Aerobic Exercise Effects on Cognition: A Functional Near Infrared Spectroscopy Systematic Review

• ¹ Temple University, Psychoogy, United States
• ² Spaulding Rehabilitation Hospital, Department of Physical Medicine and Rehabilitation, United States

Introduction: Past studies demonstrate that exercise can improve cognitive and neural functioning. Functional Near-Infrared Spectroscopy (fNIRS) is a cost-effective, novel tool to examine brain activity with several advantages, including superior temporal resolution to functional magnetic resonance imaging (fMRI), superior spatial resolution to electroencephalography (EEG), and is capable of handling motion artifacts allowing for ambulatory applications. Design: The present study is a systematic review designed to describe the effects of aerobic activity on cognitive performance. A PubMed search was conducted using the terms “Functional Near-Infrared Spectroscopy” AND “Physical Activity.” All search results were then manually screened. Studies must have included an aerobic intervention or physical fitness level assessment as well as a cognitive task correlated with fNIRS activity. Results: Included Studies: 107 studies were screened, and nine were included in the systematic review. Reasons of exclusion are documented in flowchart (Figure1) using the PRISMA guides (Moher et al., 2009). A summary of study characteristics can be seen in Table 1. Six studies (3 young adult samples and 3 older adult samples) included an acute aerobic exercise session using an ergometer (Byun et al., 2014; Decroix et al., 2016; Hyodo et al., 2012, 2016; Tsujii, Komatsu, & Sakatani, 2013; Yamazaki et al., 2017). One study used actigraphy to compare more and less active children (Mücke, Andrä, Gerber, Pühse, & Ludyga, 2017) . Another looked at a weight-loss intervention program (that included aerobic exercise) for overweight and obese adolescents and young adults (Xu et al., 2017). Finally, a randomized controlled trial investigated the effects of an 8-week aerobic “dance” video game training intervention compared to a control “balance” intervention group, involving stretching and balance training (Eggenberger, Wolf, Schumann, & de Bruin, 2016). Findings Summarized: Inhibitory Control: Six studies (1 adolescent and young adult, 2 adult, and 3 older adult samples) measured performance on the Stroop task, an executive function task of inhibition. Exercise appears to increase left prefrontal oxygenated hemoglobin (oxy-HB) levels in healthy younger adults (Byun et al., 2014; Decroix et al., 2016) and older adults with higher physical activity levels (Hyodo et al., 2016). However, in general older adult populations, contralateral brain activity (on the right side) may serve to compensate for deterioration in left brain regions, such as the left dorsolateral prefrontal cortex (DLPFC) (Hyodo et al., 2012). In an adolescent and young adult sample, increased oxy-HB in brain regions within the left hemisphere (as well as bilateral DLPC) during the Stroop task was positively correlated with weight loss (Xu et al., 2017). Exercise also improved reaction times during Stroop interference conditions for all studies that measured differences pre- and post-exercise (Byun et al., 2014; Decroix et al., 2016). Working Memory: Two studies examined working memory as an outcome. A study in older adults (Tsujii et al., 2013) found an increase in reaction time in a working memory task and increased oxy-HB levels in the left prefrontal cortex during an exercise condition compared with the control condition. A study in younger adults (Yamazaki et al., 2017) found that exercise improved spatial working memory on easier but not more difficult trials. There was no difference in brain activity during the cognitive trials; however, during exercise, those who did improve on the easier spatial working memory trials (responders) had greater activation in the right ventrolateral prefrontal cortex compared to those who did not improve (non-responders). Given the differences in analysis, fNIRS parameters, study population, and intensity levels of these two studies, no global generalizations can be concluded. Neither of these studies reported effect sizes. Other cognitive outcomes: One study comparing physically active children and non-physically active children found no differences in performance during a semantic verbal fluency test (VFT), a phonetic VFT, and a mental arithmetic task. In addition, there were no corresponding brain activation (oxy-HB levels) differences between physically active and non-physically active children (P ≥ 0.268, η2 ≤ 0.02). Conclusion: fNIRS technology shows that acute exercise can influence cognitive performance and is related to increases in prefrontal activation. Specifically, inhibitory tasks (i.e., the Stroop task) appear to be related to increases in left prefrontal activity for younger and more active older adults but right-compensatory activation in older adults. Effects sizes of exercise on cognitive task performance were not reported for any of the studies, however, one reported an effect size (ES) of .66 when looking at brain activity increases after exercise compared to baseline, when averaged across subjects that took a supplement and those who took a placebo (Decroix et al., 2016). Another study found a significant positive correlation (r=.43) between improved Stroop performance and elevated right prefrontal activity in older adults (Eggenberger et al., 2016), whereas a mediation analysis showed, left-lateralized DLPFC mediated the relationship between physical fitness level and Stroop interference time in an older adult sample (R2= 0.289, p < 0.001) (Hyodo et al., 2016). More studies using similar protocols that report effect sizes are needed to quantify the strength of the relationship between exercise and cognition as well as to confirm the current findings.