Content uploaded by Jutta Mata
Author content
All content in this area was uploaded by Jutta Mata on Feb 19, 2014
Content may be subject to copyright.
Exercise Holds Immediate Benefits for Affect and Cognition in Younger
and Older Adults
Candice L. Hogan
Stanford University
Jutta Mata
Stanford University and Max Planck Institute for Human
Development, Berlin, Germany
Laura L. Carstensen
Stanford University
Physical activity is associated with improved affective experience and enhanced cognitive processing.
Potential age differences in the degree of benefit, however, are poorly understood because most studies
examine either younger or older adults. The present study examined age differences in cognitive
performance and affective experience immediately following a single bout of moderate exercise.
Participants (144 community members aged 19 to 93) were randomly assigned to one of two experi-
mental conditions: (a) exercise (15 min of moderate intensity stationary cycling) or (b) control (15 min
completing ratings of neutral IAPS images). Before and after the manipulation, participants completed
tests of working memory and momentary affect experience was measured. Results suggest that exercise
is associated with increased levels of high-arousal positive affect (HAP) and decreased levels of
low-arousal positive affect (LAP) relative to control condition. Age moderated the effects of exercise on
LAP, such that younger age was associated with a drop in reported LAP postexercise, whereas the effects
of exercise on HAP were consistent across age. Exercise also led to faster RTs on a working memory task
than the control condition across age. Self-reported negative affect was unchanged. Overall, findings
suggest that exercise may hold important benefits for both affective experience and cognitive perfor-
mance regardless of age.
Keywords: physical activity, emotion, working memory, aging, n-back task
Health experts often remark that if exercise came in pill form it
would be the most sought-after drug on the market. For decades,
research has frequently identified exercise as an important tool for
enhancing a range of physical indices from balance, bone density,
strength, and endurance to lipid profiles, blood pressure, and
cardiovascular health (Atha, 1981; Bassey & Ramsdale, 1994;
Hickson, Bomze, & Holloszy, 1977; Judge, Lindsey, Underwood,
& Winsemius, 1993; Thompson et al., 2003). More recent studies
have associated exercise with improved brain health (e.g., in-
creased secretion of neuroprotective factors, such as brain-derived
neurotrophic factor; Zoladz et al., 2009); and improved profiles for
markers of cellular aging (e.g., telomere length and autophagy; He
et al., 2012; Ludlow et al., 2008).
Though most research focuses on potential benefits to physical
health, there is mounting evidence that exercise benefits affective
experience and cognitive performance. Researchers report reliable
associations in younger adults between exercise and increased
positive affect (see Reed & Ones, 2006 for a meta-analysis) and
reductions in negative affect and other depressive symptoms (e.g.,
Mead et al., 2009 for a meta-analysis). Moderate and vigorous
exercise bouts ranging from 5 to 30 min are associated with
improved psychological well-being and positive affective re-
sponses relative to controls (Barton & Petty, 2010; Cox, Thomas,
Hinton, & Donahue, 2006; Daley & Welch, 2004; Hansen, Ste-
vens, & Coast, 2001). In a review of 25 studies employing the
Profile of Mood States in laboratory exercise studies, results indi-
cated that exercise is typically associated with reductions in ten-
sion, anger, depression, and confusion (Berger & Motl, 2000).
Mata, Hogan, Joorman, Waugh, and Gotlib (2013) tested whether
exercise mitigates consequences of exposure to emotional stressors
in individuals recovered from major depressive disorder and
healthy control participants. Participants were randomly assigned
to an exercise or rest condition before exposure to two sad mood
inductions. Recovered depressed participants who had exercised
and healthy controls showed no increase in negative affect in
response to repeated sad mood inductions, whereas recovered
depressed participants who had not exercised reported higher
Candice L. Hogan, Department of Psychology, Stanford University;
Jutta Mata, Department of Psychology, Stanford University; Center for
Adaptive Rationality, Max Planck Institute for Human Development, Ber-
lin, Germany; Laura L. Carstensen, Department of Psychology, Stanford
University.
Research reported in this publication was supported by the National
Institute on Aging of the National Institutes of Health under Award
Number R37-AG008816 to Laura L. Carstensen. The content is solely the
responsibility of the authors and does not necessarily represent the official
views of the National Institutes of Health.
Correspondence concerning this article should be addressed to Candice
L. Hogan, Department of Psychology, Stanford University, 450 Serra Mall,
Bldg 420 Jordan Hall, Stanford, CA 94306. E-mail: cdlowder@
stanford.edu
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Psychology and Aging © 2013 American Psychological Association
2013, Vol. 28, No. 2, 587–594 0882-7974/13/$12.00 DOI: 10.1037/a0032634
587
negative affect following the second sad mood induction, suggest-
ing that exercise may serve as a protective factor against exposure
to emotional stressors. In addition, studies have found that in-
creased levels of exercise in everyday life produce increased
positive affect in healthy college students (Giacobbi, Hausenblas,
& Frye, 2005) and in young adults with major depressive disorder
(Mata et al., 2012).
Exercise is also associated with improvements in cognitive
performance in younger and older adults. Over the last several
decades, numerous studies have tested both the effects of single,
acute bouts of exercise and longer term (e.g., 3- or 6-month)
interventions. In a review of 43 studies assessing performance on
various cognitive tasks following single, acute bouts of exercise,
exercise has been linked to improvements in cognitive perfor-
mance in young adults for tasks ranging from simple reaction time
(RT) to response inhibition to creative thinking (Tomporowski,
2003). Hillman, Snook, and Jerome (2003) examined college stu-
dents’ event-related brain potentials and performance on an Erik-
sen flanker task following 30 min of treadmill exercise. They
found that exercise was related to greater P3 amplitude (an indi-
cator of allocation of cognitive resources), suggesting that acute
bouts of cardiovascular exercise may facilitate the allocation of
attentional and memory resources and thereby benefit executive
functioning (Hillman et al., 2003). A limitation of research testing
benefits to cognitive performance following exercise has been the
focus on younger adult samples. For example, Chang and col-
leagues’ (2012) meta-analysis identified 79 studies that assessed
cognitive performance in association with acute exercise, and of
these, only six studies included participants aged 60 years or older,
while 42 studies sampled participants aged 20 to 30 years. In
contrast, longer-term interventions have focused mainly on older
adults. Colcombe and colleagues (2006) randomly assigned sed-
entary older adults to participate in either vigorous, aerobic exer-
cise training or a stretching control group three times per week for
six months. Participants underwent MRI both before and after
exercise training. Following the intervention, participants in the
aerobic training group showed significant increases in both gray-
and white-matter brain regions compared with control participants.
More recent evidence suggests that volumetric differences in brain
regions may mediate cognitive performance differences, such as
task switching (Verstynen et al., 2012). In a review of 18 inter-
ventions examining the effects of aerobic training on cognitive
function in older adults, Colcombe and Kramer (2003) found a
moderate effect size (.48) for training benefits to a variety of
cognitive processes, especially executive control processes (e.g.,
working memory, inhibitory processes, multitasking).
All told, evidence for the benefits of exercise on both affective
experience and cognitive performance points to exercise as an
effective, low-cost intervention for improving both affective and
cognitive health. However, there has been some suggestion in the
literature that the effects of exercise may be weaker for older than
for younger adults. The influence of exercise on affective experi-
ence appears somewhat mixed in studies focusing on older adults,
with some studies reporting affective benefits from exercise and
others reporting reductions in positive affective states following
exercise (Arent, Landers, & Etnier, 2000; Blumenthal et al., 1989;
Focht, Knapp, Gavin, Raedeke, & Hickner, 2007; Focht, Gauvin,
& Rejeski, 2004). In addition, Ruuskanen and Ruoppila (1995)
observed that the impact of exercise on well-being was weaker for
participants aged 76 and older than for participants aged 65 to 75
years of age. A recent meta-analysis of studies on the effects of
physical activity on well-being at advanced ages also concluded
that the benefits of exercise are weakened with age, with a gradual
decrease in the degree of benefits at older ages (Netz, Wu, Becker,
& Tenenbaum, 2005). However, findings suggesting a diminishing
effect of exercise in older adults are difficult to interpret due to a
paucity of studies directly comparing effects of the same type and
duration of exercise on affective experience and cognitive perfor-
mance across age groups.
Among studies that have directly compared age groups, results
have been mixed. One 4-week daily diary study, which assessed
emotion and physical activity in younger and older adults, con-
cluded that younger adults reap more emotional benefits from light
leisure-time physical activities than older adults (Ready, Marquez,
& Akerstedt, 2009). A study of British citizens aged 18 to 94 tested
associations among self-reported walking, age, and cognitive func-
tion, and observed a significant interaction of walking and age on
simple and choice RT tasks, suggesting that walking may help to
attenuate the association between older age and slowed RT (Em-
ery, Huppert, & Schein, 1995). More recently, Whitbourne, Neu-
pert, and Lachman (2008) observed a relatively stronger reduction
in daily perceived memory failures associated with leisure-time
physical activity in older versus younger adults. However, though
diary studies offer many advantages, the reliance on self-reports
about activities in day to day life lack precision, especially regard-
ing dose (i.e., observed differences may derive from different
amounts and intensities of exercise). Kamijo and colleagues (2009)
compared the effects of light and moderate exercise in 12 older and
12 younger adults on RTs on a test of interference control (i.e.,
flanker task) and observed marginal differences such that both
older and younger adults demonstrated faster RTs following mod-
erate exercise compared with baseline. However, because this
study was conducted over three separate sessions at 12-day inter-
vals, it is unclear whether changes in flanker performance were
solely attributable to exercise. Thus, conclusions about the effec-
tiveness of exercise-based interventions remain highly tentative.
Research utilizing objective measurements of exercise, as well as
proximal measures of affective experience and cognitive perfor-
mance in association with exercise across different age groups, is
needed to reconcile mixed evidence.
To recapitulate, the body of research comparing the effects of
exercise in younger and older adults is sparse, and existing age-
comparative studies have been largely observational, mostly rely-
ing on self-report, as well as delayed measurement of affect and
cognition, such as single, end-of-day rating of affect, regardless of
timing of physical activity. The purpose of the present study is to
systematically compare the effects of a single, acute bout of
exercise compared with a nonexercise control condition, on both
affect and cognitive performance across a broad age range to
determine whether and how age may moderate these effects.
Method
Participants
Residents from the community were recruited via advertise-
ments on Internet bulletin boards and university kiosks to partic-
ipate in a lab-based study on exercise. Once participants contacted
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
588
HOGAN, MATA, AND CARSTENSEN
research staff to indicate their interest, they were screened for
eligibility using the Physical Activity Readiness Questionnaire
(PAR-Q) to assess presence of conditions that might make phys-
ical activity too risky to participate, such as heart or joint problems
(Thomas, Reading, & Shephard, 1992). Only participants who
answered “no” to all questions or received verbal permission from
his or her doctor were included in the study. The telephone version
of the Mini-Mental State Examination (MMSE) was administered
to screen for potential cognitive impairment (Newkirk et al., 2004).
Only participants scoring greater than or equal to 23 points on the
26-point scale were enrolled in the study. Participants aged 19 to
93 years (N ⫽ 144) completed the study. Participants received $20
as compensation. Means (M) and standard deviations (SD) for
demographic characteristics and descriptive measures are pre-
sented in Table 1.
Measures
Affect assessment. Ekkekakis and Petruzzello (2002) have
recommended the use of measures that capture the affective cir-
cumplex, including both high- and low-arousal positive and neg-
ative states for studies testing the affective effects of exercise;
therefore a modified version of the emotion sampler used by
Carstensen and colleagues (2011) and Carstensen, Pasupathi,
Mayr, & Nesselroade (2000) was selected for this study. Thirteen
emotion words (angry, anxious/worried, sad, fatigued, bored,
quiet, activated, enthusiastic, excited, calm, content, relaxed,
happy) representative of the affective circumplex described by
Barrett and Russell (1999) were selected for affect assessment. At
each assessment, these words appeared sequentially on a computer
screen, and participants were asked to indicate the degree to which
they were feeling each emotion in that moment on a scale ranging
from 1 (Very little or not at all)to5(Extremely). Word order was
randomized at each assessment for each participant.
This measure was administered twice during the study: (a) At
baseline (immediately after completing informed consent and be-
fore completing the 2-back task (described in the next section),
approximately 5 min before completing the exercise or control
condition) and (b) immediately following the experimental or
control condition. Composite affect scores were created by aver-
aging across low-arousal positive (LAP) words (calm, content,
relaxed; ␣ ⱖ .72) and high-arousal positive (HAP) words (acti-
vated, excited, enthusiastic; ␣ ⱖ .74). Negative affective states
showed very low variability at both measurement points (all me-
dians ⫽ 1, with the exception of fatigue after experimental ma-
nipulation, for which median ⫽ 2), and were therefore not con-
sidered further in the analyses.
Cognitive performance. Working memory was assessed us-
ing an n-back task. During this task, numbers (0 –9) are presented
one at a time on a computer screen, and participants are asked to
indicate whether each new number appearing on the screen
matches the one seen n items previously. In our study, participants
completed two blocks of a 1-back version of the task as practice in
the baseline assessment before completing four blocks of the
2-back version.
After participating in the exercise or control condition, partici-
pants completed four additional blocks of the 2-back task. Each
block was comprised of 22 trials, and stimuli were presented for
500 ms, followed by presentation of a blank screen for 2,500 ms.
Items (numbers 0 –9) for all trials were randomly generated, such
that 33% of the items in each block were targets and the remaining
67% were nontargets (Gray, 2001; Huxhold, Li, Schmiedek, &
Lindenberger, 2006; Scheibe & Blanchard-Fields, 2009). Accu-
racy and RTs were recorded for each trial.
The first two trials of each 2-back block were, by definition,
nontargets and were therefore excluded from analyses. Average
accuracy was calculated across the four blocks at baseline and
again following experimental condition. Average RTs for correct
responses were calculated as the average RT across the four blocks
for baseline and following experimental condition. Trials with RTs
of less than 100 ms were treated as inaccurate responses.
Perceived exertion. Perceived physical exertion was assessed
during exercise and control conditions using Borg’s Rating of
Perceived Exertion (RPE) scale, a 15-point scale ranging from 6 to
20 (6 ⫽ no exertion at all, 20 ⫽ maximal exertion; Borg, 1970).
Questionnaires. Participants also completed items assessing
demographic characteristics and reported typical weekly physical
activity with the Paffenbarger Physical Activity Inventory (PPAI;
Paffenbarger, Wing, & Hyde, 1978).
Procedures
Participants were assigned by stratified randomization within
gender (male or female) and age group (young: 19 –39 years,
middle age: 40 –64 years, older age: 65 years or older) to either the
exercise or control condition. Participants were asked not to eat or
smoke for two hours prior to their study session, to refrain from
exercise on the day of study, and to wear comfortable clothing and
shoes to the session.
At the start of the session, participants provided informed con-
sent and were outfitted with a Polar heart rate monitor (Polar
Electro, Inc., Oula, Finland). Resting heart rate was recorded after
three min of seated rest. Next, participants completed a baseline
affect assessment, followed by the 2-back task. After completing
these tasks, participants immediately completed either the control
or exercise condition.
Experimental condition. Stationary bicycling was used as
the mode of exercise because it is suitable for participants of all
ages. Participants were asked to pedal at a pace of 50 rpm (visible
Table 1
Demographic Characteristics of Participants, Separated by
Exercise and Control Condition
Variable Exercise; n ⫽ 71 Control; n ⫽ 73
Age (years); M (SD) 51.34 (21.74) 50.79 (19.96)
Q
1
21–29 19–28
Q
2
30–49 29–53
Q
3
50–72 54–69
Q
4
73–93 70–87
Gender (% male) 47.9 50.7
Ethnicity (% European American) 76.1 69.9
Education (% college graduates) 62 67.1
BMI; M (SD) 27.06 (6.19) 26.71 (5.41)
Physical activity (total kcal/wk);
M (SD) 2857.44 (2804.96) 2441.88 (1968.28)
Note. Q
1
⫽ first quartile; Q
2
⫽ second quartile; Q
3
⫽ third quartile; Q
4
⫽
fourth quartile; BMI ⫽ body mass index; physical activity ⫽ Paffenbarger
Physical Activity Questionnaire.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
589
EXERCISE BENEFITS AFFECT AND COGNITION
on the display screen of the bike). A warm-up phase (M ⫽ 4.21
min, SD ⫽ 1.42 min) was used to gradually increase workload to
raise each participant’s heart rate to a moderate-intensity exercise
level of approximately 50% of heart rate reserve (HRR; Karvonen,
Kentala, & Mustala, 1957). HRR is preferable to the simpler
age-predicted percent heart rate because it takes into account the
possible range of an individual’s heart rate. Participants exercised
at this level for 15 min before completing a 3-min cool-down
during which the workload was removed. Every 5 min, partici-
pants’ heart rate was recorded from the heart rate monitor and they
were asked to rate their perceived physical exertion using Borg’s
RPE scale. Moderate exercise intensity is defined as 40% to 60%
HRR (American College of Sports Medicine, 2009); average ex-
ercise intensity of participants in our study was 47% HRR (SD ⫽
14.7%). RPE scale ratings between 11 and 13 represent moderate
subjective exercise intensity; exercise participants’ average rating
in our study was 12.74 (SD ⫽ 1.35), indicating that they indeed
perceived the bout of exercise as moderate.
1
Control condition. Participants in the control condition were
asked to provide subjective picture quality ratings for 90 neutral
International Affective Picture System images (IAPS; Lang, Brad-
ley, & Cuthbert, 1997). Each image was presented for 10 s, after
which participants were given5stoprovide quality ratings on a 1
(very poor)to7(very high) scale, such that the minimum amount
of time to complete this task was 15 min and the maximum was
22.5 min (M ⫽ 17.77 min, SD ⫽ .67 min), which roughly matched
the duration of the exercise condition. As in the exercise condition,
participants’ heart rate and perceived physical exertion were re-
corded every 5 min. Average rating of perceived exertion for
control participants was 7.62 (SD ⫽ 1.81), indicating that this task
was not perceived as physically effortful.
After completing either the exercise or control condition, par-
ticipants immediately completed the affect assessment, followed
by the 2-back task. Participants completed the PPAI (Paffenbarger,
Wing, & Hyde, 1978) and demographic questions and were
thanked and debriefed before leaving the session.
Data Analysis
Independent samples t-tests and chi-square tests of contingency
were used to test whether baseline differences existed between
exercise and control participants’ demographic characteristics, in-
cluding age, body mass index (BMI), physical activity levels,
gender, education, or ethnicity. Independent samples t-tests were
also used to test for baseline differences between exercise and
control participants in all dependent measures (HAP affect, LAP
affect, 2-back accuracy, and 2-back RT). Both accuracy and RT
data from the 2-back task were skewed and therefore log-
transformed to approximate normal distribution prior to analyses.
Data from two participants (one from the exercise condition and
one from the control condition) were excluded from 2-back anal-
yses because their initial scores were ⱖ 3 SDs lower than those of
other participants. Change scores for HAP affect, LAP affect,
2-back accuracy, and 2-back RT were computed as the standard-
ized residuals obtained by regressing postmanipulation scores on
baseline scores. To test whether age moderated the effects of
exercise on HAP affect, we regressed change in HAP affect on
condition, age, and the interaction of condition and age. We
repeated these procedures for change in LAP affect, change in
2-back accuracy, and change in 2-back RT. Age was centered prior
to analyses to facilitate interpretation of interaction terms and to
reduce the possibility of multicollinearity in regression equations
as recommended by Cohen, Cohen, West, and Aiken (2003).
When significant interactions were observed, we tested the simple
slopes for the association between condition and age (Aiken &
West, 1991). Effect sizes are presented for independent samples
t-tests, Cohen’s d, calculated as (M
1
–M
2
)/SD
pooled
, and chi-square
tests, phi (), calculated as (
2
/ N)
1/2
.
Results
Demographic Characteristics of Participants
There were no baseline differences between exercise and control
participants in age, t(142) ⫽ 0.16, p ⫽ .876, d ⫽ 0.03, BMI, t(142) ⫽
0.36, p ⫽ .722, d ⫽ 0.06, physical activity levels, t(142) ⫽ 1.03,
p ⫽ .304, d ⫽ 0.17; gender,
2
(1) ⫽ .11, p ⫽ .737, ⫽⫺0.03;
education,
2
(6) ⫽ 5.50, p ⫽ .481, ⫽.20; or ethnicity,
2
(5) ⫽
8.52, p ⫽ .130, ⫽.24.
Effects of Experimental Condition and Age on
Affective Experience
Means and standard deviations for baseline and follow-up mea-
sures of affect and cognitive performance are presented in Table 2.
There were no significant differences between conditions for base-
line HAP states, t(142) ⫽⫺0.21, p ⫽ .834, d ⫽⫺0.04, or LAP
states, t(142) ⫽ 0.07, p ⫽ .944, d ⫽ 0.01. We regressed baseline
affect composite scores on age to test for initial differences of age
in affect ratings. Results indicated that age did not significantly
predict baseline HAP (b
ⴱ
⫽ 0.03, p ⫽ .750) or LAP (b
ⴱ
⫽⫺0.05,
p ⫽ .546).
In terms of HAP, we observed an effect of condition such that
exercise participants reported a greater increase in HAP than did
controls (b
ⴱ
⫽ 0.34, p ⬍ .001). We did not observe an effect of age
on change in HAP (b
ⴱ
⫽⫺0.26, p ⫽ .278). We observed an
Age ⫻ Condition interaction (b
ⴱ
⫽ 0.45, p ⫽ .046); ⌬R
2
⫽ .02. To
understand this interaction, we tested the statistical significance of
the slopes of the simple regression lines representing relations
between age and HAP using dummy coding for condition. There
was a significant positive slope for control condition (b
ⴱ
⫽ 0.36,
p ⫽ .002) but not for exercise (b
ⴱ
⫽ 0.05, p ⫽ .632), indicating
that the effect of the control condition on HAP was moderated by
age, but the effect of exercise on HAP was relatively equal across
the age range sampled (see Figure 1).
1
To test whether exercise intensity influenced affect in the current
study, we regressed change in HAP affect and LAP affect on HRR
percentage and RPE for exercise participants. There was no association
between HRR % and change in HAP affect (b
ⴱ
⫽ .08, p ⫽ .480) or LAP
affect (b
ⴱ
⫽ .08, p ⫽ .533). There was also no relation between RPE and
change in HAP affect (b
ⴱ
⫽⫺.07, p ⫽ .569). We observed an effect of
RPE on change in LAP affect such that higher perceived exertion was
associated with less LAP affect following exercise (b
ⴱ
⫽⫺.31, p ⫽ .009).
We next included RPE as a covariate in our model predicting change in
LAP by condition, age, and the interaction of condition and age; however,
RPE was not a significant predictor of change in LAP within this model
(b
ⴱ
⫽⫺.20, p ⫽ .207) and inclusion of RPE did not significantly improve
the fit of the model, F(1, 139) ⫽ 1.61, p ⫽ .207.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
590
HOGAN, MATA, AND CARSTENSEN
Testing the effect on LAP, we observed a marginal effect of
condition such that participants who exercised reported greater
decrease in LAP than those in the control condition (b
ⴱ
⫽⫺.14,
p ⫽ .075). Interestingly, we also observed a significant interaction
of age and condition (b
ⴱ
⫽ 0.55 p ⫽ .028); ⌬R
2
associated with the
interaction term ⫽ .03. Age predicted greater increase in LAP
(b
ⴱ
⫽ 0.74, p ⫽ .004). To understand the interaction, we tested the
statistical significance of the slopes of the simple regression lines
representing relations between age and LAP using dummy coding
for condition. There was a significant positive slope for exercise
(b
ⴱ
⫽ 0.38, p ⫽ .001) but not for control (b
ⴱ
⫽ 0.02, p ⫽ .863),
indicating that the effect of the exercise condition on LAP was
moderated by age, but the change in LAP associated with the
control condition was relatively equal across the age range sam-
pled (see Figure 1).
Effects of Experimental Condition and Age on
Cognitive Performance
There was no baseline difference between conditions for 2-back
accuracy, t(140) ⫽⫺0.28, p ⫽ .783, d ⫽⫺0.05, but there was a
trend for a difference in RT such that participants in the exercise
condition were initially slower than participants in the control
condition, t(140) ⫽ 1.74, p ⫽ .083, d ⫽ 0.29. To test for initial
differences of age in cognitive performance, we regressed baseline
2-back accuracy and RT on age. As expected, age was signifi-
cantly related to accuracy on the n-back task (b
ⴱ
⫽⫺0.28, p ⬍
.001) and RT (b
ⴱ
⫽ .38, p ⬍ .001), such that older age compared
with younger age predicted lower accuracy and slower responding,
respectively. We did not observe a significant effect of age on
change in accuracy (b
ⴱ
⫽⫺0.39, p ⫽ .135), of condition on
change in accuracy (b
ⴱ
⫽⫺0.09, p ⫽ .274), or a significant
interaction of age and condition on 2-back accuracy (b
ⴱ
⫽ 0.23,
p ⫽ .377; see Figure 2). Regarding change in 2-back RT, we
observed an effect of condition such that exercise was associated
with a greater reduction in RT relative to the control condition
(b
ⴱ
⫽ 0.20, p ⫽ .014). Age was not associated with change in
2-back RT (b
ⴱ
⫽ 0.10, p ⫽ .692) independent of condition, and we
did not observe a significant interaction of condition and age on
2-back RT (b
ⴱ
⫽ 0.22, p ⫽ .379), indicating that the effect of
exercise on reduction in RT was relatively consistent across the
age range sampled (see Figure 2).
Relations Between Experimental Condition, Affect,
and Cognitive Performance
Using a series of linear regression models, we tested whether the
magnitude of change in HAP or LAP affect was associated with
change in 2-back accuracy or RT as a function of experimental
condition. We did not observe any significant associations (all
Table 2
Descriptive Statistics for Baseline and Follow-Up Affect and Cognitive Performance Variables
Variable Exercise; n ⫽ 71 Control; n ⫽ 73 Full Sample; n ⫽ 144
Baseline HAP; M (SD) 2.67 (.91) 2.70 (.92) 2.68 (.91)
Follow-up HAP; M (SD) 3.19 (.85) 2.70 (.97) 2.94 (.94)
Baseline LAP; M (SD) 3.54 (.85) 3.53 (.92) 3.53 (.88)
Follow-up LAP; M (SD) 3.32 (.88) 3.52 (.76) 3.42 (.82)
Baseline 2-back accuracy; M (SD) .88 (.08) .88 (.08) .88 (.08)
Follow-up 2-back accuracy; M (SD) .90 (.09) .91 (.08) .90 (.08)
Baseline 2-back RT (ms); M (SD) 1029.54 (281.15) 947.93 (304.56) 988 (295.09)
Follow-up 2-back RT (ms); M (SD) 949.11 (279) 944.27 (333.62) 946 (306.88)
Note. HAP ⫽ high-arousal positive affect; LAP ⫽ low-arousal positive affect. Scores were transformed prior to analyses (see Data Analysis section);
RT ⫽ reaction time.
Figure 1. Effects of experimental condition and age on HAP and LAP affect. Note. Shading represents ⫾ 1
SEM. High-arousal positive (HAP) affect and low-arousal positive (LAP) affect change scores are presented as
standardized residuals obtained by regressing postmanipulation scores on baseline scores.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
591
EXERCISE BENEFITS AFFECT AND COGNITION
ps ⬎ .442) for change in HAP affect on change in 2-back accuracy
or RT, or any interactions of condition and change in HAP affect
on change in these measures of cognitive performance. We also
did not observe any significant associations (all ps ⬎. 493) for
change in LAP affect on change in 2-back accuracy or RT, or any
interactions of condition and change in LAP affect on change in
these measures of cognitive performance. Thus, data from the
present study suggest that magnitude of change in HAP or LAP
affect is not related to change in cognitive performance as a
function of condition.
2
Discussion
A growing literature demonstrates that exercise benefits both
affective experience and cognitive performance. The purpose of
this study was to test whether age may moderate these effects. Our
results suggest that a single bout of exercise appears to have
comparable and positive effects on both affective experience and
cognitive performance, independent of participants’ age. We ob-
served that a single bout of moderate exercise was associated with
increased levels of HAP affect and that the effect was consistent
across the age range sampled. Results were slightly different for
LAP affect. Younger age was associated with a drop in reported
LAP affect, whereas older age was associated with the mainte-
nance, and even slight increase in LAP, postexercise. Regarding
effects of acute exercise on cognitive performance, we found that,
independent of age, exercise resulted in significant improvement in
2-back RT compared with control participants.
In the current study, age was found to moderate the experience
of calm states following exercise, such that younger adults re-
ported decreased LAP affect, but older adults appeared to maintain
LAP even after exercise. This finding highlights the importance of
measuring the impact of exercise on both valence and arousal
components of affect. Although many studies in younger adults
observe shifts in HAP states (e.g., feelings of energy), the rela-
tionship between exercise and LAP states is less clear. To promote
more precise understanding of the relation between affect and
exercise, the current study followed recommendations from Ek-
kekakis and Petruzello (2002) by using an affective measure aimed
at capturing the full affective circumplex. These methods allowed
us to assess the influence of exercise on both high- and low-arousal
affective states. In addition, recent findings suggest that, as people
age, they may come to experience, value, and seek out low-arousal
positive states, such as feeling calm, to a greater degree than
younger adults (Scheibe, English, Tsai, & Carstensen, 2013).
Therefore, as demonstrated by the present findings, it may be
particularly important to consider both low- and high-arousal
affective changes in response to exercise to best characterize
affective effects of exercise across the life span.
In the present study, we observed that the effects of exercise
compared with a control condition on cognitive performance ap-
peared relatively stable across the age range sampled. Specifically,
we found that, independent of age, exercise resulted in significant
improvement in 2-back RT compared with control condition, and
we did not observe an effect of exercise or control condition on
change in 2-back accuracy. Although results from the present
study suggest that changes in affect and cognitive performance
observed in the exercise condition occur independently, future
research should replicate these results and further investigate po-
tential underlying mechanisms to better understand how exercise
confers psychological benefits.
It is important to acknowledge three limitations of the current
research. First, this study tested effects of one acute bout of
exercise. It remains to be shown whether and how repeated exer-
cise participation may differentially benefit people of different
ages. Although age may not moderate the effects of a single bout
of exercise on cognitive performance, further research is needed to
test whether longer term effects are comparable across the adult
2
At the suggestion of an anonymous reviewer, we also examined a
potential speed–accuracy trade-off that may have differed between the age
groups. We regressed change in 2-back RT, age, and the interaction of
2-back RT and age on change in 2-back accuracy for participants randomly
assigned to the exercise condition. We did not observe a significant
association between change in 2-back RT or change in 2-back accuracy
(b
ⴱ
⫽ 0.16, p ⫽ .235) or an interaction between change in 2-back RT and
age on 2-back accuracy (b
ⴱ
⫽ 0.005, p ⫽ .452); however, we agree with
the reviewer that future research should assess the presence of speed–
accuracy trade-offs in studies examining the effects of exercise on
cognitive performance.
Figure 2. Effects of condition and age on 2-back accuracy and reaction time. Note. Shading represents ⫾ 1
SEM. 2-back accuracy and RT change scores are presented as standardized residuals obtained by regressing
postmanipulation scores on baseline scores. Negative scores represent faster RT.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
592
HOGAN, MATA, AND CARSTENSEN
age range. Second, all study participants were in relatively good
health (e.g., did not report conditions that would make exercising
potentially dangerous, such as a heart condition); thus, it remains
unclear whether these results might generalize to those in poorer
health. A third limitation concerns the temporal separation be-
tween baseline and follow-up measures of affect and cognitive
performance for the exercise participants compared with control
participants. Despite the same overall session duration, the timing
of the affect and cognitive performance measures postexercise/
postfiller task were not exactly aligned between the exercise and
no-exercise conditions: Due to need for a warm-up phase in the
exercise condition, participants in the exercise condition started
these measurements about 4 min and 24 s later than in the control
condition, which—albeit unlikely— could have potentially af-
fected our findings.
Findings from the current study help to characterize the nature
of the effects of exercise on affective experience and cognitive
performance across the adult age range. Under controlled, exper-
imental conditions, individuals across the adult age range experi-
enced comparable benefits. The relatively large sample of individ-
uals drawn from the community and ranging in age from 19 to 93
represents an additional strength of this study and may enhance the
generalizability of our findings. Future research is needed to fur-
ther clarify how these findings may map on to exercise, affect, and
cognition in daily life. The findings add to a growing body of
research pointing to the importance of exercise for both physical
and psychological health by suggesting that, even as individuals
age, exercise remains an important contributor to psychological
health, including affective experience and cognitive performance.
References
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and
interpreting interactions. Newbury Park, CA: Sage.
American College of Sports Medicine. (2009). General principles of ex-
ercise prescription. In W. R. Thompson, N. F. Gordon, & L. S. Pesca-
tello (Eds.), ACSM’s guidelines for exercise testing and prescription (8th
ed., pp. 152–182). New York, NY: Lippincott Williams & Wilkins.
Arent, S. M., Landers, D. M., & Etnier, J. L. (2000). The effects of exercise
on mood in older adults: A meta-analytic review. Journal of Aging and
Physical Activity, 8, 407– 430.
Atha, J. (1981). Strengthening muscle. Exercise Sport Science Review, 9,
1–78.
Barrett, L. F., & Russell, J. A. (1999). Structure of current affect. Current
Directions in Psychological Science, 8, 10 –14.
Barton, J., & Petty, J. (2010). What is the best dose of nature and green
exercise for improving mental health? A multi-study analysis. Environ-
mental Science & Technology, 44, 3947–3955. doi:10.1021/es903183r
Bassey, E. J., & Ramsdale, S. J. (1994). Increase in femoral bone density
in young women following high-impact exercise. Osteoporosis Interna-
tional, 4, 72–75.
Berger, B. G., & Motl, R. W. (2000). Exercise and mood: A selective
review and synthesis of research employing the profile of mood states.
Journal of Applied Sport Psychology, 12, 69 –92. doi:10.1080/
10413200008404214
Blumenthal, J. A., Emery, C. F., Madden, D. J., George, L. K., Coleman,
R. E., Riddle, M. W.,...Williams, R. S. (1989). Cardiovascular and
behavioral effects of aerobic exercise training in healthy older men and
women. Journal of Gerontology, 44, M147–M157.
Borg, G. (1970). Perceived exertion as an indicator of somatic stress.
Scandinavian Journal of Rehabilitation Medicine, 2, 92–98.
Carstensen, L. L., Pasupathi, M., Mayr, U., & Nesselroade, J. (2000).
Emotional experience in everyday life across the adult life span. Journal
of Personality and Social Psychology, 79, 644 –655. doi:IO.1037//O022-
3514.79.4.64
Carstensen, L. L., Turan, B., Scheibe, S., Ram, N., Ersner-Hershfield, H.,
Samanez-Larkin, G. R.,...Nesselroade, J. R. (2011). Emotional
experience improves with age: Evidence based on over 10 years of
experience sampling. Psychology and Aging, 26, 21–33. doi:10.1037/
a0021285
Chang, Y. K., Labban, J. D., Gapin, J. I., & Etnier, J. L. (2012). The effects
of acute exercise on cognitive performance: A meta-analysis. Brain
Research, 1453, 87–101. doi:10.1016/j.brainres.2012.02.068
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple
regression/correlation analysis for the behavioral sciences (3rd ed.).
Mahwah, NJ: Erlbaum.
Colcombe, S. J., Erickson, K. I., Scalf, P. E., Kim, J. S., Prakash, R.,
McAuley, E.,...Kramer, A. F. (2006). Aerobic exercise training
increases brain volume in aging humans. The Journals of Gerontology:
Series A. Biological Sciences and Medical Sciences, 61, 1166 –1170.
doi:10.1093/gerona/58.2.M176
Colcombe, S. J., & 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
Cox, R. H., Thomas, T. R., Hinton, P. S., & Donahue, O. M. (2004).
Effects of acute 60% and 80% VO2max bouts of aerobic exercise on
state anxiety of women of different age groups across time. Research
Quarterly for Exercise and Sport, 75, 165–175.
Daley, A. J., & Welch, A. (2004). The effects of 15 min and 30 min of
exercise on affective responses both during and after exercise. Journal of
Sports Sciences, 22, 621– 628. doi:10.1080/02640410310001655778
Ekkekakis, P., & Petruzzello, S. J. (2002). Analysis of the affect measure-
ment conundrum in exercise psychology: IV. A conceptual case for the
affect circumplex. Psychology of Sport and Exercise, 3, 35–63. doi:
10.1016/S1469-0292(01)00028-0
Emery, C. F., Huppert, F. A., & Schein, R. L. (1995). Relationships among
age, exercise, health, and cognitive function in a British sample. The
Gerontologist, 35, 378 –385. doi:10.1093/geront/35.3.378
Focht, B. C., Gauvin, L., & Rejeski, W. J. (2004). The contribution of daily
experiences and acute exercise to fluctuations in daily feeling states
among older, obese adults with knee osteoarthritis. Journal of Behav-
ioral Medicine, 27, 101–121. doi:10.1023/B:JOBM.0000019847
.80315.4d
Focht, B. C., Knapp, D. J., Gavin, T. P., Raedeke, T. D., & Hickner, R. C.
(2007). Affective and self-efficacy responses to acute aerobic exercise in
sedentary older and younger adults. Journal of Aging and Physical
Activity, 15, 123–138.
Giacobbi, P. R., Hausenblas, H. A., & Frye, N. (2005). A naturalistic
assessment of the relationship between personality, daily life events,
leisure-time exercise, and mood. Psychology of Sport and Exercise, 6,
67– 81. doi:10.1016/J.Psychsport.2004.10.009
Gray, J. R. (2001). Emotional modulation of cognitive control: Approach–
withdrawal states double dissociate spatial from verbal two-back task
performance. Journal of Experimental Psychology: General, 130, 436 –
452. doi:10.1037/0096-3445.130.3.436
Hansen, C. J., Stevens, L. C., & Coast, J. R. (2001). Exercise duration and
mood state: How much is enough to feel better? Health Psychology, 20,
267–275. doi:10.1037/0278-6133.20.4.267
He, C., Bassik, M. C., Moresi, V., Sun, K., Wei, Y., Zou, Z.,...Levine,
B. (2012). Exercise-induced BCL2-regulated autophagy is required for
muscle glucose homeostasis. Nature, 481, 511–515. doi:10.1038/
nature10758
Hickson, R. C., Bomze, H. A., & Holloszy, J. O. (1977). Linear increase in
aerobic power induced by strenuous program of endurance exercise.
Journal of Applied Physiology, 42, 372–376.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
593
EXERCISE BENEFITS AFFECT AND COGNITION
Hillman, C. H., Snook, E. M., & Jerome, G. J. (2003). Acute cardiovas-
cular exercise and executive control function. International Journal of
Psychophysiology, 48, 307–314. doi:10.1016/S0167-8760(03)00080-1
Huxhold, O., Li, S. C., Schmiedek, F., & Lindenberger, U. (2006). Dual-
tasking postural control: Aging and the effects of cognitive demand in
conjunction with focus of attention. Brain Research Bulletin, 69, 294 –
305. doi:10.1016/j.brainresbull.2006.01.002
Judge, J. O., Lindsey, C., Underwood, M., & Winsemius, D. (1993).
Balance improvement in older women: Effects of exercise training.
Physical Therapy, 73, 254 –262.
Kamijo, K., Hayashi, Y., Sakai, T., Yahiro, T., Tanaka, K., & Nishihira, Y.
(2009). Acute effects of aerobic exercise on cognitive function in older
adults. The Journals of Gerontology Series B: Psychological Sciences
and Social Sciences, 64, 356 –363. doi:10.1093/geronb/gbp030
Karvonen, M. J., Kentala, E., & Mustala, O. (1957). The effects of training
on the heart rate: A longitudinal study. Annales Medicinae Experimen-
talis et Biologiae Fenniae, 35, 307–315.
Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1999). International
affective picture system (IAPS): Technical manual and affective ratings.
Gainsville, FL: The Center for Research in Psychophysiology, Univer-
sity of Florida.
Ludlow, A. T., Zimmerman, J. B., Witkowski, S., Hearn, J. W., Hatfield,
B. D., & Roth, S. M. (2008). Relationship between physical activity
level, telomere length, and telomerase activity. Medicine and Science in
Sports and Exercise, 40, 1764 –1771. doi:10.1249/MSS.0b013e
31817c92aa
Mata, J., Hogan, C. L., Joorman, J., Waugh, C. E., & Gotlib, I. H. (2013).
Acute exercise attenuates negative affect following repeated sad mood
inductions in persons who have recovered from depression. Journal of
Abnormal Psychology, 122, 45–50. doi:10.1037/a0029881
Mata, J., Thompson, R. J., Jaeggi, S., Buschkuehl, M., Jonides, J., &
Gotlib, I. H. (2012). Walk on the bright side: Physical activity and affect
in major depressive disorder. Journal of Abnormal Psychology, 121,
297–308. doi:10.1037/a0023533
Mead, G. E., Morley, W., Campbell, P., Greig, C. A., McMurdo, M., &
Lawlor, D. A. (2009). Exercise for depression. Cochrane Database of
Systematic Reviews, 3. doi:10.1002/14651858
Netz, Y., Wu, M. J., Becker, B. J., & Tenenbaum, G. (2005). Physical
activity and psychological well-being in advanced age: A meta-analysis
of intervention studies. Psychology and Aging, 20, 272–284. doi:
10.1037/0882-7974.20.2.272
Newkirk, L. A., Kim, J. M., Thompson, J. M., Tinklenberg, J. R., Yesav-
age, J. A., & Taylor, J. L. (2004). Validation of a 26-point telephone
version of the Mini-Mental State Examination. Journal of Geriatric
Psychiatry and Neurology, 17, 81– 87. doi:10.1177/0891988704264534
Paffenbarger, R. S., Wing, A. L., & Hyde, R. T. (1978). Physical activity
as an index of heart attack risk in college alumni. American Journal of
Epidemiology, 108, 161–175.
Ready, R. E., Marquez, D. X., & Akerstedt, A. (2009). Emotion in younger
and older adults: Retrospective and prospective associations with sleep
and physical activity. Experimental Aging Research, 35, 348 –368. doi:
10.1080/03610730902922184
Reed, J., & Ones, D. S. (2006). The effect of acute aerobic exercise on
positive activated affect: A meta-analysis. Psychology of Sport and
Exercise, 7, 477–514. doi:10.1016/j.psychsport.2005.11.003
Ruuskanen, J. M., & Ruoppila, I. (1995). Physical activity and psycholog-
ical well-being among people aged 65 to 84 years. Age and Ageing, 24,
292–296. doi:10.1093/ageing/24.4.292
Scheibe, S., & Blanchard-Fields, F. (2009). Effects of regulating emotions
on cognitive performance: What is costly for young adults is not so
costly for older adults. Psychology and Aging, 24, 217–223. doi:
10.1037/a0013807
Scheibe, S., English, T., Tsai, J. L., & Carstensen, L. L. (2013). Striving to
feel good: Ideal affect, actual affect, and their correspondence across
adulthood. Psychology and Aging, 28, 160 –171. doi:
10.1037/a0030561
Thomas, S., Reading, J., & Shephard, R. J. (1992). Revision of the physical
activity readiness questionnaire (PAR-Q). Canadian Journal of Sport
Sciences, 17, 338 –345.
Thompson, P. D., Buchner, D., Pina, I. L., Balady, G. J., Williams, M. A.,
Marcus, B.H...Wenger, N. K. (2003). Exercise and physical activity
in the prevention and treatment of atherosclerotic cardiovascular disease:
A statement from the Council on Clinical Cardiology (Subcommittee on
Exercise, Rehabilitation, and Prevention) and the Council on Nutrition,
Physical Activity, and Metabolism (Subcommittee on Physical Activ-
ity). Circulation, 107, 3109 –3116. doi:10.1161/01.CIR.0000075572
.40158.77
Tomporowski, P. D. (2003). Effects of acute bouts of exercise on cogni-
tion. Acta Psychologica, 112, 297–324. doi:10.1016/S0001-
6918(02)00134-8
Verstynen, T. D., Lynch, B., Miller, D. L., Voss, M. W., Prakash, R. S.,
Chaddock, L.,...Erickson, K. I. (2012). Caudate nucleus volume
mediates the link between cardiorespiratory fitness and cognitive flexi-
bility in older adults. Journal of Aging Research, 2012, 1–11. doi:
10.1155/2012/939285
Whitbourne, S. B., Neupert, S. D., & Lachman, M. E. (2008). Daily
physical activity: Relation to everyday memory in adulthood. Journal of
Applied Gerontology, 27, 331–349. doi:10.1177/0733464807312175
Zoladz, J. A., Pilc, A., Majerczak, J., Grandys, M., Zapart-Bukowska, J., &
Duda, K. (2008). Endurance training increases plasma brain-derived
neurotrophic factor concentration in young healthy men. Journal of
Physiology and Pharmacology, 59, 119 –132.
Received December 4, 2012
Revision received March 11, 2013
Accepted March 11, 2013 䡲
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
594
HOGAN, MATA, AND CARSTENSEN
A preview of this full-text is provided by American Psychological Association.
Content available from Psychology and Aging
This content is subject to copyright. Terms and conditions apply.
