Content uploaded by Fergus I M Craik
Author content
All content in this area was uploaded by Fergus I M Craik on Nov 18, 2017
Content may be subject to copyright.
Neuropsychologia 45 (2007) 459–464
Note
Bilingualism as a protection against the onset of symptoms of dementia
Ellen Bialystok a,b,∗, Fergus I.M. Craikb,c, Morris Freedman b,d,e
aDepartment of Psychology, York University, 4700 Keele Street, Toronto, Ontario, Canada M3J 1P3
bRotman Research Institute at Baycrest, Canada
cDepartment of Psychology, University of Toronto, Canada
dDivision of Neurology, Baycrest, Canada
eDepartment of Medicine, Division of Neurology, Mt. Sinai Hospital, University Health Network, and University of Toronto, Canada
Received 19 June 2006; received in revised form 23 October 2006; accepted 24 October 2006
Abstract
This study examined the effect of lifelong bilingualism on maintaining cognitivefunctioning and delaying the onset of symptoms of dementia in old
age. The sample was selected from the records of 228 patients referred to a Memory Clinic with cognitive complaints. The final sample consisted of
184 patients diagnosed with dementia, 51% of whom were bilingual. The bilinguals showed symptoms of dementia 4 years later than monolinguals,
all other measures being equivalent. Additionally, the rate of decline in Mini-Mental State Examination (MMSE) scores over the 4 years subsequent
to the diagnosis was the same for a subset of patients in the two groups, suggesting a shift in onset age with no change in rate of progression.
© 2006 Elsevier Ltd. All rights reserved.
Keywords: Bilingualism; Cognitive reserve; Dementia
Factors that may delay onset of dementia are of great impor-
tance given the social and economic burden of this disorder.
According to Brookmeyer, Gray and Kawas (1998), a 2-year
delay in onset of Alzheimer’s disease (AD) would reduce the
prevalence in the United States by 1.94 million after 50 years,
and delays as short as 6 months could have substantial public
health implications. Many of the factors predisposing a person
to dementia are biological (Corder et al., 1993), so the search for
methods to delay onset has focused largely on pharmacological
and other biologically-based therapies. There is growing evi-
dence, however, that some environmental factors can maintain
cognitive functioning in older adults and mitigate the effects of
illnesses that produce dementia. Notably, research on “cognitive
reserve” has demonstrated that lifestyle factors, such as physical
activity, stimulating leisure involvement, and social engagement
play a role in postponing the onset of AD and other demen-
tias (Alexander et al., 1997;Scarmeas, Levy, Tang, Manly, &
Stern, 2001;Scarmeas & Stern, 2003;Singh-Manoux et al.,
2003; Stern, 2002).
The literature on brain reserve is somewhat contentious, but
several recent large-scale reviews have provided a context for
∗Corresponding author. Tel.: +1 416 736 5115x66109; fax: +1 416 736 5814.
E-mail address: ellenb@yorku.ca (E. Bialystok).
conflicting results. The overall notion is that some factors allow
a person to function within a normal cognitive range, despite
the presence of brain pathology that would usually be associ-
ated with dementia. For example, Mortimer (1997) found that
between 10 and 40% of autopsy cases showing brain pathology
exceeding the criteria for AD had shown no signs of cognitive
impairment before death. Similarly a population-based study
carried out by the UK Medical Research Council (quoted by
Valenzuela & Sachdev, 2006a) found that more than 30% of
individuals with mild and severe AD pathology at autopsy had
shown no previous signs of cognitive impairment. Valenzuela
and Sachdev (2006a) distinguish between, ‘neurological brain
reserve’ and ‘behavioral brain reserve.’ Proponents of neurologi-
cal brain reserve argue that peak brain volume can ameliorate the
effects of brain pathology on cognitive performance and signs of
dementia. This type of brain reserve is thus presumably biologi-
cal and possibly genetic in origin. On the other hand, behavioral
brain reserve (also referred to as cognitive reserve, the term used
in the present report) suggests that sustained complex mental
activity protects against dementia in terms of both incidence
(Valenzuela & Sachdev, 2006a) and the rate of cognitive decline
in elderly individuals (Valenzuela & Sachdev, 2006b).
In their review of behavioral brain reserve, Valenzuela and
Sachdev (2006a) found strong evidence for protection against
0028-3932/$ – see front matter © 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.neuropsychologia.2006.10.009
460 E. Bialystok et al. / Neuropsychologia 45 (2007) 459–464
dementia provided by education, high occupational status,
high levels of premorbid intelligence, and mentally stimulat-
ing leisure activities. Importantly, most of the studies included
in the review found significant effects of the protective vari-
able in question after co-varying out age and other brain-reserve
measures. One surprising conclusion of the review was that “it
is evident that mentally stimulating leisure activity is the most
robust brain-reserve measure, since all these studies showed a
significant protective effect even after controlling for age, educa-
tion, occupation and other potential confounds” (Valenzuela &
Sachdev, 2006a, p. 447). The authors found an overall decrease
in incident dementia of 46% after a median follow-up interval
of 7.1 years; these figures are based on a total of over 29,000
individuals from 22 studies. It is important to note the authors’
caution that higher levels of behavioral brain reserve may simply
delay the onset of dementia, rather than reduce incidence in an
absolute manner. In a comparison piece, the same authors con-
ducted a review of studies of longitudinal cognitive change and
factors ameliorating cognitive decline in the elderly (Valenzuela
& Sachdev, 2006b). This second review was based on a new
sample of 18 studies involving more that 47,000 individuals; the
main finding was that higher levels of behavioral brain reserve
were related to decreased rates of cognitive decline. The contrib-
utors to behavioral brain reserve were again higher levels of edu-
cation, occupation, and stimulating leisure and social activities.
A review of longitudinal studies by Fratiglioni, Paillard-Borg,
and Winblad (2004) also found strong evidence for the role of
social, mental, and physical activities in protecting against all
types of dementia. Similarly, Staff, Murray, Deary, and Whalley
(2004) concluded that “more education and a more cognitively
complex occupation predict higher cognitive ability in old age
than would be expected for a person’s childhood ability and
accumulated brain burden” (p. 1196). From these reviews, it
seems clear that complex mental activity across the lifespan acts
to at least delay the incidence of dementia. Two sets of questions
follow from this conclusion; first, what types of mental activity
yield this protective function and do they have features in com-
mon? Second, what is the mechanism linking mental activity to
its neuroprotective function?
The present report contributes to the first question by demon-
strating a further type of mental activity that may be associated
with a delay in the appearance of the symptoms of dementia.
The activity in question is the constant use of two languages
over many years. Bilingualism has been shown to enhance atten-
tion and cognitive control in both children (Bialystok, 2001)
and older adults (Bialystok, Craik, Klein, & Viswanathan, 2004;
Bialystok, Craik, & Ryan, 2006). In these studies, lifelong bilin-
guals who use both languages in their daily lives showed an
advantage in a variety of tasks involving attentional control.
The suggested interpretation is that the use of two languages
requires a mechanism to control attention to the relevant lan-
guage and ignore or inhibit interference from the competing
language (Green, 1998). This experience provides continual
practice in attentional control and results in its earlier devel-
opment in children, improved functioning in adults, and slower
decline in older age. Therefore, bilingualism might contribute
to cognitive reserve and protect older adults from decline in
the context of dementia. This hypothesis was examined in the
present study.
Cognitive reserve is considered to provide a general pro-
tective function, possibly due to enhanced neural plasticity,
compensatory use of alternative brain regions, or enriched brain
vasculature (Fratiglioni et al., 2004). Since cognitive reserve
occurs independently of any specific conditions, such as AD,
our study sample included all cases of dementia regardless of
diagnosis.
1. Method
We examined the records of consecutive 228 patients who were referred to
the Memory Clinic at Baycrest in Toronto, Canada, between 2002 and 2005 with
cognitive complaints. In addition to a medical history, physical examination, and
mental status evaluation, patients were usually assessed with CT, SPECT, and
screening blood tests. There were two exclusion criteria that reduced this sample
by 44 patients, leaving a final sample of 184. First, 23 patients (12 monolin-
guals and 11 bilinguals) received a diagnosis that did not include dementia (e.g.,
depression, mild cognitive impairment), so were not considered further. Second,
21 patients could not be clearly classified as monolingual or bilingual (see below)
and were therefore excluded as well. Of the remaining 184 patients, 132 patients
met criteria for probable AD by consensus among a group of medically qualified
Clinic staff, including at least one neurologist, using NINCDS–ADRDA criteria
(McKhann et al., 1984). A further 52 patients were diagnosed with other demen-
tias, including possible AD, dementia due to other neurodegenerative disorders,
and cerebrovascular disease.
The age of onset of cognitive impairment was determined by the interview-
ing neurologist at the first clinic visit who asked patients and their families or
caregivers when symptoms were first noticed. Although this approach involves
a subjective estimate by patients and their families, it was applied equally to
all cases. We see no reason to expect that monolinguals and bilinguals or their
families should differ systematically in this judgment, and therefore no reason to
expect systematic bias in the clinical information that was recorded. Moreover,
the histories were taken prior to any knowledge of the current study.
The files contained the following information about language history: lan-
guages spoken, English fluency, place of birth, date of birth, and year of immi-
gration to Canada. This information, without any other details, was given to 11
judges experienced in conducting behavioral research with bilinguals who classi-
fied each patient as monolingual or bilingual. The criterion for bilingualism was
that patients had spent the majority of their lives, at least from early adulthood,
regularly using at least two languages. The judges did not reach a consensus for
21 patients, so these were eliminated from further analyses. Inter-rater reliabil-
ity was .95 (S.D. = 0.04) for designating an individual as monolingual and .81
(S.D. =0.08) for designating as bilingual. Immigration occurred predominantly
in the 1940s (n= 14), 1950s (n= 25), and 1960s (n= 17). The bilinguals included
speakers of 25 different first languages, of which the most common examples
were Polish (n= 20), Yiddish (n=13), German (n=12), Romanian (n= 8), and
Hungarian (n= 7). Many of these individuals were bilingual prior to arriving
in Canada. The data also included scores from Mini-Mental State Examination
(MMSE) (Folstein, Folstein, & McHugh, 1975) at the initial appointment, years
of education, and occupation.
To summarize, the final sample consisted of 184 patients, of whom 91 were
monolingual and 93 were bilingual. There were 66 patients in each language
group diagnosed with probable AD, comprising 73% of monolinguals and 71%
of the bilinguals. AD was sometimes accompanied by other conditions, such as
cerebrovascular disease, depression, psychosis, meningioma, and sleep apnea.
2. Results
The mean values for the relevant variables are reported in
Table 1. The difference between monolinguals and bilinguals
of 4.1 years in age of onset of symptoms of dementia is sig-
nificant, F(1,178) = 9.16, p< 0.003, with no difference between
E. Bialystok et al. / Neuropsychologia 45 (2007) 459–464 461
Table 1
Mean value (and standard deviation) for descriptors for each language group
Language group NAge of onset Age at first appointment Years of education MMSE at first appointment Occupation status
Monolingual 91 71.4(9.6) 75.4(9.3) 12.4(3.8) 21.3(6.4) 3.3(1.5)
Men 43 70.8 (9.5) 76.2 (9.1) 12.9 (4.8) 20.5 (6.8) 3.6 (1.5)
Women 48 71.9 (9.8) 74.7 (9.5) 11.9 (2.8) 22.0 (5.9) 3.0 (1.4)
Bilingual 93 75.5(8.5) 78.6(8.4) 10.8(4.2) 20.1(7.1) 3.0(1.6)
Men 38 76.1 (5.9) 79.4 (6.3) 10.8 (4.8) 20.7 (7.5) 3.0 (1.7)
Women 55 75.1 (9.9) 78.1 (9.6) 10.7 (3.7) 19.6 (6.8) 3.1 (1.4)
men and women, F< 1. The power of this effect with α= 0.05
is 0.87. In addition, bilinguals were 3.2 years older than mono-
linguals at the time of the initial clinic appointment, a differ-
ence that was also significant, F(1,180) = 5.93, p< 0.02, with no
difference between men and women, F< 1. We also analyzed
these data separately for diagnoses of AD and other demen-
tias. The delayed onset of symptoms for bilinguals was signif-
icant both in the subsample of 132 patients with probable AD,
F(1,128) = 7.07, p< 0.009, with a delay of 4.3 years, and for
the other dementias, F(1,47) = 3.81, p< 0.04, with a delay of
3.5 years.
The scores out of 30 on the MMSE, administered at the ini-
tial clinic visit, showed no difference between patients in the
two language groups, F(1,161) = 1.29, n.s., and no difference
between men and women, F< 1. We nonetheless carried out an
analysis of covariance on the age of onset data, using gender as
the covariate. This analysis yielded adjusted scores for age of
onset of 71.4 years for monolinguals and 75.5 years for bilin-
guals; this difference was significant, F(1,179) = 9.13, p< 0.003.
Research on cognitive reserve has pointed to formal educa-
tion as a protective factor against the onset of dementia (Bennett
et al., 2003; Staff et al., 2004). A comparison of the years
of schooling for the participants in the two groups (Table 1)
indicates that the bilinguals had significantly fewer years of edu-
cation, F(1,171) = 7.06, p< 0.009, with no difference by gender,
F< 1. An analysis of covariance on age of onset data using years
of education as the covariate yielded adjusted age of onset scores
of 71.7 years for monolinguals and 75.2 years for bilinguals; this
difference was significant, F(1,170) = 6.18, p< 0.02. The lower
value for years of schooling for the bilingual group may reflect
differences in opportunity more than ability, given that many
individuals in that group came from Europe and their lives were
disrupted by World War II.
The bilinguals varied widely in their cultural experiences, but
we considered whether there might be a general bias in which
bilinguals avoided seeking medical attention longer than mono-
linguals. However, Table 1 shows that the interval between onset
of symptoms and the first clinic visit is actually shorter (3.0
years) for the bilinguals than for the monolinguals (3.8 years), a
difference that was almost significant, F(1,178) = 3.69, p= 0.06.
There was a main effect of gender, F(1,178) = 8.84, p< 0.003,
in which men postponed clinic visits longer than women, and
an interaction of language group and gender, F(1,178) = 5.35,
p< 0.02, because monolingual males were particularly slow in
seeking medical attention.
On a related point, most of the bilinguals (81/93) were immi-
grants to Canada, whereas most of the monolinguals (78/91)
were not, and it is possible that immigrants delayed visiting
the clinic. We conducted a two-way ANOVA for language
group and immigration status on the interval between onset
of symptoms and time of appointment and found no effect
of either language group, F(1,177) = 3.33, p= 0.07, or immi-
gration status, F(1,177) = 1.85, n.s. A further analysis exam-
ined the age of onset of symptoms of dementia for only
those individuals in the two language groups who were immi-
grants (monolinguals = 13, bilinguals = 81). The age of onset
for monolinguals was 63.8 years and for bilinguals it was 75.3
years, a difference that was highly significant, F(1,92) = 17.96,
p< 0.0001.
Higher occupation status and more intellectually stimulating
work is associated with retained cognitive function in old age
(Staff et al., 2004) and reduced effects of dementia (Seidler et al.,
2004). We classified the occupations of the patients using the sys-
tem developed by Human Resources and Skills Development,
Canada (2001). Occupations are classified on a five-point scale,
with higher numbers associated with higher status. The scale
reflects the placement of occupations in terms of the two dimen-
sions of skill level and social class associated with each. Women
who did not work outside the home and patients for whom no
occupation was listed in the records were not included in this
analysis. For this reason, 37 patients (13 monolingual and 24
bilingual) were excluded from the analysis of occupational sta-
tus. The mean rankings for occupational status for the remaining
147 patients, reported in Table 1, did not differ between the lan-
guage groups, F< 1. An analysis of covariance on age of onset
data using occupational status as the covariate yielded adjusted
age of onset scores of 71.3 years for monolinguals and 74.2 years
for bilinguals. This difference was significant, F(1,142) = 3.93,
p< 0.05, with bilinguals continuing to show a later age of onset
of symptoms.
Finally, there is evidence that patients with higher education
catch up with their less-educated peers by exhibiting faster rates
of cognitive decline in the 5 years after diagnosis (Scarmeas,
Albert, Manly & Stern, 2006). Therefore, we examined the
progression of the dementia after the initial appointment for
the 25 monolinguals and 24 bilinguals who received further
MMSE tests over the next 4 years. Fig. 1 indicates no group
difference in rate of decline. A regression analysis showed that
the year of testing subsequent to diagnosis predicted MMSE
score, t(1) = −3.74, p< 0.003, and that the rate of decline for
462 E. Bialystok et al. / Neuropsychologia 45 (2007) 459–464
Fig. 1. Slope of decline in MMSE scores for subset of 25 monolinguals and 24
bilinguals across 4 years, indicating equivalent rates of decline for both groups.
the two language groups was equivalent in that there was
no interaction between language group and slope, t(1) = 1.10,
p= 0.28.
3. Discussion
The bilingual patients in our sample exhibited a delay of 4.1
years in the onset of symptoms of dementia in comparison to
monolinguals. The implications of such a delay are substan-
tial. From the meta-analysis based on 16 European and North
American studies presented by Hy and Keller (2000), we calcu-
lated that a delay of 4 years at the age of our patients translates
into a reduction of prevalence of 47%. Interestingly, in their
review, Valenzuela and Sachdev (2006a) report a very similar
figure for the reduction in incidence of dementia associated with
stimulating mental activities (46%). There are currently no phar-
macological interventions that have shown comparable effects.
The interpretation that bilingualism delays the onset of
dementia depends on evidence that the monolingual and bilin-
gual groups do not differ apart from their language abilities.
The primary data relevant to this point is that the MMSE scores
were equivalent for the patients in the two groups at the ini-
tial visit to the Memory Clinic, indicating comparable levels
of impairment. In addition, we considered cultural differences,
immigration, formal education, and employment status. None
of these could account for the results. Analyses of covariance
on the age of onset data were conducted, using gender, years
of education, and employment status as the covariate, with the
result that age of onset remained significantly different between
the language groups in all cases. Nonetheless, we acknowledge
that the present data were gathered retrospectively from clinical
records, and that there is necessarily some degree of subjectivity
involved in the estimate of age of onset of dementia. There is
also some degree of subjectivity on the part of patients and their
families in terms of deciding when it is necessary to seek help
for the condition, potentially delaying the first visit and increas-
ing the age at which the diagnosis is made. Although we see no
reason for a systematic bias between the groups in this respect,
the present report should be treated as suggestive rather than as
definitive.
The protective effect of bilingualism found in the present
study cannot be generalized to individuals who have some
knowledge of another language but are not fully bilingual. All
the patients classified as bilingual in the present study were fluent
in English as well as another language and have used both lan-
guages regularly for most of their lives. A study by LoGiudice,
Hassett, Cook, Flicker, and Ames (2001) showed that patients
living in Australia who spoke another language and had poor
English skills arrived at a memory clinic at a later stage in the
disease than did the native English-speaking patients. In con-
trast, the patients in the present study had fluent English skills
and arrived at the memory clinic at the same level of cognitive
impairment as the English-speaking monolinguals.
Dementia can be affected by a variety of factors, includ-
ing genetic, neurobiological, environmental, intellectual, and
lifestyle. The finding that bilingualism delays the onset of
dementia by 4 years in our sample falls into the last category.
These results demonstrate how a psychological factor can affect
a biologically-based disease state. Given the findings on behav-
ioral brain reserve reviewed by Valenzuela and Sachdev (2006a,
2006b) the present findings are not entirely surprising. Some
caution, however, is needed in interpreting the results. First,
the finding is correlational and not the result of an experimen-
tal design with random assignment to groups. Such a design is
impossible in research of this type; people are monolingual or
bilingual for reasons that have nothing to do with our exper-
imental inquiry. However, unlike the correlation between AD
and reduced activity levels in midlife (Friedland et al., 2001)
where causality could operate in either direction, it is quite
unlikely that the early effects of AD predispose a person to be
monolingual, especially since most of these patients became
bilingual at an early age. Second, the measure of “age of onset”
is somewhat subjective, relying on the report of the patient, fam-
ily members, and caregivers. However, the measure and means
of obtaining the data were equivalent for all the cases so we
see no reason to expect systematic bias. Having considered and
rejected some obvious confounding factors, the data are strongly
in line with our hypothesis that bilingualism delays the onset of
symptoms of dementia. Third, the evidence on post-diagnosis
decline (Fig. 1) is fragmentary, but the evidence suggests that
the time between initial appearance of symptoms and the first
clinical appointment is the same for the two groups, as is the
subsequent rate of cognitive decline. This pattern suggests that
bilingualism simply shifts the function relating level of impair-
ment to passage of time by 4.1 years, a pattern also reported
by Salthouse (2006) for the beneficial effects of stimulating
activities on normal age-related cognitive decline. In contrast
to this result, the second review by Valenzuela and Sachdev
(2006b) reports results in favor of the notion that behavioral
brain reserve slows the rate of cognitive decline in older people.
However, another recent report (Scarmeas et al., 2006) found that
whereas higher levels of education delay the onset of dementia,
the rate of decline is actually faster post-diagnosis in the highly
educated group. Further elucidation of this point is obviously
needed.
E. Bialystok et al. / Neuropsychologia 45 (2007) 459–464 463
The present results identifying an experience that may help
to maintain cognitive functioning in older age are impor-
tant because of their implications for research showing that
such cognition-enhancing experiences affect both structural vol-
ume and functional organization of the central nervous system
(Draganski et al., 2004; Karni et al., 1995; Maguire et al., 2000;
Steven & Blakemore, 2004). However, unlike most of the stud-
ies in which plasticity traced to such experiences is specific
to particular skills, the present findings suggest that extensive
experience of one kind can have widespread effects on general
cognitive functioning. The speculative conclusion (following
Fratiglioni et al., 2004;Scarmeas & Stern, 2003;Staff et al.,
2004;Valenzuela & Sachdev, 2006a, 2006b) is that bilingualism
does not affect the accumulation of pathological factors associ-
ated with dementia, but rather enables the brain to better tolerate
the accumulated pathologies. The mechanisms underlying such
protective effects are not yet clear, and the present results yield
no direct evidence on such mechanisms, but plausible candidates
listed by Valenzuela and Sachdev (2006a) include increases in
resting phosphocreatine levels, increased generation of neurons,
synapses and arborized dendrites, and functional reorganization
of brain networks. In general, it is increasingly clear that biologi-
cal factors interact with environmental experiences to determine
cognitive outcomes; the present findings suggest that bilingual-
ism is one experiential factor that can provide a positive benefit
in this respect.
Acknowledgements
This work was supported by a grant from the Canadian Insti-
tutes of Health Research to E.B. and F.C., and by CIHR group
grant # MGC 14974. M.F. is supported by the Saul A. Silver-
man Family Foundation, Toronto, Canada, as part of a Canada
International Scientific Exchange Program (CISEPO) project.
We thank Sharyn Kreuger, Johanna Lake and Bernard Kadosh
for their assistance in compiling data from the files; Gigi Luk
and Mythili Viswanathan for their assistance with data analyses.
We are grateful to Endel Tulving and Donald Stuss for helpful
comments on the manuscript.
References
Alexander, G. E., Furey, M., Grady, C., Pietrini, P., Brady, D. R., Mentis, M. J.,
et al. (1997). Association of premorbid intellectual function with cerebral
metabolism in Alzheimer’s disease: Implications for the cognitive reserve
hypothesis. American Journal of Psychiatry,154, 165–172.
Bennett, D. A., Wilson, R. S., Schneider, J. A., Evans, D. A., Mendes de Leon,
C. F., Arnold, S. E., et al. (2003). Education modifies the relation of AD
pathology to level of cognitive function in older persons. Neurology,60,
1909–1915.
Bialystok, E. (2001). Bilingualism in development: Language, literacy, and cog-
nition. New York: Cambridge University Press.
Bialystok, E., Craik, F.I. M., Klein, R., & Viswanathan, M. (2004). Bilingualism,
aging, and cognitive control: Evidence from the Simon task. Psychology and
Aging,19, 290–303.
Bialystok, E., Craik, F. I. M., & Ryan, J. (2006). Executive control in a
modified anti-saccade task: Effects of aging and bilingualism. Journal of
Experimental Psychology: Learning, Memory, and Cognition,32, 1341–
1354.
Brookmeyer, R., Gray, S., & Kawas, C. (1998). Projections of Alzheimer’s
disease in the United States and the public health impact of delay-
ing disease onset. American Journal of Public Health,88, 1337–
1342.
Corder, E. H., Saunders, A. M., Strittmatter, W. J., Schmechel, D. E., Gaskell,
P. C., Small, G. W., et al. (1993). Gene dose of apoliproprotein E type 4
allele and the risk of Alzheimer’s disease in late onset families. Science,
261, 921–923.
Draganski, B., Gaser, C., Busch, V., Schuierer, G., Bogdahn, U., & May,
A. (2004). Changes in grey matter induced by training. Nature,427,
311–312.
Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). “Mini-mental state”. A
practical method for grading the cognitive state of patients for the clinician.
Journal of Psychiatric Research,12, 189–198.
Fratiglioni, L., Paillard-Borg, S., & Winblad, B. (2004). An active and socially
integrated lifestyle in late life might protect against dementia. Lancet Neu-
rology,3, 343–353.
Friedland, R. P., Fritsch, T., Smyth, K. A., Koss, E., Lerner, A. J., Chen, C.
H., et al. (2001). Patients with Alzheimer’s disease have reduced activi-
ties in midlife compared with healthy control-group members. Proceedings
of the National Academy of Sciences of the United States of America,98,
3440–3445.
Green, D. W. (1998). Mental control of the bilingual lexico-semantic system.
Bilingualism: Language and Cognition,1, 67–81.
Human Resources and Skills Development, Canada (2001). National Occu-
pational Classification. Retrieved May 13, 2006. from http://www23.hrdc-
drhc.gc.ca/2001/e/groups/index.shtml.
Hy, L. Y., & Keller, D. M. (2000). Prevalence of AD among whites: A summary
by levels of severity. Neurology,55, 198–204.
Karni, A., Meyer, G., Jezzard, P., Adams, M. M., Turner, R., & Ungerleider, L.
G. (1995). Functional MRI evidence for adult motor cortex plasticity during
motor skill learning. Nature,377, 155–158.
LoGiudice, D., Hassett, A., Cook, R., Flicker, L., & Ames, D. (2001). Equity
of access to a memory clinic in Melbourne? Non-English speaking back-
ground attenders are more severely demented and have increased rates of
psychiatric disorders. International Journal of Geriatric Psychiatry,16,
327–334.
Maguire, E. A., Gadian, D. G., Johnsrude, I. S., Good, C. D., Ash-
burner, J., Frackowiak, R. S. J., et al. (2000). Navigation-related struc-
tural changes in the hippocampi of taxi drivers. Proceedings of the
National Academy of Sciences of the United States of America,97, 4398–
4403.
McKhann, G., Drachman, D. A., Folstein, M., Katzman, R., Price, D., & Stad-
lan, E. M. (1984). Clinical diagnosis of Alzheimer’s disease: Report of the
NINCDS-ADRDA work group under the auspices of Department of Health
and Human Services Task Force on Alzheimer’s disease. Neurology,34,
939–944.
Mortimer, J. A. (1997). Brain reserve and the clinical expression of Alzheimer’s
disease. Geriatrics,52, S50–S53.
Salthouse, T. A. (2006). Mental exercise and mental aging: Evaluating the
validity of the “use it or lose it” hypothesis. Perspectives on Psychologi-
cal Science,1, 68–87.
Scarmeas, N., Albert, S. M., Manly, J. J., & Stern, Y. J. (2006). Education
and rates of cognitive decline in incident Alzheimer’s disease. Journal of
Neurology, Neurosurgery and Psychiatry,77, 308–316.
Scarmeas, N., Levy, G., Tang, M., Manly, J., & Stern, Y. (2001). Influence
of leisure activity on the incidence of Alzheimer’s disease. Neurology,57,
2236–2242.
Scarmeas, N., & Stern, Y. (2003). Cognitive reserve and lifestyle. Journal of
Clinical and Experimental Neuropsychology,25, 625–633.
Seidler, A., Nienhaus, A., Bernhardt, T., Kauppinen, T., Elo, A.-L., & Fr¨
olich,
L. (2004). Psychosocial work factors and dementia. Occupational and Envi-
ronmental Medicine,61, 962–971.
Singh-Manoux, A., Richards, M., & Marmot, M. (2003). Leisure activities and
cognitive function in middle age: Evidence from the Whitehall II study.
Journal of Epidemiology and Community Health,57, 907–913.
Staff, R. T., Murray, A. D., Deary, I. J., & Whalley, L. J. (2004). What provides
cerebral reserve? Brain,27, 1191–1199.
464 E. Bialystok et al. / Neuropsychologia 45 (2007) 459–464
Stern, Y. (2002). What is cognitive reserve? Theory and research application of
the reserve concept. Journal of the International Neuropsychological Soci-
ety,8, 448–460.
Steven, M. S., & Blakemore, C. (2004). Cortical plasticity in the adult human
brain. In M. S. Gazzaniga (Ed.), The cognitive neurosciences III (pp.
1243–1254). Cambridge: MIT Press.
Valenzuela, M. J., & Sachdev, P. (2006a). Brain reserve and dementia: A sys-
tematic review. Psychological Medicine,36, 441–454.
Valenzuela, M. J., & Sachdev, P. (2006b). Brain reserve and cognitive decline:
A non-parametric systematic review. Psychological Medicine,36, 1065–
1073.