Elsevier

Brain Research

Volume 1453, 9 May 2012, Pages 87-101
Brain Research

Review
The effects of acute exercise on cognitive performance: A meta-analysis

https://doi.org/10.1016/j.brainres.2012.02.068Get rights and content

Abstract

There is a substantial body of literature related to the effects of a single session of exercise on cognitive performance. The premise underlying this research is that physiological changes in response to exercise have implications for cognitive function. This literature has been reviewed both narratively and meta-analytically and, although the research findings are mixed, researchers have generally concluded that there is a small positive effect. The purpose of this meta-analysis was to provide an updated comprehensive analysis of the extant literature on acute exercise and cognitive performance and to explore the effects of moderators that have implications for mechanisms of the effects. Searches of electronic databases and examinations of reference lists from relevant studies resulted in 79 studies meeting inclusion criteria. Consistent with past findings, analyses indicated that the overall effect was positive and small (g = 0.097 n = 1034). Positive and small effects were also found in all three acute exercise paradigms: during exercise (g = 0.101; 95% confidence interval [CI]; 0.041–0.160), immediately following exercise (g = 0.108; 95% CI; 0.069–0.147), and after a delay (g = 0.103; 95% CI; 0.035–0.170). Examination of potential moderators indicated that exercise duration, exercise intensity, type of cognitive performance assessed, and participant fitness were significant moderators. In conclusion, the effects of acute exercise on cognitive performance are generally small; however, larger effects are possible for particular cognitive outcomes and when specific exercise parameters are used.

Introduction

A growing body of research has been designed to further our understanding of how a single bout of exercise (also referred to as acute exercise) affects cognitive performance. This research is based upon the premise that physiological responses to exercise have an impact on cognitive functioning which can be assessed using behavioral measures. The physiological responses that have been implicated in the cognitive literature include changes in heart rate (Allard et al., 1989, ⁎Davranche et al., 2005, ⁎Davranche et al., 2006, ⁎Hillman et al., 2003, ⁎Kamijo et al., 2004a, ⁎Kamijo et al., 2004b, McMorris and Graydon, 2000), levels of brain-derived neurotrophic factor (Ferris et al., 2007, Winter et al., 2007), and changes in plasma catecholamines (Chmura et al., 1994). Researchers have typically assessed the effects of acute exercise on cognition using one of three exercise paradigms: maximal intensity exercise, submaximal intensity exercise, or exercise in conjunction with hydration status. In a narrative review, Tomporowski and Ellis (1986) described this literature as being limited by a lack of consistency in methodology and by a failure to use theory-based approaches. Since then, this literature has grown and has been reviewed narratively on several occasions (Brisswalter et al., 2002, McMorris and Graydon, 2000, Tomporowski, 2003a, Tomporowski, 2003b). Although individual empirical studies have yielded inconsistent results, the consensus of the narrative reviewers is that there is a positive effect of acute exercise on cognitive performance (Brisswalter et al., 2002, McMorris and Graydon, 2000, Tomporowski, 2003a, Tomporowski, 2003b). However, given the heterogeneity of the findings, the use of meta-analytic techniques is warranted to statistically summarize empirical findings. Meta-analysis also allows for the testing of moderating variables that may yield information regarding potential mechanisms of the effects.

The effects of acute exercise have been examined meta-analytically in three previous reviews (Etnier et al., 1997, Lambourne and Tomporowski, 2010, Sibley and Etnier, 2003). Etnier et al. (1997) indicated that acute exercise has a significant small positive effect on cognitive performance (ES = 0.16). Sibley and Etnier (2003) limited their review to studies testing the effects of acute exercise on cognitive performance in children and also reported a significant small effect (ES = 0.37). However, these earlier meta-analyses have two limitations. First, numerous empirical studies on acute exercise and cognition have been conducted since the publication of these earlier reviews. Second, neither review was designed specifically to examine acute exercise; thus, moderators particularly relevant to the effects of acute exercise on cognitive performance were not examined. Thus, these early meta-analyses provided important direction for future research by establishing that acute exercise has reliable small effects on cognitive performance, but the failure to test relevant moderators means that conclusions relative to potential mechanisms of the effects could not be drawn.

More recently, Lambourne and Tomporowski (2010) conducted a meta-analytic review specifically focused on acute exercise studies (n = 40), in which they included studies testing the effects in healthy young adults and measuring cognitive performance prior to exercise and then either during exercise (n = 21) or following exercise (n = 29). Their results indicated that exercise had a detrimental effect on cognitive performance during exercise (d = -0.14), but improved cognitive performance after exercise (d = 0.20). Moderator variables particularly relevant to acute exercise were examined and included exercise intensity and duration, timing of the cognitive task administration, exercise mode, cognitive task type, and study design. Findings relevant to these moderators could provide insights as to the potential mechanisms of the effect and guidance for exercise recommendations to garner the largest cognitive benefits. However, the scope of the review was limited by the decisions to focus on young adults and to only include studies using “within-subjects, repeated measures” (Lambourne and Tomporowski, 2010, p. 14) designs. These decisions clearly impact the generalizability of the conclusions and result in the exclusion of numerous relevant studies that might contribute to our understanding of the effects of acute exercise on cognitive performance.

Thus, the purpose of this meta-analysis is to provide a more comprehensive review of the extant literature on acute exercise and cognitive performance. The use of less restrictive inclusion criteria allows for the testing of additional relevant moderators and provides a more broad review of the literature. In addition, by including participants of all ages, the effects of acute exercise for children and older adults can be explored. This may be particularly relevant because past reviews have suggested that larger effects from chronic exercise can be observed for these age groups (Angevaren et al., 2008, Colcombe and Kramer, 2003, Etnier et al., 1997, Sibley and Etnier, 2003) and it is not known if this pattern of results would also apply to acute exercise.

Prior to testing the effects of moderators, studies in this analysis were separated into subsets based upon the three particular paradigms that were used (during exercise, immediately following exercise, and after a delay of longer than 1 min). This decision was based upon the fact that the potential underlying physiological mechanisms are affected directly by exercise and that the effects dissipate following exercise cessation. Thus, it is expected that findings for cognitive performance would differ for studies testing the effects of acute exercise on cognitive performance during exercise as compared to those testing the effects at various time points following exercise (Lambourne and Tomporowski, 2010). Four primary moderators were then examined for studies within each of these paradigms. The primary moderators were selected based upon the fact that they have been previously identified as important variables to consider when examining the effects of physical activity on cognitive performance and included exercise intensity, timing of test administration relative to exercise, cognitive task type, and initial fitness level of the participants.

Exercise intensity is a moderator that has frequently been considered in acute exercise studies. The frequent attention given to exercise intensity is due to its relevance to understanding mechanisms of the effects. The inverted-U hypothesis and drive theories both suggest that exercise intensity will influence the size of the effect. In particular, the inverted-U hypothesis predicts that moderate intensity exercise will have the greatest benefits while the drive theories suggest that the largest effects will be observed at high intensity. Clearly when considering mechanisms such as heart rate, catecholamines, and brain-derived neurotrophic factor (BDNF), the intensity level of the exercise is important for determining the amount of change in these physiological mechanisms that will be achieved and this may then be important for predicting the behavioral effects as well. As an example, studies testing the effects of acute exercise on circulating BDNF indicate that high intensity protocols result in larger increases than do low-intensity protocols (Knaepen et al., 2010). Thus, if BDNF is a mediator of the effects of acute exercise on cognitive performance, intensity would be expected to influence behavioral outcomes.

The specific timing of the cognitive test administration is a second moderator of interest. This moderator has been found to influence the effect of acute exercise on cognitive performance by young healthy adults (Lambourne and Tomporowski, 2010). This moderator also has implications for mechanisms of the effects because of the specific way in which the mechanisms are impacted by exercise. For example, exercise has transient effects on BDNF, therefore the timing of the cognitive assessment may be critical in terms of the effects mediated by BDNF.

The relationship between acute exercise and cognitive performance might also be dependent upon the nature of the cognitive task. Etnier et al. (1997) reported that acute exercise had large beneficial effects on motor skills, academic achievement, and when tested using a composite from a variety of tests (ES ranging from 1.20 to 1.47); however, it had negative effects on tasks related to reasoning and verbal skills (ES ranging from − 0.06 to − 0.02). At the time of Etnier et al.'s review, many acute exercise studies used reaction time tasks (Fleury and Bard, 1987, Hogervorst et al., 1996, McMorris, 1995, McMorris and Keen, 1994, Travlos and Marisi, 1995) and visual recognition tasks (Bard and Fleury, 1978, Fleury et al., 1981). Recently, researchers have begun to examine the effects of acute exercise on executive function or frontal-lobe dependent measures (Chang and Etnier, 2009a, Chang and Etnier, 2009b, Dietrich and Sparling, 2004, Sibley et al., 2006, Tomporowski et al., 2005). This may reflect interest in testing the transient hypofrontality hypothesis (Dietrich, 2006) and past meta-analytic evidence suggesting that cognitive task type moderates the effects of both acute and chronic exercise (Angevaren et al., 2008, Colcombe and Kramer, 2003, Etnier et al., 1997, Lambourne and Tomporowski, 2010).

A final primary moderator that is of interest is the initial fitness level of the participants (see Chodzko-Zajko, 1991, Tomporowski, 2003b, Tomporowski and Ellis, 1986). Again, this relates to the potential mechanisms of the effects. For example, some evidence suggests that the BDNF response to acute exercise is dependent upon the participants' level of training (Castellano and White, 2008, Schulz et al., 2004, Zoladz et al., 2008). Thus, if BDNF is a mediator, one might expect fitness level to moderate the behavioral effects of acute exercise.

Thus, this meta-analysis is designed to extend beyond the existing meta-analytic reviews by using broader inclusion criteria than have been used previously and by including moderators that have not been previously examined and that have implications for mechanisms of the effects. Based upon the findings of previous meta-analyses (Etnier et al., 1997, Lambourne and Tomporowski, 2010, Sibley and Etnier, 2003), it is hypothesized that acute exercise will have a significant small beneficial effect on cognitive task performance after exercise, but will negatively affect cognitive performance during exercise. We expect the effects of acute exercise on cognitive performance to be moderated by exercise intensity, timing of the test administration, cognitive task type, and fitness levels.

Section snippets

Description of studies

A total of 79 studies and 1034 effect sizes were included in the meta-analytic review. This represented data from 2072 subjects. The average age of the samples was reported in 61 studies and was 28.51 (SD = 17.21) with most effects coming from studies testing young adults (20–30 years, n = 42 studies) and fewer testing the effects on children (5–20 years, n = 9 studies), adults (30–60 years, n = 4 studies), and older adults (> 60 years, n = 6 studies). Effects were calculated from samples consisting only of

Discussion

When summarized meta-analytically, results from 79 studies indicate that a single bout of exercise has a small positive effect (d = 0.097) on cognitive performance that is significantly different from zero. This is consistent with conclusions drawn by narrative reviewers (Brisswalter et al., 2002, McMorris and Graydon, 2000, Tomporowski, 2003a, Tomporowski, 2003b) and with the small positive effect reported by Etnier et al. (1997). The first moderator examined in this review was the acute

Data collection

Several steps were taken to obtain all possible data relevant to the effects of acute exercise on cognitive performance. First, studies identified from previous reviews (Brisswalter et al., 2002, Etnier et al., 1997, McMorris and Graydon, 2000, Tomporowski, 2003b, Tomporowski and Ellis, 1986) were considered for inclusion. Second, computerized searches of the electronic data bases of Sports Discus, Psych Info, Pub Med, ERIC, and High Wire were conducted. Searches were conducted using the

Acknowledgments

We acknowledge the Office of Research for the School of Health and Human Science at the University of North Carolina at Greensboro for the provision of funding to support the completion of this study.

References3 (48)

  • P.D. Tomporowski

    Effects of acute bouts of exercise on cognition

    Acta Psychol.

    (2003)
  • B. *Winter et al.

    High impact running improves learning

    Neurobiol. Learn. Mem.

    (2007)
  • F. Allard et al.

    The effect of exercise on visual attention performance

    Hum. Perform.

    (1989)
  • American College of Sports Medicine

    ACSM's Guidelines for Exercise Testing and Prescription

    (2010)
  • M. Angevaren et al.

    Physical activity and enhanced fitness to improve cognitive function in older people without known cognitive impairment

    Cochrane Database Syst. Rev.

    (2008)
  • C. *Bard et al.

    Influence of imposed metabolic fatigue on visual capacity components

    Percept. Mot. Skills

    (1978)
  • J. Brisswalter et al.

    Effects of acute physical exercise characteristics on cognitive performance

    Sports Med.

    (2002)
  • Y.K. Chang et al.

    Exploring the dose–response relationship between resistance exercise intensity and cognitive function

    J. Sport Exerc. Psychol.

    (2009)
  • J. *Chmura et al.

    Choice reaction time during graded exercise in relation to blood lactate and plasma catecholamine thresholds

    Int. J. Sports Med.

    (1994)
  • W.J. Chodzko-Zajko

    Physical fitness, cognitive performance, and aging

    Med. Sci. Sports Exerc.

    (1991)
  • C. Cian et al.

    Influences of variations in body hydration on cognitive function: effects of hyperhydration, heat stress, and exercise-induced dehydration

    J. Psychophysiol.

    (2000)
  • S. Colcombe et al.

    Fitness effects on the cognitive function of older adults: a meta-analytic study

    Psychol. Sci.

    (2003)
  • K. *Davranche et al.

    Information processing during physical exercise: a chronometric and electromyographic study

    Exp. Brain Res.

    (2005)
  • Cited by (1261)

    View all citing articles on Scopus
    1

    Present address: Graduate Institute of Coaching Science, National Taiwan Sport University, No.250, Wenhua 1st Rd., Guishan Shiang, Taoyuan County 333, Taiwan, ROC.

    2

    Present address: Department of Kinesiology and Health Education, Southern Illinois University Edwardsville, Edwardsville, IL 62026, USA.

    3

    References marked with an asterisk indicate studies included in the meta-analysis.

    View full text