Wilbur, J., Marquez, D. X., Fogg, L., Wilson, R. S., Staffileno, B. A., Hoyem, R. L., Morris, M. C.

, Bustamante, E. E., Manning, A. F. (2012). The relationship between physical activity and cognition in
Staffileno,
older Latinos. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 10.1093/geronb/gbr137
67(5), 525–534, doi:10.1093/geronb/gbr137. Advance Access published on February 9, 2012

The Relationship Between Physical Activity and

Cognition in Older Latinos
JoEllen Wilbur,1 David X. Marquez,2 Louis Fogg,3 Robert S. Wilson,4,5 Beth A. Staffileno,6
Ruby L. Hoyem,1 Martha Clare Morris,7 Eduardo E. Bustamante,2 and Alexis F. Manning,1

1Women, Children, and Family Nursing, College of Nursing, Rush University, Chicago, Illinois.

Downloaded from https://academic.oup.com/psychsocgerontology/article-abstract/67/5/525/657933 by guest on 05 May 2019

2Department of Kinesiology and Nutrition College of Applied Health Sciences, University of Illinois at Chicago.
3Community, Systems and Mental Health Nursing, College of Nursing,
4Department of Neurological Sciences, 5Department of Behavioral Sciences,
6Adult Health and Gerontological Nursing, College of Nursing and
7Section of Nutrition and Nutritional Epidemiology, Department of Internal Medicine,

Rush University, Chicago, Illinois.

Objectives. The purpose of this study was to explore the relationship between minutes spent participating in light and
moderate/vigorous-intensity physical activity (PA) and cognition in older Latinos, controlling for demographics, chronic
health problems, and acculturation.

Method. A cross-sectional study design was used. Participants were self-identified Latinos, without disability, who
had a score less than 14 on a 21-point Mini-Mental State Examination. Participants were recruited from predominantly
Latino communities in Chicago at health fairs, senior centers, and community centers. PA was measured with an acceler-
ometer, worn for 7 days. Episodic memory and executive function (inference control, inattention, and word fluency) were
measured with validated cognitive tests.

Results. Participants were 174 Latino men (n = 46) and women (n = 128) aged 50–84 years (M = 66 years). After
adjusting for control variables (demographics, chronic health problems) and other cognitive measures, regression analy-
ses revealed that minutes per day of light-intensity PA (r = −.51), moderate/vigorous PA (r = −.56), and counts per minute
(r=−.62) were negatively associated with lower word fluency.

Discussion. Findings suggest that the cognitive benefits of both light-intensity PA and moderate/vigorous PA may be
domain-specific.

Key Words: Cognition—Latino/Hispanic—PA.

Background Physical activity (PA) is thought to be one means of

The number of older adults in the United States is rapidly attenuating cognitive decline and improving cognitive ability
growing, and it is expected to grow fastest among Latinos. among older adults (Colcombe & Kramer, 2003). Research
Between 2008 and 2030, the Latino population 65 years and providing support for the beneficial effects of PA on cogni-
older is expected to increase by 224% compared with just a tive function in older adults spans several decades but has
65% increase of the non-Latino White population 65 years recently been advanced by findings from observational
and older (Greenberg, 2009). Unfortunately, data from the prospective epidemiological studies (Hillman, Erickson, &
Health and Retirement Study, a large nationally representa- Kramer, 2008). Hamer and Chida (2009) identified 10 pro-
tive study of the 55 and older population, indicate that older spective cohort studies with healthy participants at baseline
Latinos are twice as likely as non-Latino Whites to exhibit that examined PA and cognitive decline. In nine of the 10
cognitive impairment with age (17.5% vs. 8.8%; Alzheimer’s studies, an inverse relationship was found between PA and
Association, 2010). Moreover, there is epidemiological cognitive decline. None of the studies, however, indicated
evidence linking age-associated cognitive impairment to that Latinos were included or mentioned translation of
diabetes (Kodl & Seaquist, 2008), for which Latinos in the instruments. The two studies that included race as a covari-
United States are more burdened compared with the domi- ate simply identified White versus non-White populations
nant culture (non-Latino White men 6.7% vs. Latino men (Albert et al., 1995; Sturman et al., 2005). Therefore, to
11.0%; non-Latino White women 5.6% vs. Latina women our knowledge, a majority of what is known about the PA–
10.9%; Roger et al., 2011). Given the costs associated with cognition relationship has been derived from the non-
care of the older people, it is imperative that we obtain a Latino dominant culture, and little is known about the possible
better understanding of modifiable health behaviors that can influence of acculturation on cognition in addition to the
prevent or stabilize cognitive impairment through the middle known influence of age and education (Plassman et al.,
and older years of life of our growing Latino population. 2007).
©
©The
All
TheAuthor
Author 2012.
All rights
2012. Published
rights reserved.
Published by
reserved. For
by Oxford
For permissions,
Oxford University
permissions, please
University Press
please e-mail:
Press on
on behalf
behalf of
ofThe
The Gerontological
Gerontological Society
e-mail: journals.permissions@oup.com.
journals.permissions@oup.com.
Society of
ofAmerica.
America. 5251
Received
Received January
January 21,
21, 2011;
2011;Accepted
Accepted November
November 19,
19, 2011
2011
Decision
Decision Editor:
Editor: Robert
RobertWest,
West, PhD
PhD
2526 WILBUR
WILBUR
Closer scrutiny of these 10 studies revealed selected
measurement issues that may have particular bearing for
Latinos. First, all relied on self-report measures that focused
heavily on leisure-time or structured PA obtained through
exercise and sport. According to 2009 National Health In-
terview Survey, however, 44% of adult Latinos versus 28%
of non-Latino White adults reported no light, moderate, or
vigorous leisure PA (Roger et al., 2011). Furthermore, data
from Latinos indicate that the prevalence of leisure-time PA
is lowest among immigrants, and, although it increases over
time since immigration, it remains lower than in nonimmi-
grants (Tremblay, Bryan, Peréz, Ardern, & Katzmarzyk,
2006). Conversely, the Latino population has been shown to
engage in a high level of nonleisure-time lifestyle PAs that
are a part of everyday life (Berrigan, Troiano, McNeel,
DiSogra, & Ballard-Barbash, 2006). Unfortunately, the few
studies that included lifestyle PA were restricted to a single
item that queried participation in yard or garden work
(Albert et al., 1995; Broe et al., 1998; Sturman et al., 2005;
van Gelder et al., 2004). None of the studies addressed
true lifestyle PA that incorporates all activity dimensions:
leisure-time, household, occupational, and transportation
PA. To gain a better understanding of the cognition–PA
relationship for Latinos, we need measures of PA that ade-
quately capture their lifestyle activity.
Another limitation of this group of studies examining the
PA–cognition relationship has been the failure to adequately
define intensity, frequency, and duration of PA. Therefore, it
is unknown if the 2008 Physical Activity Guidelines for
Americans by the U.S. Department of Health and Human
Services (2008) calling for 150 min of moderate PA per
week are protective for cognitive decline as well as for
improving cardiovascular health. Only one study (Singh-
Manoux, Richards, & Marmot, 2003) used a self-report
measure that assigned PAs a metabolic equivalence inten-
sity code (metabolic equivalent of task [MET], with 1.0 being
approximately equal to energy cost while laying quietly)
and then summed the amount of time spent in moderate/
vigorous-intensity PA to provide a calculation of the
minutes/hours per week of moderate/vigorous-intensity PA.
They found that adults who spent the least amount of time
in moderate-intensity (<2 hr) or high-intensity activity (<1
hr) over three 10-year points in time had greater cognitive
decline, thus suggesting a protective effect for cognition
as well as cardiovascular disease. A major limitation of
self-report measures of PA, however, is that they are largely
inaccurate for assessing the quantity of light- and moderate-
intensity unstructured lifestyle activity frequently performed
by older adults and immigrant populations (Brach, Kriska,
Glynn, & Newman, 2008). The optimum intensity and
duration of lifestyle PA needed to impact cognitive function
requires further investigation and confirmation with objec-
tive measures.
Accelerometers, PA monitors that can detect/record the
actual magnitude of acceleration, offer a more robust objective
alternative to self-report (Hawkins et al., 2009). They are
small and unobtrusive and can be worn throughout the
day without interfering with everyday movement. Another
unique feature is their ability to capture light-intensity PAs,
such as household cleaning, which is difficult to recall on
self-report questionnaires. Although it is well-established
that moderate-intensity PA reduces the risk of cardiovascu-
lar disease and diabetes (U.S. Department of Health and
Human Services, 2008), recent work has shown that light-
intensity PA is positively associated with physical health
Downloaded from https://academic.oup.com/psychsocgerontology/article-abstract/67/5/525/657933 by guest on 05 May 2019
and well-being (of which cognitive ability is one compo-
nent) in persons more than 65 years of age (Buman et al.,
2010). To date, the only study we identified that used accel-
erometers to relate PA with cognition found that total daily
activity counts were related to higher cognitive levels
(Buchman, Wilson, & Bennett, 2008). Unfortunately, inten-
sity and duration of PA in minutes per day or week were not
reported.
Finally, an additional issue was found with measures of
cognition, in that over half of the 10 studies relied on either
a single global measure of cognition (Ho, Woo, Sham,
Chan, & Yu, 2003; Lytle, Vander Bilt, Pandav, Dodge, &
Ganguli, 2004; Richards, Hardy, & Wadsworth, 2003;
Schuit, Feskens, Launer, & Kromhout, 2001; Yaffe et al.,
1999) or a composite measure (Albert et al., 1995; Sturman
et al., 2005; Weuve et al., 2004). There is literature that sug-
gests that some cognitive domains are affected by PA more
than others (Colcombe & Kramer, 2003). Therefore, use of
global and composite measures may obscure relationships
between PA and cognition that differ based upon the cogni-
tive domain.
The purpose of this study was to explore the relationship
between minutes spent participating in light- and moderate/
vigorous-intensity PA and different forms of cognition in
older Latinos, controlling for sex and for factors shown to
be associated with cognition. Our study extends previous
literature because it examines an understudied, at-risk,
rapidly growing population and uses accelerometers, an
innovative, comprehensive means of measuring PA.
Method
Design, Setting, and Sample
A descriptive, cross-sectional study design was used.
Inclusion criteria were 50 years of age or older, Latino, no
ambulation disability or assistive walking device, no evidence
of suspected dementia, and a Chicago metropolitan address.
To assess for suspected dementia, we used a modification of
the 30-point Mini-Mental State Examination (MMSE), a
global measure of cognition, for telephone administra-
tion (Folstein, Folstein, & McHugh, 1975). The modified
21-point MMSE did not include items that could not be done
over the telephone (9 points): orientation to room, naming a
pencil and watch, following a three-stage command, reading
 RELATIONSHIP
RELATIONSHIP BETWEEN
BETWEEN PA
PA AND
AND COGNITION
COGNITION
and obeying a command, writing a sentence, and copying a
design. To date 1,451 older people in the Rush Memory
and Aging Project have completed the standard 30-point
MMSE as part of their baseline evaluation (Bennett et al.,
2005). The Pearson correlation between the standard version
and the 21-point MMSE is 0.968 (p < .001). Previous studies
on cognition and aging using the 30-point MMSE with par-
ticipants of low-educational attainment have identified one-
third incorrect as a cutpoint for impaired/poor cognition (Raji,
Al Snih, Ostir, Markides, & Ottenbacher, 2010); therefore,
below 14 was our cutpoint for suspected dementia.
Recruitment sites (5 senior centers, 2 community health
care centers, 15 health fairs, 1 senior housing center) were
located in 11 Chicago community areas that have a moderate-
to-large Latino population (10%–89%) and in 1 suburban
community that is 8% Latino but adjacent to a community
that is 18% Latino (City of Chicago, n.d.). Recruitment
materials, available in Spanish and English, were reviewed
by four bilingual adults who met the study eligibility crite-
ria but did not participate in the study. They reviewed both
to determine comprehension and consistency between the
versions. Sign-up sheets were available at the study sites.
Overall, 342 persons signed up on a recruitment sheet at
all data collection sites; 282 were reached, and, of these,
224 agreed to be screened. Of those screened, 25 did not
meet the inclusion criteria (11 a score of <14 on the short-
ened MMSE, 6 too young, 6 used a walking assistive
device, 1 not Latino, 1 lived outside area); of the 199 eligi-
ble volunteers, 18 decided not to participate and 7 could not
schedule a time for an assessment, leaving 174 who partic-
ipated. There were no significant differences between
eligible volunteers who did and did not participate on
either mean age (both 66 years) or MMSE scores (18.73
vs. 19.31).
Measures
Measures of background characteristics and cognitive
function had been previously translated and were available
in Spanish or English.
Background characteristics.—Background character-
istics included demographics, chronic health problems, and
acculturation.
Demographics. Demographic variables were age, sex (1 =
men, 2 = women), and educational level (1 = no high school
degree, 2 = high school degree or higher). Descriptive
demographic characteristics included marital status, employ-
ment status, and country of origin.
Chronic health problems. Both diabetes and stroke place
persons at risk for cognitive decline (Haan et al., 2003). Par-
ticipants were asked to respond yes or no if they were on
medication for diabetes and/or hypertension. Blood pressure
5273
was taken with the Omron HEM-907XL automated blood
pressure machine (Omron Health Care Inc., Vernon Hills,
IL). It has reliability and reproducibility of mercury sphyg-
momanometer measurements (El Assaad, Topouchian, Darne,
& Asmar, 2002). Blood pressure measurements followed
the American Heart Association Recommendations for
human blood pressure determination (Pickering et al.,
2005), and the average of three measurements was recorded.
Consistent with National Health and Nutrition Examination
Survey (NHANES), a participant was identified as having
Downloaded from https://academic.oup.com/psychsocgerontology/article-abstract/67/5/525/657933 by guest on 05 May 2019
hypertension if he or she was taking medication for hyper-
tension or if the systolic blood pressure was ≥140 mmHg or
diastolic blood pressure was ≥90 mmHg (Roger et al.,
2011). Chronic health problems were 0 “does not have
hypertension or diabetes,” 1 “has either hypertension or
diabetes,” or 2 “has both hypertension and diabetes.”
Acculturation. The Acculturation Rating Scale for
Mexican Americans—II measures orientation toward the
Latino and the Anglo culture independently using two
subscales: a Latino orientation subscale (17 items, origi-
nal α = .88) and an Anglo orientation subscale (13 items,
original α = .83; Cuellar, Arnold, & Maldonado, 1995).
We revised this measure to include all Latino subgroups
(e.g., Puerto Ricans, Dominicans). Acculturation was
determined by subtracting the mean score for the Latino
subscale from the mean score for the Anglo subscale.
Based on the scores, Cuellar and colleagues (1995) iden-
tified five acculturative categories: very Latino oriented
(less than −1.33), Latino oriented to approximately bal-
anced bicultural (greater than or equal to −1.33 and less
than or equal to −.07), slightly Anglo oriented (greater
than −.07 and <1.19), strongly Anglo oriented (≥1.19 and
<2.45), and very assimilated (>2.45). Good reliability
was demonstrated in our study for the Spanish (α = .81)
and the English (α = .73) versions.
Cognitive function.—Cognitive function was assessed
with a battery of seven performance tests: two of episodic
memory and five of executive function processes. Executive
function tests were selected to represent domains of execu-
tive function including interference control, inattention,
and verbal fluency (Murphy, Barkley, & Bush, 2001). All
tests were chosen because they have been shown to test
functions that decline in old age (Wilson, Beckett et al., 2002),
shown to test functions that are likely to be related to PA
(Kramer, Erickson, & Colcombe, 2006), and translated
into Spanish and shown to be valid across ethnic and
socioeconomic backgrounds (Krueger, Wilson, Bennett, &
Aggarwal, 2009). The battery of tests met the practical
demands of the study (i.e., brief, portable, acceptable to
older people).
The bilingual data collectors participated in 1 day of
training with follow-up observation. They were certified on
each test using performance-based criteria before they
4528 WILBUR
WILBUR
began data collection. The cognitive function tests are
administered in the following fixed order: East Boston
Memory Test (immediate recall), Stroop Neuropsychological
Screening Test, East Boston Memory Test (delayed recall),
numbers comparison, and category fluency.
East Boston Memory Test. Episodic memory was mea-
sured with the East Boston Memory Test, which involves
immediate and delayed recall of a brief story (Albert et al.,
1991). The story, which has 12 key elements, was read
slowly to the participant. The participant was asked to
immediately recall the story. After an approximate 3-min
delay (during which time the Stroop was administered), the
participant was asked to recall the story. Scores ranged from
0 to 12 story ideas recalled on immediate and delayed
recall. The episodic summary score was the mean of the
immediate and delayed recall.
Stroop Neuropsychological Screening Test. Interference
control was measured with the modified color–word task
(Wilson et al., 2005) of the Stroop Neuropsychological
Screening Test (Trenerry, Crosson, DeBoe, & Leber, 1989).
The participant named the color of ink in which a list of
color names were provided. In some cases, the color of ink
in which the word was printed was the same as the word and
for others the color differed from the word. For example,
the word red was in blue ink. The scores were the number of
colors correctly identified in 30 s, the number of incorrect
responses in 30 s, and the number of colors answered cor-
rectly minus the number of incorrect responses (Wilson
et al., 2005). Prior to administration of the color–word task,
the participant read a list of color names as quickly as pos-
sible for 30 s. According to Trenerry and colleagues (1989),
the color task may have a priming effect on the interference
shown in the color–word task. Three participants did not
complete the measure due to literacy problems.
Numbers Comparison Test. The Numbers Comparison
Test (Ekstron, French, Harmen, & Kermen, 1976), a measure
of inattention, requires participants to classify pairs of 3- to
10-digit sequences as the same or different. The attention
score was the number of pairs classified correctly in 90 s
minus the number classified incorrectly.
Category Fluency Test. Word fluency was assessed by
asking participants to generate as many examples as possible
from two semantic categories (animals, fruits, vegetables)
in separate 60-s trials (Welsh et al., 1994). The total number
of unique examples was the performance measure. A word
fluency score was arrived at by summing the number of
animals generated to the number of fruits and vegetables
generated.
Physical Activity.—PA was monitored with the portable
GT1M ActiGraph accelerometer (Tryon & Williams, 1996).
The ActiGraph monitor provides a valid assessment of PA
in adult men and women during treadmill walking/running
and daily activity (Hendelman, Miller, Baggett, Debold, &
Freedson, 2000). It records vertical accelerations as “counts.”
Participants were instructed to wear the monitor on their
hips for seven consecutive days. Participants were given an
accelerometer log on which they recorded the hours they
wore it. The monitor was initialized and set to collect count
values at 1-min sampling intervals (epochs). To be included
in the analyses, ≥10 hr of data or wear time had to be available
Downloaded from https://academic.oup.com/psychsocgerontology/article-abstract/67/5/525/657933 by guest on 05 May 2019
in a 24-hr period on at least 3 days (Matthews, Ainsworth,
Thompson, & Bassett, 2002). As in Troiano and colleagues
(2008), nonwear time was defined as at least 60 consecutive
minutes of 0 activity counts.
PA intensity was categorized by cutpoints of Miller,
Strath, Swartz, and Cashin, (2010) used with persons 60–69
years of age. The mean number of minutes of PA was
defined as: sedentary <100 counts/min, light 100–1,565
counts/min (<3 METs), moderate 1,566–6,139 counts/min
(3.0–6.0 METs), and vigorous ≥6,140 counts/min (≥6.1
METs). We report the number of minutes of light-intensity
PA per day and combine the number of minutes of moderate
and vigorous PA per day due to low participation in vigor-
ous activity. In addition, we provide the number of activity
counts per minute, which is an estimate that corrects for the
amount of wear time.
Procedures
Interested volunteers were screened in person or over the
telephone for age, ethnicity, and disability and for cognition
with the MMSE. Eligible persons were scheduled for an ap-
pointment at a location of their choosing, including their
homes (n = 65, 37%), one of the community recruitment
sites (n = 99, 57%), or our program offices (n = 10, 6%). To
closely simulate the privacy of the community sites and pro-
gram office setting, participants who chose their home were
asked to identify a room and time when there would be no
outside interruptions. At the appointment, a bilingual staff
member further explained the study and read the consent
form aloud in the participant’s preferred language. After
consent was given, the health and demographics question-
naire was administered followed by assessment of blood
pressure, resting heart rate, height, weight, and waist circum-
ference. Then the cognitive function measures were adminis-
tered. Next, additional questionnaires related to determinants
of PA and health (e.g., barriers to PA, social support, and
nutritional history), to be reported in a later manuscript
related to these issues, were administered. Once the ques-
tionnaires were completed, participants watched a demonstra-
tion of the accelerometer and were then asked to demonstrate
their competence. Participants wore their accelerometer for
the next 7 days. A second appointment was made to download
the accelerometer data, go over the results, and complete
any questionnaires administered after the cognitive function
 RELATIONSHIP
RELATIONSHIP BETWEEN
BETWEEN PA
PA AND
AND COGNITION
COGNITION 5295

Table 1. Background Characteristics correlation analyses to determine how the dependent vari-
n % ables were related to the predictors.
Age
50–59 54 31.0
60–69 52 29.9
Results
70–79 51 29.3
≥80 17 9.8 Background Characteristics
Education The mean age of the participants was 66 years (SD =
Did not complete high school 99 56.9 9.12; range 50–84 years; Table 1). The majority of the
Completed high school or GED 29 16.7
Some college or vocational school 26 14.9 participants were women (74%), unmarried (59%), and not

Downloaded from https://academic.oup.com/psychsocgerontology/article-abstract/67/5/525/657933 by guest on 05 May 2019

Completed college or higher 20 11.5 working (76%). Over half (56%) of the participants had not
Sex graduated from high school. The main country of origin was
Men 46 26.4
Mexico (55%). A total of 144 (83%) participants chose to
Women 128 73.6
Marital respond in Spanish. The mean acculturation score was
Never married 13 7.5 −1.47 (SD = 1.16, range −3.23 to 1.98), which is in the
Married 72 41.4 “very Latino-oriented” range. Nearly 70% of the partici-
Widowed 46 26.4
Separated 13 7.5
pants had a self-reported chronic health problem. Eight par-
Divorced 26 14.9 ticipants had diabetes only, 78 had hypertension only, and
Member of an unmarried couple 4 2.3 35 had both hypertension and diabetes.
Employment
Working 42 24.1
Retired 89 51.2 PA and Cognitive Function
Other 43 24.7
Country of origin Table 2 shows the mean score and standard deviation for
USA 29 16.7 minutes of accelerometer-assessed PA and cognitive function
Mexico 95 54.6 domains. Of the 174 participants, 151 had valid accelerom-
Puerto Rico 24 13.8
Central American 2 1.1
eter data, 4 lost their accelerometers, 6 had no valid data due
South American 11 6.3 to accelerometer malfunctions, and 13 had only 1 or 2 days
Other/missing 13 7.5 of valid data. More minutes were spent in light-intensity
Acculturation levels than in moderate/vigorous-intensity PA. There were no sig-
Very Latino oriented 107 62.6
Latino to balanced bicultural 39 22.8
nificant differences on background characteristics or cogni-
Slightly Anglo bicultural 21 12.3 tive function tests between the participants who had valid
Strongly Anglo 4 2.3 accelerometer data and those who did not. Test scores from
Chronic health problems
one of the cognitive function domains (episodic memory)
0 52 30.1
1–2 121 69.9 had a negatively skewed distribution, whereas the compo-
Note. N = 174.
nents of executive function (interference control, inattention,
word fluency) were relatively symmetric.

tests that remained. At the end of the second visit, partici-
pants were given the oral and written results of their blood
pressure, weight, and height information and PA informa-
tion from the American Heart Association in Spanish or
English. Participants were compensated $10 at the end of
the first appointment and received another $10 at the end of
the second appointment.
Analyses
All analyses were conducted using SPSS V16.0. The
bivariate correlations between background/control measures
and PA with the cognitive measures were conducted to
determine if further multiple regression analysis was justified.
Next, a multivariate multiple regression was done with
episodic memory, interference control, inattention, and the
two-word fluency measures (animals and fruits and vegeta-
bles) as the dependent variables and the background and PA
variables as predictors. This was followed by a canonical
Correlations Among Background Characteristics, PA, and
Cognitive Function
Age was significantly negatively correlated with light-
intensity PA, minutes of moderate/vigorous-intensity PA,
and activity counts (Table 2). Correlations of small magni-
tude were found between chronic health problems and
minutes of light-intensity PA a day (r = −.221). There was a
correlation between sex and moderate/vigorous-intensity
PA, with men having more minutes in moderate/vigorous-
intensity PA than women.
Correlations were assessed between the PA measures and
cognitive measures. In addition, we looked at the composite
scores for episodic memory (Boston immediate and Boston
delay) and PA measures. The pattern of correlations for
the component and summary/composite scores for episodic
memory was nearly identical across all three PA measures.
Consequently, the subsequent analyses examined only the
episodic composite.
 Table 2. Correlations Among Background Characteristics, Physical Activity (PA), and Cognition
M (SD) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.
Background characteristics
1. Age
2. Sexa −.012
3. Educationb −.268*** .000
4. Chronic health problems .296*** −.203* −.117
5. Acculturation −.020 −.010 .277*** .176*
PA
6. Lightc 259.4 (73.3) −.438*** .096 .139 −.221** .087
7. Moderate/vigorousc 31.2 (30.5) −.318*** −.255** −.036 −.093 .036 .321***
8. Counts per minute 267.0 (181.9) −.280*** −.112 −.022 −.151 .041 .368*** .865***
Cognition
9. Boston immediate 8.8 (2.0) −.000 .165* .054 −.121 .078 .085 .013 .014
10. Bostondelay 8.6 (2.0) −.023 .216** .076 −.117 .112 .091 .056 .067 .771***
6530 WILBUR
WILBUR

11. Episodic memory 8.7 (2.0) −.012 .203* .069 −.126 .101 .093 .037 .043 .939*** .943***
12. Interference control 14.8 (9.5) −.486*** .026 .258** −.140 .192* .234** .150 .137 .295*** .319*** .327***
13. Inattention 16.7 (7.3) −.311*** .002 .494*** −.086 .244** .119 .062 .058 .197* .139 .178* .495***
14. Number of animals 17.2 (4.5) −.247** .009 .244** −.040 .028 .253** .231** .240** .276** .277** .294*** .389*** .401***
15. Number of fruits and 17.1 (5.2) −.211** .306*** .204* −.164* .029 .246** .079 .114 .204* .263** .249** .334*** .341*** .574***
vegetables
16. Word fluency 34.3 (8.7) −.260** .165* .254** −.109 .032 .281*** .181* .205* .273*** .305*** .308*** .410*** .421*** .868*** .905***
Notes. n = 151.
a 1. = men; 2. = women.
b 1. = no high school degree; 2. = high school degree or higher.
c Minutes per day at named intensity.

*p ≤ .05; **p ≤ .01; ***p ≤ .001.

Downloaded from https://academic.oup.com/psychsocgerontology/article-abstract/67/5/525/657933 by guest on 05 May 2019

 RELATIONSHIP
RELATIONSHIP BETWEEN
BETWEEN PA
PA AND
AND COGNITION
COGNITION 5317

Table 3. The Correlations Between Background and Physical Table 3 also displays the extent to which the four func-
Activity (PA) Predictor Variables and Canonical Functions tions are correlated with cognitive functioning. Functions
Canonical functions 1 and 3 both predict interference control, and inattention
Predictor variables 1 2 3 4 Function 2 is most correlated with episodic memory and
Background word fluency (fruits and vegetables recalled). Function 4 is
Age .81 −.04 .54 .05 most highly correlated with the two measures of word flu-
Sex .05 −.92 −.12 .04
Education −.72 −.13 .63 −.10
ency, the number of fruit and vegetables recalled (r = −.44)
Chronic health problems .19 −.41 .18 .15 and the number of animals recalled (r =−.83). Function 4 is
Acculturation −.38 −.07 .21 .40 composed of the three measures of PA and one measure of
PA acculturation. From these results, it is clear that there is a

Downloaded from https://academic.oup.com/psychsocgerontology/article-abstract/67/5/525/657933 by guest on 05 May 2019

Light −.34 −.24
Moderate/vigorous −.23 −.13
Counts per minute −.21 .05
Dependent variables
Episodic memory −.09 −.56
Interference control −.83 −.08
Inattention −.84 −.12
Word fluency animals −.48 −.10
Word fluency fruits and vegetables −.38 −.75
Note. Loadings above .40 are marked in bold.
Light-intensity PA had significant positive correlations
with interference control. Minutes of light-intensity PA per
day, minutes of moderate/vigorous PA per day, and PA
counts/minute all had significant positive correlations with
word fluency. PA was not significantly correlated with epi-
sodic memory or inattention.
Age had significant negative correlations, and education had
significant positive correlations with interference control, inat-
tention, and word fluency (animals, fruits, and vegetables), but
not with episodic memory. Women had higher scores on tests
of episodic memory and word fluency. Chronic health prob-
lems were significantly negatively correlated with only number
of fruits and vegetables. Acculturation had significant positive
correlations with both interference control and inattention but
not with episodic memory or word fluency.
Regression of Cognitive Function on Background
Characteristics and PA
The overall multivariate regression analysis showed a
significant relationship among the background measures,
PA measures, and dependent cognitive measures. For the
multivariate regression, the approximate F-ratio, based on
Pillai’s trace was 3.38 (degrees of freedom = 40, 710; p <
.001; Bock, 1975). From this analysis, four canonical func-
tions were extracted, and they combined to account for
82.7% of the variance associated with the five dependent
variables. The percent of variance associated with these
four functions was 36.4%, 18.5%, 9.8%, and 18.0%, respec-
tively. Table 3 displays the correlations between the eight
predictor variables and the four functions. Age and education
load on the first and third functions. Sex and chronic health
problems load on the second function. PA (minutes of light
activity r = −.51, minutes of moderate PA r = −.56, counts
per minute r = −62) and acculturation (r = .40) load on the
fourth function.
−.35 −.51
−.23 −.56
relationship between PA and word fluency.
−.23 −.62
.04 .01 Discussion
−.45 .11 The relationship between PA and cognitive function was
.51 .03
examined in a sample of older Latinos who had low levels
.00 −.83
−.12 −.44 of education, were highly Latino acculturated, and most of
whom were born in Mexico. These individuals were recruited
from predominantly Latino Chicago communities and rep-
resent a population for which little is known about their
cognition and its relationship to PA. Another distinctive fea-
ture of this study is that PA was assessed with accelerometers.
We know of only one other study to date that has used accel-
erometers to measure PA and examined its relationship with
cognition, and that study was restricted to a primarily
non-Latino White (92%) sample (Buchman et al., 2008).
Problems with misrepresentation of PA in Latinos using
self-report PA measures are well-documented (Ham &
Ainsworth, 2010). Although Latinos are shown to have
lower levels of leisure-time PA than non-Latino White par-
ticipants, recent evidence suggests that moderate/vigorous
PA measured with an accelerometer is actually higher for
Latinos than for non-Latino Whites (Troiano et al., 2008).
This is thought to be due to the inclusion of occupational
and transportation PA captured with accelerometers. There-
fore, accelerometers may better capture the influence of
lifestyle PA on cognition in this population.
Overall, scores on the cognitive tests from the current
sample are comparable to previous work of Krueger and
colleagues (2009) with older, nondemented community-
based Latinos. With the exception of interference control,
all cognitive function tests were slightly higher in our study
(episodic memory: immediate recall 8.8 vs. 8.5, delayed
recall 8.6 vs. 7.4; interference control: 14.8 vs. 15.0; inat-
tention: number comparison 16.7 vs. 13.3; word fluency:
category fluency 34.3 vs. 30.5). This slight discrepancy may
be due to the lower age limit in our study compared with that
of Krueger and colleagues (2009; 50 vs. 55 years of age).
Mean PA counts of Latinos in our study (263) were compa-
rable to NHANES findings for Mexican Americans 60 years
of age and older (283.5 men; 214.6 women) but higher than
for non-Latino Whites (221.6 men; 208.7 women; Troiano
et al., 2008). Participants in our study spent a lot of time per-
forming light-intensity activities, similar to older adult partici-
pants in the senior quality-of-life study (Buman et al., 2010).
8532 WILBUR
WILBUR
We examined correlations between all study variables to
identify congruence with prior studies on age, education,
sex, PA, and cognition. As expected, age was negatively
correlated and education positively correlated with the
majority of the dimensions of cognitive function (Plassman
et al., 2007). The lack of a correlation between age and epi-
sodic memory was surprising. There is evidence suggesting
that frequent participation in cognitively stimulating activ-
ities is associated with reduced risk of dementia (Wilson
et al., 2005). Therefore, we speculate that the rich tradition
of oral history that is characteristic of the Latino community
may provide some protective effects on episodic memory
(Caspe, 2009).
As demonstrated in prior studies, there was a negative
correlation between chronic health problems and all cogni-
tive measures, but only one component of word fluency was
significant (Kodl & Seaquist, 2008; Plassman et al., 2007).
Findings related to sex have been unequivocal, with several
earlier studies finding women to be at greater risk for
dementia than men (Bachman et al., 1992; Canadian Study
of Health and Aging Working Group, 1994), but recent U.S.
prevalence data found no sex differences in cognitive func-
tion (Plassman et al., 2007). We found women perform better
on episodic memory and word fluency. Consistent with
prior studies on PA and aging in Latinos, we found a signif-
icant negative correlation between age and PA, and men
tended to be more active than women (Troiano et al., 2008).
We selected tests that assess executive function because
prior studies show it has the largest fitness-induced cogni-
tive benefits (Colcombe & Kramer, 2003). Accordingly, we
found PA (minutes of both light and moderate/vigorous PA)
correlated significantly with two of the three domains of
executive function (word fluency and interference control).
Not surprisingly, there was no relationship with episodic
memory, which does not tap into executive function.
Subsequently, in the regression analyses, only word fluency
retained a relationship with PA, and the effect was larger for
those with a more Latino orientation. It is uncertain why PA
was more strongly related to rule-based retrieval of acquired
knowledge (identification of animals, fruits, vegetables)
compared with other aspects of executive function such as
interference control and numbers comparison. Comparing
our findings with studies of cohorts with healthy people at
baseline was a challenge due to considerable variation in
cognitive tests across studies and the use of single global
and composite scores. There is a call for researchers to identify
a small group of standard measurement instruments to allow
for better comparison across studies (Snowden et al., 2011).
Our findings seem to suggest that some domains of executive
function may be more susceptible to PA and that, not all, may
be equally affected. Further research is needed in identifying
the effects of PA on specific domains of executive function.
We believe that this is one of the first studies to objec-
tively measure light-intensity PA in relation to cognitive
function. Our findings suggest that light-intensity PA may
provide cognitive benefits for word fluency as does moder-
ate/vigorous PA. This could be particularly important for
older adults who, like participants in our study, tend to
spend more time participating in light-intensity than more
vigorous activities. We must keep in mind that our estimates
between moderate PA and cognition may be somewhat mis-
leading because of the floor effects. It may be that increases
in moderate PA will lead to substantial increases in cogni-
tive functioning, but we were not able to find this relation-
ship due to the limited range of moderate/vigorous PA in
Downloaded from https://academic.oup.com/psychsocgerontology/article-abstract/67/5/525/657933 by guest on 05 May 2019
this study. However, this needs further study using prospec-
tive research designs before substantive recommendations
can be made.
Despite the unique findings, there are several limitations
to this study. A major limitation is that it is cross-sectional;
thus, we cannot identify the true direction of associations
between PA and cognitive function. Another limitation is
that the study is composed of volunteers, so the findings have
to be cautiously generalized beyond the study population.
The participants, however, were recruited from multiple
communities in the Chicago area, thus reflecting those of
Mexican, Puerto Rican, and other Latino backgrounds.
The strengths of this study include the inclusion of highly
Latino–acculturated participants who had the opportunity to
respond in their preferred language. Also, using accelerom-
eters allowed us to assess lifestyle activity (i.e., activity dur-
ing leisure, work, homemaking, walking for transportation).
Unfortunately, the accelerometer does not allow us to
identify the specific types of activity in which participants
engaged.
In summary, we found an association between higher levels
of PA and higher levels of cognitive function for word
fluency in older Latinos similar to other cross-sectional and
epidemiological studies. Most importantly, we found that
cognitive benefits may be gained from both light-intensity
and moderate/vigorous activity. Longitudinal work needs to
be done to identify the effects of PA on the various domains
of executive function.
Funding
This work was supported by the Rush University Medical Center Pilot
Funding.
Acknowledgments
We acknowledge Janet Flores, research assistant, who assisted in
recruitment and data collection. We thank Kevin Grandfield for editorial
assistance.
Correspondence
Correspondence should be addressed to JoEllen Wilbur, PhD, College
of Nursing, Rush University Medical Center, 600 S Paulina, Suite 1062A,
Chicago, IL 60612. E-mail: joellen_wilbur@rush.edu.
References
Albert, M. S., Jones, K., Savage, C. R., Berkman, L., Seeman, T.,
Blazer, D., . . . Rowe, J. W. (1995). Predictors of cognitive change
in older persons: MacArthur studies of successful aging. Psychol-
ogy and Aging, 10, 578–589. doi:10.1037/0882-7974.10.4.578
 RELATIONSHIP
RELATIONSHIP BETWEEN
BETWEEN PA
PA AND
AND COGNITION
COGNITION
Albert, M. S., Smith, L. A., Scherr, P. A., Taylor, J. O., Evans, D. A., &
Funkenstein, H. H. (1991). Use of brief cognitive tests to identify
individuals in the community with clinically diagnosed Alzheimer’s
disease. International Journal of Neuroscience, 57, 167–178.
doi:10.3109/00207459109150691
Alzheimer’s Association. (2010). 2010 Alzheimer’s disease facts and
figures. Alzheimer’s and Dementia, 6, 158–194. doi:10.1016/
j.jalz.2010.01.009
Bachman, D. L., Wolf, P. A., Linn, R., Knoefel, J. E., Cobb, J., Belanger, A.,
. . . White, L. R. (1992). Prevalence of dementia and probable
senile dementia of the Alzheimer type in the Framingham Study.
Neurology, 42, 115–119. Retrieved from http://www.neurology.org/
Bennett, D. A., Schneider, J. A., Buchman, A. S., Mendes de Leon, C.,
Bienias, J. L., & Wilson, R. S. (2005). The rush memory and aging
project: Study design and baseline characteristics of the study cohort.
Neuroepidemiology, 25, 163–175. doi:10.1159/000087446
Berrigan, D., Troiano, R. P., McNeel, T., DiSogra, C., & Ballard-Barbash, R.
(2006). Active transportation increases adherence to activity recom-
mendations. American Journal of Preventive Medicine, 31, 210–216.
doi:10.1016/j.amepre.2006.04.007
Bock, R. D. (1975). Multivariate statistical methods in behavioral research.
New York: McGraw-Hill.
Brach, J. S., Kriska, A. M., Glynn, N. W., & Newman, A. B. (2008). Physical
activity and the older adult: Measurement, benefits, and risks. Current
Cardiovascular Reports, 2, 305–310. doi:10.1007/s12170-008-0056-6
Broe, G. A., Creasey, H., Jorm, A. F., Bennett, H. P., Casey, B., Waite, L. M., . . .
Cullen, J. (1998). Health habits and risk of cognitive impairment and
dementia in old age: A prospective study on the effects of exercise,
smoking and alcohol consumption. Australian and New Zealand Journal
of Public Health, 22, 621–623. doi:10.1111/j.1467-842X.1998.tb01449.x
Buchman, A. S., Wilson, R. S., & Bennett, D. A. (2008). Total daily activity is
associated with cognition in older persons. American Journal of Geri-
atric Psychiatry, 16, 697–701. doi:10.1097/JGP.0b013e31817945f6
Buman, M. P., Hekler, E. B., Haskell, W. L., Pruitt, L., Conway, T. L., Cain,
K. L., . . . King, A. C. (2010). Objective light-intensity physical activity
associations with rated health in older adults. American Journal of
Epidemiology, 172, 1155–1165. doi:10.1093/aje/kwq249
Canadian Study of Health and Aging Working Group. (1994). Canadian
study of health and aging: Study methods and prevalence of dementia.
Canadian Medical Association Journal, 150, 899–913. Retrieved
from http://www.cmaj.ca/
Caspe, M. (2009). Low-income Latino mother’s booksharing styles and
children’s emergent literacy development. Early Childhood Re-
search Quarterly, 24, 306–324. doi:10.1016/j.ecresq.2009.03.006
City of Chicago. (n.d.). Zoning and land use planning: Community Area
2000 Census Profiles. Retrieved from http://www.cityofchicago.org/
content/city/en/depts/dcd/supp_info/community_area_2000census
profiles.html
Colcombe, S., & Kramer, A. F. (2003). Fitness effects on the cognitive
function of older adults: A meta-analytic study. Psychological Science,
14, 125–130. doi:10.1111/1467-9280.t01-1-01430
Cuellar, I., Arnold, B., & Maldonado, R. (1995). Acculturation Rating
Scale of Mexican Americans II—A revision of the original ARSMA
Scale. Hispanic Journal of Behavioral Sciences, 17, 275–304.
doi:10.1177/07399863950173001
Ekstron, R. B., French, J. W., Harmen, H. H., & Kermen, D. (1976). Manual
for kit of factor-referenced cognitive test. Princeton, NJ: Educational
Testing Service.
El Assaad, M. A., Topouchian, J. A., Darne, B. M., & Asmar, R. G. (2002).
Validation of the Omron HEM-907 device for blood pressure mea-
surement. Blood Pressure Monitoring, 7, 237–241. doi:10.1097/
00126097-200208000-00006
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. doi:10.1016/
0022-3956(75)90026-6
5339
Greenberg, S. (2009). A profile of older Americans: 2009. Retrieved from
http://www.aoa.gov/AoAroot/Aging_Statistics/Profile/2009/docs/
2009profile_508.pdf
Haan, M. N., Mungas, D. M., Gonzalez, H. M., Ortiz, T. A., Acharya, A., &
Jagust, W. J. (2003). Prevalence of dementia in older Latinos: The
influence of type 2 diabetes mellitus, stroke and genetic factors. Journal
of American Geriatrics Society, 51, 169–177. doi:10.1046/j.1532-
5415.2003.51054.x
Ham, S. A., & Ainsworth, B. E. (2010). Disparities in data on Healthy
People 2010 physical activity objectives collected by accelerometry
and self-report. American Journal of Public Health, 100, S263–S268.
Downloaded from https://academic.oup.com/psychsocgerontology/article-abstract/67/5/525/657933 by guest on 05 May 2019
doi:10.2105/AJPH.2009.180075
Hamer, M., & Chida, Y. (2009). Physical activity and risk of neurodegen-
erative disease: A systematic review of prospective evidence. Psy-
chological Medicine, 39, 3–11. doi:10.1017/S0033291708003681
Hawkins, M. S., Storti, K. L., Richardson, C. R., King, W. C., Strath, S. J.,
Holleman, R. G., & Kriska, A. M. (2009). Objectively measured
physical activity of USA adults by sex, age, and racial/ethnic groups:
A cross-sectional study. International Journal of Behavioral Nutrition
and Physical Activity, 6, 31. doi:10.1186/1479-5868-6-31
Hendelman, D., Miller, K., Baggett, C., Debold, E., & Freedson, P. (2000).
Validity of accelerometry for the assessment of moderate intensity
physical activity in the field. Medicine and Science in Sports and
Exercise, 32, S442–S449. doi:10.1097/00005768-200009001-00002
Hillman, C. H., Erickson, K. I., & Kramer, A. F. (2008). Be smart, exercise
your heart: Exercise effects on brain and cognition. Nature Reviews
Neuroscience, 9, 58–65. doi:10.1038/nrn2298
Ho, S. C., Woo, J., Sham, A., Chan, S. G., & Yu, A. L. (2003). A 3-year
follow-up study of social, lifestyle and health predictors of cognition
impairment in a Chinese older cohort. International Journal of
Epidemiology, 30, 1389–1396. doi:10.1093/ije/30.6.1389
Kodl, C. T., & Seaquist, E. R. (2008). Cognitive dysfunction and diabetes
mellitus. Endocrine Reviews, 29, 494–511. doi:10.1210/er.2007-0034
Kramer, A. F., Erickson, K. I., & Colcombe, S. J. (2006). Exercise, cognition,
and the aging brain. Journal of Applied Physiology, 101, 1237–1242.
doi:10.1152/japplphysiol.00500.2006
Krueger, K. R., Wilson, R. S., Bennett, D. A., & Aggarwal, N. T. (2009).
A battery of tests for assessing cognitive function in older Latino
persons. Alzheimer Disease and Associated Disorders, 23, 384–388.
doi:10.1097/WAD.0b013e31819e0bfc
Lytle, M. E., Vander Bilt, J., Pandav, R. S., Dodge, H. H., & Ganguli, M.
(2004). Exercise level and cognitive decline: The MoVIES project.
Alzheimer Disease and Associated Disorders, 18, 57–64. doi:00002093-
200404000-00003
Matthews, C. E., Ainsworth, B. E., Thompson, R. W., & Bassett, D. R., Jr.
(2002). Sources of variance in daily physical activity levels as
measured by an accelerometer. Medicine and Science in Sports
and Exercise, 34, 1376–1381. doi:10.1097/00005768-200208000-
00021
Miller, N. E., Strath, S. J., Swartz, A. M., & Cashin, S. E. (2010). Estimating
absolute and relative physical activity intensity across age via accel-
erometry in adults. Journal of Aging and Physical Activity, 18, 158–170.
Retrieved from http://journals.humankinetics.com/japa
Murphy, K. R., Barkley, R. A., & Bush, T. (2001). Executive functioning
and olfactory identification in young adults with attention deficit-
hyperactivity disorder. Neuropsychology, 15, 211–220. doi:10.1037//
0894-4105.15.2.211
Pickering, T. G., Hall, J. E., Appel, L. J., Falkner, B. E., Graves, J., Hill,
M. N., . . . Rocella, E. J. (2005). Recommendations for blood pressure
measurement in humans and experimental animals: Part 1: Blood pres-
sure measurement in humans: A statement for professionals from the
Subcommittee of Professional and Public Education of the American
Heart Association Council on High Blood Pressure Research. Hyper-
tension, 45, 142–161. doi:10.1161/01.CIR.0000154900.76284.F6
Plassman, B. L., Langa, K. M., Fisher, G. G., Heeringa, S. G., Weir, D. R.,
Ofstedal, M. B., . . . Wallace, R. B. (2007). Prevalence of dementia in
534 WILBUR
10 WILBUR
the United States: The aging, demographics, and memory study.
Neuroepidemiology, 29, 125–132. doi:10.1159/000109998
Raji, M. A., Al Snih, S., Ostir, G. V., Markides, K. S., & Ottenbacher, K. J.
(2010). Cognitive status and future risk of frailty in older Mexican
Americans. The Journals of Gerontology, Series A: Biological
Sciences and Medical Sciences, 65, 1228–1234. doi:10.1093/gerona/
glq121
Richards, M., Hardy, R., & Wadsworth, M. E. (2003). Does active leisure
protect cognition? Evidence from a national birth cohort. Social
Science and Medicine, 56, 785–792. doi:S0277953602000758
Roger, V. L., Go, A. S., Lloyd-Jones, D. M., Adams, R. J., Berry, J. D.,
Brown, T. M., . . . Wylie-Rosett, J. (2011). Heart disease and stroke
statistics—2011 update: A report from the American Heart Asso-
ciation. Circulation, 123, e18–e209. doi:10.1161/CIR.0b013e3
182009701
Schuit, A. J., Feskens, E. J., Launer, L. J., & Kromhout, D. (2001). Physi-
cal activity and cognitive decline, the role of the apolipoprotein
e4 allele. Medicine and Science in Sports and Exercise, 33, 772–777.
Retrieved from http://journals.lww.com/acsm-msse/pages/default.
aspx
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.
doi:10.1136/jech.57.11.907
Snowden, M., Steinman, L., Mochan, K., Grodstein, F., Prohaska, T. R.,
Thurman, D. J., . . . Anderson, L. A. (2011). Effect of exercise
on cognitive performance in community-dwelling older adults:
Review of intervention trials and recommendations for public health
practice and research. Journal of American Geriatrics Society, 59,
704–716. doi:10.1111/j.1532-5415.2011.03323.x
Sturman, M. T., Morris, M. C., Mendes de Leon, C. F., Bienias, J. L.,
Wilson, R. S., & Evans, D. A. (2005). Physical activity, cognitive
activity, and cognitive decline in a biracial community population.
Archives of Neurology, 62, 1750–1754. doi:10.1001/archneur.62.
11.1750
Tremblay, M. S., Bryan, S. N., Peréz, C. E., Ardern, C. I., & Katzmarzyk,
P. T. (2006). Physical activity and immigrant status: Evidence from
the Canadian Community Health Survey. Canadian Journal of Public
Health, 97, 277–282. Retrieved from http://www.ncbi.nlm.nih.gov/
entrez / query. fcgi ? cmd = Retrieve & db = PubMed & dopt = Citation &
list_uids=16967745
Trenerry, M., Crosson, B., DeBoe, J., & Leber, W. (1989). Stroop Neuro-
psychological Screening Test manual. Adessa, FL: Psychological
Assessment Resources (PAR).
Troiano, R. P., Berrigan, D., Dodd, K. W., Masse, L. C., Tilert, T., &
McDowell, M. (2008). Physical activity in the United States measured
by accelerometer. Medicine and Science in Sports and Exercise, 40,
181–188. doi:10.1249/mss.0b013e31815a51b3
Tryon, W. W., & Williams, R. S. (1996). Fully proportional actigraphy:
A new instrument. Behavior Research Methods, Instruments, and
Computers, 28, 392–403. doi:10.3758/BF03200519
U.S. Department of Health and Human Services. (2008). 2008 Physical
Downloaded from https://academic.oup.com/psychsocgerontology/article-abstract/67/5/525/657933 by guest on 05 May 2019
activity guidelines for Americans. Washington DC: Author.
van Gelder, B. M., Tijhuis, M. A., Kalmijn, S., Giampaoli, S., Nissinen, A., &
Kromhout, D. (2004). Physical activity in relation to cognitive
decline in elderly men: The FINE Study. Neurology, 63, 2316–2321.
doi:10.1016/S0197-4580(04)81579-3
Welsh, K. A., Butters, N., Mohs, R. C., Beekly, D., Edland, S., Fillenbaum,
G., & Heyman, A. (1994). The consortium to establish a registry for
Alzheimer’s disease (CERAD). Part V. A normative study of the
neuropsychological battery. Neurology, 44, 609–614.
Weuve, J., Kang, J. H., Manson, J. E., Breteler, M. M., Ware, J. H., &
Grodstein, F. (2004). Physical activity, including walking, and cogni-
tive function in older women. The Journal of the American Medical
Association, 292, 1454–1461. doi:10.1001/jama.292.12.1454
Wilson, R. S., Barnes, L. L., Krueger, K. R., Hoganson, G., Bienias, J. L., &
Bennett, D. A. (2005). Early and late life cognitive activity and cog-
nitive systems in old age. Journal of the International Neuropsycho-
logical Society, 11, 400–407. doi:10.1017/S1355617705050459
Wilson, R. S., Beckett, L. A., Barnes, L. L., Schneider, J. A., Bach, J.,
Evans, D. A., & Bennett, D. A. (2002). Individual differences in rates
of change in cognitive abilities of older persons. Psychology and
Aging, 17, 179–193. doi:10.1037//0882-7974.17.2.179
Wilson, R. S., Mendes De Leon, C. F., Barnes, L. L., Schneider, J. A.,
Bienias, J. L., Evans, D. A., & Bennett, D. A. (2002). Participation in
cognitively stimulating activities and risk of incident Alzheimer dis-
ease. Journal of the American Medical Association, 287, 742–748.
doi:10.1001/jama.287.6.742
Yaffe, K., Blackwell, T., Gore, R., Sands, L., Reus, V., & Browner, W. S.
(1999). Depressive symptoms and cognitive decline in nondemented
elderly women: A prospective study. Archives of General Psychiatry,
56, 425–430. doi:10.1001/archpsyc.56.5.425