-point scale
OPIE
(US Census categories, 1994 )
no
no

no
no

no

NART

Sol-Padulls et al. (23) 4-point scale (qualification)

customized 4-point scale

WAIS-R (Vocabulary subtest)

Spitznagel et al. (24)

Staff et al. (25)

no
numerical scale (years of education)

WRAT
no

Stern et al. (26)

numerical scale (years of education)

no
9-point scale (GB: Office of
Population Censuses
and Surveys, 1990)
no

customized
questionnaire
customized
questionnaire
no
no

Roselli et al. (21)

Scarmeas et al. (22)

dichotomous scale (less or more

than 8 years of education)
numerical scale (years of education)

R
O
F

Tucker-Drob et al. (27) dichotomous scale (less or more

than 12 years of education)
Valenzuela &
numerical scale (years of
Sachdev (28)
education) plus courses

no
9-point ordinal scale
(Australian Standard
Classification of
Occupations, 1997)

from a systematic literature revision of CR indicators

(some examples are given in Table 1).
Education is one of the first and most commonly used
proxies in studies on CR (17-19, 29, 30). Education
plays a role in the cognitive decline in normal aging, as
well as in degenerative disease or traumatic brain injury.
Nevertheless, it is frequently recognized that higher (or
lower) education levels have an influence on adult lifestyles.
The effect is rather difficult to isolate from other protective factors such as a successful job, the awareness of
health risks, and the quality of the social environment,
amongst others. The proxy is usually indicated by the

219 Aging Clin Exp Res, Vol. 24, No. 3

no

WAIS-R (Vocabulary subtest); no

NART
Kit of factor-referenced
no
cognitive tests
no
customized
questionnaire

number of years of education (or, alternatively, the degree

of literacy), in some cases on an ordinal scale and in
others a numerical one.
Several studies have shown that occupation may provide an additive and independent source of CR throughout a persons lifetime (13, 25, 31). The last (or the
longest) job is usually taken into account. Occupation
has a different value according to the cognitive load involved. Perceived prestige and/or salary are also common
indices.
Over and above education level and occupation, epidemiological evidence has shown that premorbid en-

Cognitive Reserve Index questionnaire (CRIq)

gagement in leisure activities may also provide a separate

or synergic increase in CR (22, 23, 32, 33). Intellectual,
social and physical activities are usually considered. Several activities have been recorded by means of instruments
applying various types and numbers of items, target periods and frequencies (23, 28, 34, 35). The rationale behind these studies is that experiences acquired during
adulthood and later can affect reserve.
Intelligence is another frequently used index of CR (9, 22,
23, 36), in which I.Q. or pre-morbid I.Q. are the most common proxies used to estimate CR. Also in this case, the instruments used to evaluate I.Q. are quite heterogeneous.
Some of the most common tests are the Vocabulary Subtest
of the Wechsler Adult Intelligence Scale (WAIS, 37) and the
National Adult Reading Test (NART, 38).
Therefore, even when the same proxies are used, the
procedures and measurement scales are difficult to compare and the various studies are, so far, not always easily comparable. For the same reasons, use of the CR in
clinical settings is rather difficult.
Our research aimed at standardizing a new procedure to quantify the amount of CR accumulated by individuals throughout their lifetimes by introducing a new
questionnaire, the Cognitive Reserve Index questionnaire (CRIq), and a new index, the Cognitive Reserve Index (CRI). The relevance and potential use of the CRIq in
both experimental research and clinical practice is discussed.

WAIS (Italian version, 39); 2) the TIB (Test di Intelligenza Breve, 40, English translation: Short test of Intelligence,
a reading test very similar to the NART, 38).
CRIq scale construction
The CRIq includes some demographic data (date and
place of birth, gender, place of residence, nationality, marital status), and 20 items grouped into three sections, education, working activity, and leisure time, each of which
returns a subscore.
CRI-Education: years of education plus possible training courses (lasting at least six months); the raw score of
this section is the sum of these two values.
CRI-WorkingActivity: adulthood professions. Five different levels of working activities are available, dealing with
the degree of intellectual involvement and personal responsibility: unskilled, manual work (e.g., farmer, car
driver, call centre operator); skilled manual work (e.g.,
craftsman, clerk, hairdresser); skilled non-manual or technical work (e.g., trader, kindergarten teacher, real estate
agent); professional occupation (e.g., lawyer, psychologist,
physician); highly intellectual occupation (e.g., university
professor, judge, top manager). Working activity was
recorded as the number of years in each profession over
the lifespan. The raw score of this section was the result
of years of working activity multiplied by the cognitive level of job (from one to five), as detailed above.
CRI-LeisureTime: cognitively stimulating occupations
carried out during leisure time (out of working time or
school schedule). Sixteen items were related to various intellectual activities (e.g., reading newspapers or books,
playing music), social activities (participation in charitable
activities, going to a museum, travel) and physical activities (sports, dancing). The frequency (i.e., never/rare,
often/always) and the number of years (how long each activity had been carried out) were recorded. The raw score
of this section was the total number of years of activity for
which frequency was often/always. A score related to the
number of children was also included.
The CRIq questionnaire (Italian, French and English versions), instructions, and the Excel file for automatic calculation of scores, are available at http://cri.psy.unipd.it.

s
i
t
ur

K
e
c
i
r
LY
t
N
i
O
d
E
E
S
U
,
L
2
A
1
N
O
S
20
PER

METHOD
Participants and data collection
A total of 588 participants, randomly selected from the
general Italian population (323 women, 55%), were enrolled in this study. Their age ranged from 18 to 102
years old (50.2119.62, women 51.9120.99 and men
48.1317.65) and was arbitrarily divided into the following three groups: Young, from 18 to 44 years old
(n=246; 31.527.89), Adults, from 45 to 69 (n=212;
546.70) and Elderly, from 70 to 102 (n=120;
78.876.28). Participants were healthy and without evident neurological or psychiatric illness; no other specific exclusion criteria were used. The participants did not receive any compensation for taking part in the study.
Trained psychologists (all studying for a Masters degree) administered the CRIq in single individual sessions
lasting about 15 minutes. The CRIq is not anonymous;
however, if participants objected to registering their
names, a fictitious name was generated. All data were collected from September 2009 to June 2010.

R
O
F

Intelligence assessment
As the existence of a correlation between CR and intelligence is undoubted, in order to quantify this relationship, two tests highly correlated with intelligence
were administered: 1) the Vocabulary Test from the

Computation of CRI
The raw scores of the three sections of the CRIq
were correlated with age by the number of years an activity had been carried out (correlation r=-0.56 for education; r=0.48 for working activity, r=0.66 for leisure time
activity). In order to rule out this age effect, three linear
models were used: the raw scores of the three sections
were set as dependent variables, and age as the independent (or predictor) variable (see Fig. 1).
The three CRIq subscores (CRI-Education, CRIWorkingActivity, CRI-LeisureTime) were the residuals of
the relative linear models, standardized and transposed to

Aging Clin Exp Res, Vol. 24, No. 3 220

M. Nucci, D. Mapelli and S. Mondini

250
500

20

15

10

200
CRI-LeisureTime (rough score)

CRI-WorkingActivity (rough score)

CRI-Education (rough score)

25

150

100

50

400

300

200

s
i
t
ur

100

Fig. 1 - Scatter-plots of raw

scores of three sections of
CRIq according to age. Lines:
estimators of linear regression model, with age as predictable variable and raw
scores of sections as dependent variables.

K
e
c
i
r
LY
t
N
i
O
d
E
E
S
U
,
L
2
A
1
N
O
S
20
PER
0

20

40

60 80
Age

100

20

40

60 80
Age

100

a scale with M=100 and SD=15. This allowed all participants to be systematically compared with their corresponding age class. Lastly, CRI (total CRIq score) was the
average of the three subscores, again standardized and
transposed to a scale with M=100 and SD=15. The
higher the CRI, the higher the estimated CR. CRI could
be classified into five ordered levels: Low (less than 70),
Medium-low (70-84), Medium (85-114), Medium-high
(115-130) and High (more than 130).

R
O
F

Toward a validation of the CRIq

The CRI-LeisureTime section was progressively simplified and improved by means of the Item Response Theory (41, 42), by either excluding redundant, non-discriminative items or by joining some of them. The actual sixteen items used showed good reliability (=0.73,
95% CI [0.70, 0.76]).
The majority of elderly Italians did not have more than
five years of education, for social and/or historical reasons.
Therefore, in our database it is not surprising that elderly
people had low CRI-Education but high CRI-WorkingActivity or high CRI-LeisureTime scores. Instead, at present,
young subjects with Masters degrees (high CRI-Education) may have unskilled jobs (low CRI-WorkingActivity),
due to the current crisis of the job market. Therefore, all
three sections of the questionnaire equally co-occurred to the
CRI (reliability of CRIq =0.62, 95% CI [0.56, 0.97]). The
prediction was that, although the three subscores were
all proxies of the same construct (i.e., CR), the correlation
between them would not be very high.

221 Aging Clin Exp Res, Vol. 24, No. 3

20

40

60 80
Age

100

Because of the lack of standardized questionnaires for estimating the CR, a concurrent validation of the CRIq could
not easily be obtained. Intelligence is the construct which
is most closely related to the CR (and was thus used as a
predictive proxy in some studies). However, the CRI and IQ
are not equivalent to each other. Thus, a perfect correlation
between them is neither expected, nor desirable. In our
sample, the correlation between CRI and two tests considered as being highly correlated with intelligence (Vocabulary tests from WAIS and TIB) was around 0.45.

RESULTS
Statistical analyses are organized into three sections:
Frequencies, Correlations, ANOVAs.
Frequencies
Some main descriptive features from the data collected (588 participants) are reported, in order to shape
the whole database. The raw CRI-Education score
(12.285.34) showed frequency peaks around 5, 8, 13
and 18 years of school, easily linked to Italian levels of education. As predicted, Elderly showed a lower level of education than Adults and Young (7.334.17, 11.984.93
and 15.154.11, respectively). Women also showed a
lower level of education than men (11.585.30 and
13.135.28, respectively).
In CRI-WorkingActivity, low skilled manual work was
the most frequently recorded activity (33%), followed by
skilled manual work, professional occupation and skilled
non-manual work (20%, 18% and 15%, respectively). On-

Cognitive Reserve Index questionnaire (CRIq)

Table 2 - Percentages of types of working activity according to cognitive resources involved.

Never employed
Low skilled manual work: agricultural worker, waiter, driver,
mechanic, plumber, call centre operator, etc.
Skilled manual work: craftsman, clerk, cook, shop assistant,
tailor, nurse, professional soldier, barber/haidresser, ect.
Skilled non manual work: shopkeeper, white-collar worker,
priest or monk-nun, sales representative, estate agent, musician, etc.
Professional occupation: CEO of a small company, lawyer,
physician, psychologist, engineer, teacher, etc.
Highly responsible or intellectual occupation: CEO of large
company, politician, university professor, judge, surgeon, etc.

ly 1% of the sample had a highly responsible or intellectual occupation; 13% had been never employed. The frequencies of various working activities across gender and
age are listed in Table 2.
CRI-LeisureTime showed Housework and Driving as
the most frequently recorded activities (12%). The least
were Going to the cinema/theatre, and Charitable or artistic activities (about 2%). Table 3 lists the frequencies
across gender and age.

Total
n=588

Females
n=323

Males
n=265

Young
n=246

Adults
n=212

Elderly
n=130

13%
33%

16%
37%

9%
28%

24%
26%

1%
35%

11%
45%

20%

15%

27%

18%

23%

21%

15%

15%

15%

15%

17%

12%

18%

16%

20%

17%

23%

10%

1%

0%

2%

0%

1%

1%

s
i
t
ur

pendence between the main variables involved in CRIq.

The CRI showed a satisfactory correlation with the
three subscores, CRI-Education, CRI-WorkingActivity
and CRI-LeisureTime: r=0.77, r=0.78 and r=0.72, respectively1 (see Fig. 2). As predicted, the correlation between the subscores was not high: CRI-Education and
CRI-WorkingActivity were r=0.44; CRI-Education and
CRI-LeisureTime were r=0.30; CRI-WorkingActivity
and CRI-LeisureTime were r=0.32. The first was significantly higher than the other two (z=2.78, p<0.01 and
z=2.40, p<0.01, respectively, with the formula recommended by Cohen et al., 43).
Differences across Gender were minimal, except for the

K
e
c
i
r
LY
t
N
i
O
d
E
E
S
U
,
L
2
A
1
N
O
S
20
PER

Correlations
Results for Pearsons correlations and the Fisher test
are reported here, in order to show the degree of de-

Table 3 - Percentages of types of activities carried out during leisure time.

R
O
F

Reading newspapers and magazines

Housework (cooking, washing, ironing, etc.)
Driving (not biking)
Leisure activities (sports, hunting, dancing, cards, bowling, etc.)
Using new technologies (digital camera, computer, internet, etc.)
Social activities (parties/going out with friends,
local community events, etc.)
Cinema or theatre
Gardening, handicraft, knitting, etc.
Taking care of children or elderly
Volunteering
Artistic activities (playing an instrument, painting, writing, etc.)
Exhibitions, concerts, conferences
Holidays
Reading books
Pet care
Managing ones bank account(s)

1Following

Total
n=588

Women
n=323

Men
n=265

Young
n=246

Adults
n=212

Elderly
n=130

11%
12%
12%
4%
5%
8%

10%
18%
9%
4%
4%
7%

11%
5%
16%
6%
6%
9%

7%
9%
13%
4%
1%
11%

10%
11%
14%
4%
4%
8%

14%
16%
8%
5%
1%
6%

2%
9%
3%
2%
2%
4%
3%
7%
5%
12%

3%
10%
4%
2%
2%
3%
2%
8%
6%
10%

2%
7%
2%
1%
2%
4%
4%
6%
5%
14%

3%
3%
2%
2%
3%
4%
5%
8%
5%
11%

2%
8%
3%
1%
2%
4%
3%
8%
5%
12%

2%
13%
4%
1%
2%
2%
2%
7%
5%
12%

Fisher transformation, in a sample of 588 individuals, all Pearson's correlations greater than 0.107 were significantly different from 0, p<0.01.

Aging Clin Exp Res, Vol. 24, No. 3 222

M. Nucci, D. Mapelli and S. Mondini

60 80 100 120 140

60 80 100 120 140 160

CRI

140

0.77
140
120
100
80
60

0.78

0.72

0.44

0.30

100
80
60

CRI-Education

CRI-WorkingActivity

140

0.32
160
CRI-LeisureTime

100
80

s
i
t
ur

Fig. 2 - Upper right triangle: correlations between CRI, CRI-Education, CRI-WorkingActivity

and CRI-LeisureTime. Bottom
left triangle: relative scatter
plots. Histograms with estimated density of each variable
on diagonal.

K
e
c
i
r
LY
t
N
i
O
d
E
E
S
U
,
L
2
A
1
N
O
S
20
PER

120
80
60

60 80 100 120 140

80 100 120 140 160

correlation between CRI and CRI-LeisureTime: women

showed a higher correlation than men (rwomen=0.76;
rmen=0.68; z=2.85, p<0.01). As expected, the correlation
between CRI and CRI-Education was significantly lower
for the Elderly (relderly=0.76) with respect to the Young and
Adults (ryoung=0.81, radults=0.83; z=2.23, p<0.05 and
z=3.28, p<0.01, respectively). The correlation between
CRI and CRI-WorkingActivity was significantly the lowest for Young (ryoung=0.67, radult=0.80, relderly=0.74;
z=4.92, p<0.01 and z=2.38, p<0.01, respectively).
Adults also showed the significantly lowest correlation between CRI and CRI-LeisureTime (radults=0.62, ryoung=0.75,
relderly=0.75; z=4.24, p<0.01 for both).
The two intelligence tests used in this study (Vocabulary tests
of WAIS and TIB, number of errors) were only moderately correlated with the CRI (r=0.42 and r=-0.45, respectively).
Among the three sections of the CRIq, as expected, the
highest correlation with the two intelligence tests was with CRIEducation (r=0.44 and r=-0.43, respectively); CRIWorkingActivity (r=0.29 and r=-0.28) and CRI-LeisureTime (r=0.22 and r=-0.32) were slightly less correlated.

R
O
F

ANOVAs
Four ANOVA tests were performed, in order to verify
the effects of Age and Gender on the CRI and CRI subscores (Fig. 3, Table 4). A few post-hoc analyses were performed with the t-test. Type I error control was not provided, owing to the small number of tests.

223 Aging Clin Exp Res, Vol. 24, No. 3

Gender and Age significantly affected the CRI. Men

had a higher CRI than women (101.53 vs 98.75,
F(1,582)=5.56, p=0.02) and age also turned out to be a
significant factor (Young 97.57, Adults 106.22, Elderly
94.45, F(2,582)=32.08, p<0.01). Post-hoc analyses revealed the highest CRI in Adult (Young vs Adults t(456)=-7.27,
p<0.01; Young vs Elderly t(374)=2.13, p=0.03 and Adults
vs Elderly t(340)=6.32, p<0.01). No effect emerged from the
Age-Gender interaction (F(2,582)=0.98, p=0.37).
For the CRI-Education subscore, Gender was the only significant factor, men exceeding women (101.42 vs
98.82, F(1,582)=4.40, p=0.03) whereas Age groups and
Age-Gender interactions did not reach significance (Young
99.41, Adults 101.74 and Elderly 98.26, F(2,582)=2.02,
p=0.13 and F(2,582)=0.22, p=0.79 respectively).
ANOVA on the subscore CRI-WorkingActivity showed
Gender and Age as significant factors. Men exceeded
women (103.09 vs 97.46, F(1,582)=24.65, p<0.01) and
the three Age groups were also significant (Young 97.24,
Adults 107.00, Elderly 93.79, F(2,582)=41.54, p<0.01).
A post-hoc analysis revealed a significant difference in
CRI-WorkingActivity across the three Age groups, CRIWorkingActivity in the Adult group being highest (Young
vs Adults: t (456)=-8.82; p<0.01; Young vs Elderly:
t (374)=2.41, p=0.02; Adults vs Elderly: t(340)=6.85;
p<0.01). Age and Gender showed a significant interaction
(F(2,582)=7.84, p<0.01).
Lastly, Age was the only significant main effect on CRI-

Cognitive Reserve Index questionnaire (CRIq)

105
CRI-Education

105
100
CRI

95
90
85
Men Women

105
100
95
90
85
0

Men Women

95
90
85
0

Young Adults Elderly

CRI-LeisureTime

CRI-WorkingActivity

100

Men Women

Young Adults Elderly

105
100
95
90
85
0

Young Adults Elderly

Men Women

Young Adults Elderly

s
i
t
ur

Fig. 3 - Barplots of CRI and

subscores in Gender and Age
groups.

K
e
c
i
r
LY
t
N
i
O
d
E
E
S
U
,
L
2
A
1
N
O
S
20
PER

LeisureTime (Young 97.84, Adults 105.34, Elderly

95.36, F (2,582)=25.85, p<0.01). Post-hoc analyses
showed Adults had the highest CRI-LeisureTime with respect to Young and Elderly groups, which did not differ
(Young vs Adults: t(456)=7.04, p<0.01; Adults vs Elderly:
t(340)=5.06; p<0.01, Young vs Elderly: t(374)=1.57,
p=0.11). No significant difference was found in Gender
(men 98.93 vs women 100.87, F(2,582)=2.66, p=0.10).

DISCUSSION
This study proposes a new questionnaire, the CRIq, to
collect and quantify the amount of CR acquired during a
persons lifetime. In a single index, the CRIq conveys
three main sources of CR: education, working activity
and leisure time activities. Each of these aspects of an individuals lifetime is recorded as a subscore. The CRIq calculates each activity according to the number of years and
frequency of practice. Thus, the raw score increases

R
O
F

throughout an entire lifespan, as Stern stated CR is not

fixed; at any point in ones lifetime it results from a combination of exposures (2, p. 2017). However, by means
of a linear model, the three subscores and the final CRIq
score discards age, allowing comparisons across groups.
The data collected for 588 Italian participants indicated
that all three sections of the CRIq gathered distinct and
non-redundant information on individual lifestyles (correlations between subscores were not high and ranged between r=0.30 and r=0.44). The Italian population divided by age and gender showed several interpretable features. The lowest correlation, between CRI and CRI-Education, was found in the Elderly group. As already reported, elderly Italians generally only attended a few
years of school (particularly women) for historical and social reasons. Instead, the lowest correlation between CRI
and CRI-WorkingActivity was found in the Young group
probably because most young people were not yet work-

Table 4 - Means and standard deviations of CRI and sub-indices across gender and age.

TOTAL

Men
Total Men
Women
Total Women

Age

CRI

CRI-Education

CRI-WorkingActivity

CRI-LeisureTime

18-102

588

100 (15)

100 (15)

100 (15)

100 (15)

Young (18-44)
Adults (45-69)
Elderly (70-102)

118
107
40

98.63 (10.53)
106.19 (14.92)
97.60 (16.81)

101.09 (14.87)
102.71 (16.25)
98.99 (13.17)

97.88 (7.73)
108.81 (15.36)
103.15 (19.82)

97.92 (7.63)
102.48 (13.00)
92.42 (19.75)

20-90

265

101.52 (13.97)

101.43 (15.20)

103.09 (14.30)

98.093 (12.78)

Young (18-44)
Adults (45-69)
Elderly (70-102)

128
105
90

96.60 (8.95)
106.25 (15.96)
93.05 (19.36)

97.88 (12.59)
100.75 (16.81)
97.94 (14.98)

96.65 (6.93)
105.15 (15.38)
89.64 (18.64)

97.77 (8.34)
108.25 (15.00)
96.67 (23.12)

18-102

323

98.75 (15.71)

98.83 (14.75)

97.46 (15.10)

100.87 (16.57)

Aging Clin Exp Res, Vol. 24, No. 3 224

M. Nucci, D. Mapelli and S. Mondini

ing or had not reached top positions. Interestingly, the only correlation in which women overtook men was in
CRI-LeisureTime, in which several items (e.g., housework,
caring for grandchildren or elderly people) pertained
specifically to women. This balanced the fact that CRIWorkingActivity did not include housewife as a profession. Men had a higher CRI than women, because of differential involvement in work, low for women in the Elderly group. Instead Adults had the highest CRI, both CRIWorkingActivity and CRI-LeisureTime.
CR and intelligence are undoubtedly related, and their
measures are therefore correlated. Nevertheless, CR and
intelligence are distinct. Wechsler (44) defined intelligence as the aggregate or global capacity of the individual
to act purposefully, to think rationally, and to deal effectively with his environment. Sternberg (45) stated that intelligence is a goal-directed adaptive behavior. Thus, the
focus of intelligence is on acting and behaviour, in other
words, on intellectual performance. Instead, CR is a construct based on the idea of the storage of resources: the
potential cognitive capabilities acquired throughout life.
This clear-cut difference is reflected in their measurement. Whereas I.Q. is a measure of performance, CRI is
not. For this reason, we chose to exclude I.Q. as a CR
proxy, despite its undisputed correlation. We postulated
that CR and intelligence are two distinct constructs and
our results (correlation r=0.44) sustain our assumption.
Obviously, as expected, the highest correlation between
I.Q. and the CRI subscores of our sample was found
with CRI-Education.
The specific results obtained from our sample indicate
that CRIq is an efficient and reliable tool for measuring
CR. In particular, it is short and easy to administer,
and thus can easily be included in standard assessments
without too much effort for the subject in terms of both
time and cognitive resources. It also overcomes the
major limitations found in studies estimating CR. The
three proxies of the CRIq have already been used in other studies (23, 28), but they were usually combined in
couples or used with premorbid I.Q. (see Table 1). In a
recent paper from a broad systematic review of CR indicators, Jones et al. (46) confirmed education, occupation and leisure activities as the most frequently used
proxies of CR. The CRIq considers activities carried
out throughout adulthood (from the age of 18) (32,
28), whereas others gathered information exclusively
from current lifestyles or the past six months (22, 47).
The CRIq also takes into account not only the number of
years during which the activities were carried out, but also their frequency. Unlike other instruments which only measure activities late in life (28, 34), the CRIq can be
administered at any age. Lastly, since the measurement
of CR is not one of performance, a significant and interesting feature of CRIq is that it can also be administered to relatives or close friends, if an individual cannot

be interviewed for any reason (e.g., brain impairment,

aphasia, coma, dementia, etc.). Above all, the CRIq
provides a standardized and psychometrically controlled
measure of cognitive reserve, allowing its extensive employment in both experimental research and clinical
practice.
In fundamental research, the CRIq represents a single
index to compare data and results from different studies.
All investigations evaluating cognitive abilities could benefit from it in place of education only, in order to assess
individual performance better. The CRIq may also be
used to validate new psychometric tests (only age and education are usually considered).
In clinical settings, the CRIq could be used to diagnose
senile dementia, comparing CRI and cognitive performance. In subjects with high CRI scores, the richness of
neuronal synaptic tissue and its plasticity acts as a reserve,
compensating for the atrophy of grey matter and covering clinical pathological signs (2). Thus, poor performance on cognitive tests, but still within the normal
range, must be considered suspect in cases of a high CRI.
Conversely, poor performance combined with a low CRI
is expected. The CRIq could also enable psychologists to
gain a more complete image of their patients and their
lifestyles, including information from standard assessments which usually depend on the clinicians experience.
The Cognitive Reserve hypothesis assumes that the
fuller the life a person has had in terms of intellect, abilities and experiences, the more that person will be able to
cope with difficult cognitive tasks and social events in life.
The CRIq is the tool to quantify this cognitive, social, cultural and human capital.

s
i
t
ur

K
e
c
i
r
LY
t
N
i
O
d
E
E
S
U
,
L
2
A
1
N
O
S
20
PER
R
O
F

225 Aging Clin Exp Res, Vol. 24, No. 3

REFERENCES

1. Katzman R, Terry R, Deteresa R et al. Clinical, pathological, and

neurochemical changes in dementia: a subgroup with preserved
mental status and numerous neocortical plaques. Ann Neurol
1988; 23: 138-44.
2. Stern Y. Cognitive reserve. Neuropsychologia 2009; 47: 201528.
3. Keller JN. Age-related neuropathology, cognitive decline, and
Alzheimer's disease. Ageing Res Rev 2006; 5: 1-13.
4. Kramer AF, Bherer L, Colcombe SJ, Dong W, Greenough,
WT. Cognitive decline is related to education and occupation in
a Spanish elderly cohort. Aging Clin Exp Res 2002; 14: 132-42.
5. Stern Y. Cognitive reserve and Alzheimer disease. Alzheimer Dis
Assoc Disord 2006; 20: 69-74.
6. Valenzuela MJ, Sachdev P. Brain reserve and dementia: a systematic review. Psychol Med 2005; 35: 1-14.
7. Whalley LJ, Deary IJ, Appleton CL, Starr JM. Cognitive reserve and the neurobiology of cognitive aging. Ageing Res Rev
2004; 3: 369-82.
8. Kesler SR, Adams HF, Blasey CM, Bigler ED. Premorbid intellectual functioning, education, and brain size in traumatic brain injury: an investigation of the cognitive reserve hypothesis. Appl
Neuropsychol 2003; 10: 153-62.
9. Alexander GE, Furey ML, Grady CL, Pietrini P, Mentis MJ,

Cognitive Reserve Index questionnaire (CRIq)

10.

11.

12.

13.

14.

15.

Shapiro MB. Association of premorbid function with cerebral

metabolism in Alzheimers disease: implications for the reserve hypothesis. Am J Psychiatry 1997; 154: 165-72.
Bialystok E, Craik FI, Freedman M. Bilingualism as a protection
against the onset of symptoms of dementia. Neuropsychologia
2007; 45: 459-64.
Christensen H, Anstey KJ, Parslow R, Mackinnon A, Sachdev P.
The brain reserve hypothesis. Brain Atrophy and aging.
Gerontology 2006; 53: 82-95.
Daffner KR, Ryan KK, Williams DM et al. Age-related differences
in novelty and target processing among cognitively high performing adults. Neurobiol Aging 2005; 26: 1283-95.
Garibotto V, Borroni B, Kalbe E et al. Education and occupation
as proxies for reserve in aMCI converters and AD: FDG-PET evidence. Neurology 2008; 71: 1342-9.
Garrett DD, Grady CL, Hasher L. Everyday memory compensation: the impact of cognitive reserve, subjective memory, and
stress. Psychol Aging 2010; 25: 74-83.
Le Carret N, Auriacombe S, Letenneur L, Bergua LV, Dartigues
JF, Fabrigoule C. Influence of education on the pattern of cognitive deterioration in AD patients: the cognitive reserve hypothesis. Brain Cogn 2005; 57: 120-6.
Martino DJ, Strejilevich SA, Scpola M et al. Heterogeneity in
cognitive functioning among patients with bipolar disorder. J
Affect Disord 2008; 109: 149-56.
McDowell I, Xi G, Lindsay J, Tierney M. Mapping the connections
between education and dementia. J Clin Exp Neuropsychol
2007; 29: 127-41.
Ngandu T, Von Strauss E, Helkala EL et al. Education and dementia: what lies behind the association? Neurology 2007; 69:
1442-50.
Perneczky R, Drzezga A, Diehl-Schmid J et al. Schooling mediates brain reserve in Alzheimers disease: findings of fluoro-deoxyglucose-positron emission tomography. J Neurol Neurosurg
Psychiatry 2006; 77: 1060-3.
Ropacki SA, Bert AA, Ropacki MT, Rogers BL, Stern RA. The
influence of cognitive reserve on neuropsychological functioning
following coronary artery bypass grafting (CABG). Arch Clin
Neuropsychol 2007; 22: 73-85.
Roselli F, Tartaglione B, Federico F, Lepore V, Defazio G,
Livrea P. Rate of MMSE score change in Alzheimer's disease: influence of education and vascular risk factors. Clin Neurol
Neurosurg 2009; 111: 327-30.
Scarmeas N, Zarahn E, Anderson KE et al. Association of life activities with cerebral blood flow in Alzheimer Disease. Implications
for the cognitive reserve hypothesis. Arch Neurol 2003; 60:
359-65.
Sol-Padulls C, Bartrs-Faz D, Junqu C et al. Brain structure and
function related to cognitive reserve variables in normal aging, mild
cognitive impairment and Alzheimer's disease. Neurobiol Aging
2009; 30: 1114-24.
Spitznagel MB, Tremont G. Cognitive reserve and anosognosia
in questionable and mild dementia. Arch Clin Neuropsychol
2005; 20: 505-15.
Staff RT, Murray AD, Deary IJ, Whalley LJ. What provides
cerebral reserve? Brain 2004; 127: 1191-9.
Stern Y, Habeck C, Moeller J et al. Brain networks associated
with cognitive reserve in healtly young and old adults. Cerebr
Cortex 2005; 15: 394-402.
Tucker-Drob EM, Johnson KE, Jones RN. The cognitive reserve hypothesis: a longitudinal examination of age-associated de-

clines in reasoning and processing speed. Dev Psychol 2009; 45:

431-46.
28. Valenzuela MJ, Sachdev P. Assessment of complex mental activity
across the life span: development of the Lifetime of experiences
Questionnaire (LEQ). Psychol Med 2007; 37: 1015-25.
29. Gatz M, Svedberg P, Pedersen NL, Mortimer JA, Berg S,
Johansson B. Education and the risk of Alzheimer's disease:
findings from the study of dementia in Swedish twins. J Gerontol
B Psychol Sci Soc Sci 2001; 56: 292-300.
30. Stern Y, Alexander GE, Prohovnik I, Mayeux R. Inverse relationship between education and parietotemporal perfusion deficit
in Alzheimers disease. Ann Neurol 1992; 32: 371-5.
31. Stern Y, Alexander GE, Prohovnik I et al. Relationship between
lifetime occupation and parietal flow: implications for a reserve
against Alzheimers disease pathology. Neurology 1995; 45:
55-60.
32. Crowe M, Andel R, Pedersen NL, Johansson B, Gatz M. Does
participation in leisure activities lead to reduced risk of Alzheimer's
disease? A prospective study of Swedish twins. J Gerontol B
Psychol Sci Soc Sci 2003; 58: 249-55.

s
i
t
ur

33. Wilson RS, Bennett DA, Bienias JL et al. Cognitive activity and
incident AD in a population-based sample of older persons.
Neurology 2002; 59: 1910-4.

K
e
c
i
r
LY
t
N
i
O
d
E
E
S
U
,
L
2
A
1
N
O
S
20
PER

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.
26.

27.

R
O
F

34. Schinka JA, McBride A, Vanderploeg RD et al. Florida Cognitive

Activities Scale: intial development and validation. J Int
Neuropsychol Soc 2005; 11: 108-16.
35. Wilson, RS, Barnes, LL, Krueger, KR et al. Early and late life cognitive activity and cognitive systems in old age. J Int Neuropsychol
Soc 2005; 11: 400-7.

36. Scarmeas N, Zarahn E, Anderson KE et al. Cognitive reserve

modulates functional brain responses during memory tasks: a PET
study in healthy young and elderly subjects. Neuroimage 2003;
19: 1215-27.
37. Wechsler D. Wechsler Adult Intelligence Scale-Revised. New
York: Psychological Corp, 1981.
38. Nelson HE. National Adult Reading Test: Test Manual. Windsor,
Berks: NFER-Nelson, 1982.
39. Orsini A, Laicardi C. Wechsler Adult Intellingence Scale-RevisedWAIS-R, Contributo alla taratura italiana. Firenze: Organizzazioni
Speciali, Giunti, 1997.
40. Colombo L, Sartori G, Brivio C. Stima del quoziente intellettivo
tramite l'applicazione del TIB (test breve di Intelligenza). Giornale
Italiano di Psicologia 2002; 3: 613-37.
41. Rizopoulos D. ltm: An R package for latent variable modeling and
item response theory analyses. J Stat Software 2006; 17: 1-25.
42. Van Der Linden WJ, Hambleton RK. Handbook of modern
item response theory. New York: Springer-Verlag, 1997.
43. Cohen J, Cohen P, West SG, Aiken LS. Applied multiple regression/correlation analysis for the behavioural sciences, 3rd ed.
Mahwah, NJ: Lawrence Erlbaum, 2003.
44. Wechsler D. The Measurement of Adult Intelligence. Baltimore:
Williams & Wilkins, 1944.
45. Sternberg RJ. Handbook of Human Intelligence. New York:
Cambridge University Press, 1982.
46. Jones RN, Fong TG, Metzger E et al. Aging, brain disease, and
reserve: implications for delirium. Am J Geriatr Psychiatry 2010;
18: 117-27.
47. Hall CB, Lipton RB, Sliwinski M. Cognitive activities delay onset
of memory decline in persons who develop dementia. Neurology
2009; 73: 356-61.

Aging Clin Exp Res, Vol. 24, No. 3 226