Moderate aerobic exercise, but not anticipation of exercise, improves cognitive control.

BACKGROUND: Evidence suggests a single bout of exercise can improve cognitive control. However, many studies only include assessments after exercise. It is unclear whether exercise changes as a result, or in anticipation, of exercise.
OBJECTIVE: To examine changes in cognitive control due to moderate aerobic exercise, and anticipation of such exercise.
METHODS: Thirty-one young healthy adults (mean age 22 years; 55% women) completed three conditions (randomized order): 1) exercise (participants anticipated and completed exercise); 2) anticipation (participants anticipated exercise but completed rest); and 3) rest (participants anticipated and completed rest). Cognitive control was assessed with a modified Flanker task at three timepoints: (1) early (20 min pre-intervention, pre-reveal in anticipation session); (2) pre-intervention (after reveal); and (3) post-intervention. An accuracy-weighted response time (RTLISAS) was the primary outcome, analyzed with a linear mixed effects modeling approach.
RESULTS: There was an interaction between condition and time (p = 0.003) and between session and time (p = 0.015). RTLISAS was better post-exercise than post-rest and post-deception, but was similar across conditions at other timepoints. RTLISAS improved across time in session 1 and session 2, but did not improve over time in session 3. There were also main effects of condition (p = 0.024), session (p = 0.005), time (p<0.001), and congruency (p<0.001).
CONCLUSIONS: Cognitive control improved after moderate aerobic exercise, but not in anticipation of exercise. Improvements on a Flanker task were also observed across sessions and time, indicative of a learning effect that should be considered in study design and analyses.

RCT Entities:   Population Interventions Outcomes

1 Introduction

Cognitive control is critical across the life span as it contributes to all aspects of daily life, including occupational, functional, and social activities. Research to improve cognitive control has generally focused on early life (as part of academic performance) and older adults (to reduce age-associated cognitive decline) [1-3]. Understanding of interventions to improve cognitive functions across the life course could impact rates of late-life dementia, by providing additional cognitive reserve capacity before function is affected.

A growing body research has examined the influence of physical exercise on cognitive control [4] with exercise now being considered a viable low cost option for maintaining neurological health as we age [5]. Optimistically, habitual physical activity (as reported by questionnaires) and aerobic fitness (as measured by maximal exercise tests) are both associated with better cognitive control [6-8]. Moreover, cognitive improvements and brain changes including increase in volume in areas of the brain (e.g., hippocampus) and connectivity of some brain networks can be observed after exercise interventions [4,5]. Though evidence from clinical trials of exercise is generally more inconsistent than observational studies [7,9-11], the most inclusive and most recent meta-analyses of randomized clinical trials generally conclude that exercise training (including aerobic, resistance, or mind-body exercise) is associated with better cognitive outcomes, though there is not yet support for prevention of dementia from these trials [12-19].

Improvements in cognitive control can also be observed after a single session of aerobic exercise [20,21]. Though recent meta-analyses indicate that acute exercise elicits a small positive effect on cognitive control [20,21], there were a number of factors that moderated the magnitude of the effects [20,21]. Potential moderators may include, but are not limited to, cognitive domain and task, exercise intensity and duration, and experimental design [20].

The cognitive benefits of exercise have been observed most consistently for executive functions. Executive functions refer to higher order cognitive processes that control basic cognitive processes for the purpose of goal-directed actions, including our ability to shift, inhibit, or update during a task [22-24]. Cognitive (or executive) control describes brain processes that guide goal-directed thoughts and behaviour. Cognitive control refers to the ability to focus on goal-related information and ignore irrelevant information and inhibit automatic responses. Cognitive control can be assessed using a modified Eriksen flanker task [25,26]. A number of studies have used a Flanker task to assess changes in cognitive control with exercise [5,6,27-29]. Arousal levels may influence cognitive performance across several cognitive domains, with evidence for changes in attention [30], working memory [31,32], and long-term memory [33-35]. In turn, arousal may change both due to exercise, as well as the anticipation of exercise; in turn, these changes in arousal may influence cognitive control. Anticipation of exercise, less studied in relation to cognitive control, is accompanied by a rise in ventilation [36] and increases in plasma cortisol and norepinephrine levels in some individuals [37]. Blood pressure and systemic vascular resistance were also observed to increase prior to a handgrip test, indicating sympathetic nervous system activation [38]. These studies suggest an increased arousal prior to exercise, which could improve cognitive control pre-exercise [28,39], and at least partially account for the cognitive benefits observed with exercise.

Some study designs examining the effects of exercise on cognitive control would be vulnerable to confounding by pre-exercise arousal. Most acute studies assessing changes in executive functions have not included a pre-exercise assessment so would be unable to separate the cognitive effects of physical exercise from those related to the anticipation of exercise [29,40-45]. Even when pre and post-assessments were included, many existing studies failed to blind participants to the order of the sessions [29,40-45]. As a result, pre-intervention arousal might be different before the exercise intervention than before control sessions, confounding results.

The objective of this study was to examine the influence of the anticipation of exercise and physical exercise on Flanker task performance. We hypothesized that the anticipation of exercise and exercise itself would both improve cognitive control, as reflected by improved performance on the Flanker task. Additionally, we expected that the effect of exercise itself would be larger than the effect of anticipating exercise. To overcome the limitations of previous studies, this study used a deception condition to separate the anticipation of exercise from physical exercise itself. In addition, we include both pre- and post-intervention cognitive assessments (Flanker task).

2 Materials and methods

2.1 Participants

Thirty-one young healthy adults (aged 18–35 years) were recruited to the study by email and via posters placed around the University of Waterloo campus. Recruitment was active from July 2018 to April 2019. Participants were screened with the Physical Activity Readiness Questionnaire (PAR-Q+) to ensure safety to exercise [46]. Participants were excluded if they had: a history of heart disease (heart attack or operation, heart murmur, coronary artery disease, congenital heart disease, pacemaker); uncontrolled or hypertension; drop in blood pressure when rising from a seated position; neurological conditions (e.g. stroke, epilepsy, Parkinson’s disease, or dementia); were taking beta blockers, anticoagulants, or anticholinergics; had chronic obstructive pulmonary disease; or had musculoskeletal impairments that cause more pain during exercise than tolerable. This study was approved by a University of Waterloo research ethics committee (ORE# 31106). All participants provided written informed consent.

2.2 Study design

This study used a repeated measures design to examine the influence of three conditions (exercise, exercise anticipation, and rest) on cognitive control. In the exercise anticipation session, participants came to the session expecting to exercise, but the researchers revealed they would instead rest after the first cognitive assessment. In the exercise session, participants expected to and did exercise. In the rest condition, participants expected to and did rest. Sessions occurred at least a week apart to lessen learning effects. The order of these conditions was counterbalanced. All participants were asked to refrain from caffeine 4 hours prior to the beginning of all study sessions.

2.3 Experimental protocol

In the first session, participants reported demographics and medical history, and completed the International Physical Activity Questionnaire (IPAQ) [47]. In all sessions, participants then started the experimental protocol with a practice block of 50 trials of a modified Flanker task (described below). Subsequently, cognitive control was assessed with 200 trials of the modified Flanker task at three timepoints: (1) early (pre-reveal in the exercise anticipation condition); (2) pre-intervention (post-reveal in the exercise anticipation condition, but pre-exercise/rest; 20 min after the early assessment); and (3) post-intervention. The experimental protocol is shown in Fig 1.

Fig 1

Experimental design.

The three conditions (exercise, rest, and anticipation) were in randomized order.

During the exercise session, participants performed approximately 3 minutes of acclimatization on the bike followed by 20min of moderate intensity aerobic exercise on a stationary cycle ergometer and 3 minutes of self-paced cool down. Participants were instructed to maintain a cadence of 90±5 rpm. Cycle ergometer resistance was adjusted to maintain a rating of perceived exertion (RPE) of 13 measured on the Borg scale of perceived exertion (6–20 version) [48], consistent with moderate intensity exercise. RPE and resistance (watts) were recorded every 2 minutes. In both the exercise anticipation and rest sessions, participants completed 26 minutes of quiet rest on the stationary cycle ergometer, with only the anticipation of exercise (and reveal) varying between sessions. Heart rate was recorded every 2 minutes during all interventions using a Polar heart rate monitor.

2.4 Measures

The primary outcome was a modified Eriksen Flanker task [25], used to probe the evaluation of cognitive control. The modified Flanker Task consisted of five arrow heads displayed on the screen. The participant was asked to respond to the direction of the centre target arrow by pressing one button if it pointed left and another if it pointed right. The flanking arrows could point in the same direction as the centre arrow (congruent condition, e.g. >>>>>) or in the opposite direction from the centre arrow (incongruent condition, e.g. <<><<). There was an even distribution of congruent and incongruent trials within each 200-trial block. The modified Flanker Task was created and delivered using STIM2 software (Compumedics Neuroscan, El Paso, TX, USA).

The task was performed in a small room with a wall and a divider to prevent visual distraction. Participants were instructed to look at a small white fixation-cross in the middle of a black screen where the target stimuli appeared and to respond as quickly as possible to the stimulus when it appeared. The response pad was placed on a table centered allowing for an elbow’s length of reach. Participants responded with their left index for arrows point left and their right index for arrows pointing right. Participants were seated 120 cm away from a 24-inch computer monitor. The center of the stimulus was 27 cm from the surface of the table and had a height of approximately 10 cm. Each stimulus was displayed for 150ms with a 1000ms response window. There was a 1250 or 1750 (randomized) ms inter-trial duration. A minimum response time of 200ms was required for correct responses to eliminate anticipatory responses.

Accuracy and response time were collected by the Stim2 software. Response time was only considered for correct trials. Since accuracy varied by condition and time, modified Flanker task performance was quantified by an accuracy and variability adjusted response time score, the Linear Integrated Speed Accuracy Score (referred to as RTLISAS) [49,50]. We chose to report results using RTLISAS (as opposed to the more typical RT) to account for the observed speed-accuracy tradeoffs made by the participants. Though still a relatively new measure, RTLISAS has already been used in a number of studies of cognition over the past 3 years, see for example [51-54]. The RTLISAS score was calculated using the following equation:

RT and PE are the participants’ mean response time and proportion error (1-accuracy) for each block of flanker responses, separated by congruency. The terms sd RT and sd PE refer to the participants overall standard deviation in response time and proportion error for each block of flanker responses, again separated by congruency. If PE was zero (100% accuracy), the latter portion of the equation was set to zero. In cases of high accuracy, the changes observed in RTLISAS were more influenced by response time than by accuracy. In these cases, RTLISAS can be considered mostly as an accuracy adjusted response time measure.

The interpretation of RTLISAS is analogous to the interpretation of response time; namely, lower scores are better. This is because as PE decreases (i.e. accuracy increases), the later part of the equation shrinks towards 0. If PE is equal to zero (100% accuracy), then the measure is simply the response time.

2.5 Statistical analysis

All analyses were conducted in R v.3.6.0 [55]. Differences in RTLISAS were analyzed using linear mixed effects models built using the lme4 R package v1.1–21 [56], with significance tests provided by the lmerTest R package v3.1–0 [57]. Fixed effects of condition (exercise, exercise anticipation, rest), time (early, pre-intervention, post-intervention), congruency (congruent, incongruent), and session (first, second, third) were included. The rate of no response and gender were included as covariates. Session, condition, time, and congruency where fully interacted. Statistical significance of fixed effects was determined using Satterthwaite Type III ANOVA tests conducted on the linear mixed effects models using the ANOVA function in R. Significance for all analyses was defined as p<0.05. Data driven post-hoc contrasts were conducted using the emmeans function from the emmeans R package v1.4.1 [58]. All contrasts used the Tukey correction for multiple comparisons and the Satterthwaite method for calculating degrees of freedom. Pairwise effect sizes (as standardized mean differences, SMDs, also known as Cohen’s d) were obtained with the emmeans package. The SMDs were calculated using the pairwise differences of the model estimates, divided by the estimated population standard deviation as obtained from the residual standard deviation of the model. Effect sizes were classified according to general convention with .2 as small, .5 as medium. and .8 as large.

3 Results

3.1 Participant characteristics

Participants had an average age of 22.0 years (standard deviation, sd = 0.90, range: 20–25) and 55% [17] were female. Average activity level as measured by the IPAQ was 2797 MET-min/wk (sd = 1898, range: 360–6558). Flanker data from all 31 participants was included in the analyses though 7 participants (3 male, 4 female) were missing a proportion of their data (proportion missing: 11% to 67%).

3.2 Exercise characteristics

Characteristics of participants during exercise are displayed in . The percent of maximum heart rate was similar across all three sessions at early and pre-intervention times. However, the percent of maximum heart rate attained by participants during the intervention and post-intervention was significantly higher in the exercise session than in the resting and anticipation sessions (during intervention: 67% ± 10% for exercise versus 38% ± 5%. for rest and 39% ± 5% for deception; post-intervention: 46% ± 9% for exercise versus 38% ± 5% for rest and 39% ± 6% for deception). The average recorded RPE during exercise was 13 (sd = 1), which aligned with a moderate intensity as intended. The average mechanical power of participants during exercise was 66 watts (sd = 29).

3.3 Flanker results

There was a significant interaction effect of condition x time for RTLISAS (F(4,419.86) = 4.13, p = 0.003). The RTLISAS for the rest, deception, and exercise conditions was not significantly different at the early assessment (rest: M = 368, SEM = 8.57; deception: M = 367, SEM = 8.59; exercise: M = 367, SEM = 8.57; all pairwise p = 1.000; SMDs: 0.01–0.04) or the pre-intervention assessment (rest: M = 360, SEM = 8.58; deception: M = 355, SEM = 8.59; exercise: M = 357, SEM = 8.54; all pairwise p>0.95; SMDs: 0.08–0.21). However, in the post-assessment, the RTLISAS in the exercise condition (M = 336, SEM = 8.54) was significantly lower when compared to both the rest condition (M = 352, SEM = 8.57, t = 3.38, df = 422, p = 0.022, SMD: 0.66) and deception condition (M = 358, SEM = 8.62, t = 4.62, df = 421, p<0.001, SMD: 0.91). The rest and deception conditions were not significantly different at post-assessment (t = 1.28, df = 420, p = 0.937, SMD: 0.25). RTLISAS by condition and time is presented in Fig 2.

Fig 2

Condition by time plot.

Points represent the model estimated accuracy adjusted response time (RTLISAS) for each level of condition (rest, deception, exercise) and time (early, pre-Intervention, post-intervention).

There was also significant interaction of session x time for RTLISAS (F(4,419.89) = 3.12, p = 0.015). In the first session (regardless of session type), the RTLISAS decreased (marginally significantly) from early (M = 377, SEM = 8.54) to pre-intervention (M = 364, SEM = 8.49, t = 3.08, df = 420, p = 0.057, SMD: 0.57) and from pre-intervention to post-intervention (M = 349, SEM = 8.51, t = 3.46, df = 420, p = 0.017, SMD: 0.63). In the second session, RTLISAS decreased from early (M = 366, SEM = 8.56) to pre-intervention (M = 351, SEM = 8.58, t = 3.15, df = 420, p = 0.045, SMD: 0.90) but not from pre-intervention to post-intervention (M = 344, SEM = 8.56, t = 1.55, df = 420, p = 0.831, SMD: 0.30). In the third session, RTLISAS showed no significant differences between early (M = 360, SEM = 8.66) and pre-intervention (M = 356, SEM = 8.66, t = 0.80, df = 420, p = 0.997, SMD: 0.16) or pre-intervention and post-intervention (M = 354, SEM = 8.66, t = 0.51, df = 420, p = 1.000, SMD: 0.10). RTLISAS by session and time is presented in Fig 3.

Fig 3

Session by time plot.

Points represent the model estimated accuracy adjusted response time for each level of session (first, second, third) and time (early, pre-intervention, post-intervention).

There were also main effects of condition (F(2,423.88) = 3.78, p = 0.024), time (F(2,419.91) = 25.05, p<0.001), session (F(2,425.03) = 5.47, p = 0.005), and congruency (F(1,419.82 = 226.14, p<0.001). RTLISAS was significantly higher in the rest condition (M = 360, SEM = 8.16) compared to the exercise condition (M = 353, SEM = 8.14, t = 2.36, df = 426, p = 0.050, SMD: 0.28) and the deception condition (M = 360, SEM = 8.16) compared to the exercise condition (t = 2.42, df = 424, p = 0.042, SMD: 0.28). The difference between the rest and deception conditions was not significant (t = 0.04, df = 422, p = 1.000, SMD: 0.10). RTLISAS was also higher at time 1 (M = 368, SEM = 8.14) than time 2 (M = 357, SEM = 8.14, t = 3.98, df = 420, p<0.001, SMD: 0.44) and RTLISAS at time 2 was higher than time 3 (M = 349, SEM = 8.14, t = 3.10, df = 420, p = 0.006, SMD: 0.34). RTLISAS was also higher in the first session (M = 363, SEM = 8.13) compared to the second (M = 354, SEM = 8.16, t = 3.26, df = 428, p = 0.003, SMD: 0.39) but not the third session (M = 357, SEM = 8.20, t = 2.22, df = 427, p = 0.068, SMD: 0.27). The difference between the second and third sessions was not statistically significant (t = 1.00, df = 421, p = 0.578, SMD: 0.12). Finally, incongruent trials had higher RTLISAS (M = 374, SEM = 8.07) compared to congruent trials (M = 342, SEM = 8.07, t = 15.04, df = 420, p<0.001, SMD: 1.36).

4 Discussion

This study evaluated the impact of moderate aerobic exercise, and anticipation of exercise, on cognitive control. Our results confirmed that moderate aerobic exercise improves cognitive control, and that this benefit occurs independent of anticipation (and possible arousal) that occurs prior to exercise. In addition, secondary analyses indicated that there are significant learning effects with a modified Flanker task, both within and across sessions. Together, these results confirm aerobic exercise as part of a strategy to augment cognitive function and suggest that research using a Flanker task should carefully consider learning effects in the study design, target sample size, and analysis.

4.1 Effect of exercise and anticipation of exercise

Our study provides support for the acute cognitive benefits of exercise, by demonstrating positive changes in cognitive control with exercise (in line with our hypothesis) while addressing previous methodological shortcomings. Most prior studies examining exercise and cognitive control only included post-exercise assessments [29,40-45]. Even when pre and post-assessment measures were used, many existing studies failed to blind participants to the order of the sessions [29,40-45]. As a result, pre-intervention arousal might be different before the exercise intervention than before control sessions, thereby confounding results. As a result, in prior research, it was unclear whether cognitive control improved in anticipation of exercise or as a result of exercise. Both were reasonable given prior observations of sympathetic nervous system activation pre-exercise [36-38], and the known effects of sympathetic nervous system activation on cognitive control [28,39]. Here, we included two pre-exercise cognitive assessments and a deception condition to separate the anticipatory and physical contributions of exercise. In this study, Flanker task performance was better after exercise condition than after rest, regardless of whether participants had been anticipating exercise.

Our results counter our hypothesis that arousal associated with anticipating exercise would improve cognitive control [28,39]. In this study, Flanker task performance was similar across sessions at the early (pre-reveal) and pre-intervention times. Furthermore, Flanker task performance did not differ in the deception condition between when participants were anticipating exercise (early assessment) and when they knew they were resting (pre-intervention). There are two explanations for the lack of observed differences due to the anticipation of exercise: (1) the anticipation of exercise did not change arousal levels or (2) arousal caused by anticipating exercise did not alter cognitive control.

We hypothesized that cognitive control would improve with the anticipation of exercise since prior research indicated that several markers of arousal and sympathetic nervous system activation changed prior to exercise onset (e.g., ventilator rate, cortisol levels, blood pressure) [36-38]. However, another study noted that there was considerable individual variability in the pre-exercise stress response and arousal changes [37]. As a result, it is reasonable to think that our participants did not experience pre-exercise arousal changes, or that changes were too small to induce cognitive improvements. Although the percent of heart rate maximum was 2 percentage points higher in the deception and exercise conditions than in the rest condition in this study, differences were not statistically significant. Any alteration in arousal due to the anticipation of exercise may have been too small to be detected by heart rate, a very coarse measure of arousal and sympathetic nervous system activation. However, it is impossible to confidently determine whether or not participants experienced arousal changes due to the anticipation of exercise in this study.

The second possibility is that a change in arousal was elicited by anticipation of exercise but that it was insufficient to, or simply did not, change cognitive control as measured by the flanker. Arguably, this is less likely as there is good evidence that arousal levels impact cognitive control. For example, a meta-analysis concluded that acute increases in cortisol enhanced cognitive control in the short term [39]. In addition, catecholamine neurotransmitters are believed to be involved in the regulation of arousal and cognitive control [59-61], though we did not measure these in this study. It seems likely that if the anticipation of exercise caused a significant change in arousal, it would have had an effect on the flanker task. However, it should be noted that our Flanker task had an even number of congruent and incongruent trials as opposed to more congruent than incongruent trials which has been shown to increase the Flanker congruency effect and make the task more cognitively demanding [62]. It is possible that had we adopted this uneven distribution of congruent and incongruent trials, the effect of the anticipation of exercise may have been stronger due to the heightened difficulty level of the task.

The observation that anticipation of exercise did not result in cognitive improvement, but exercise itself did, may point to a unique or additional mechanism of action for exercise that is dissociable from arousal. For example, it is possible that the improvement in cognitive control resulted at least partially from acute changes in brain derived neurotrophic factor (BDNF) or insulin-like growth factor-1 (IGF-1) activity, as BDNF [63] and IGF-1 [64] have both been associated with improved cognitive control but are not typically associated with psychologically induced arousal. However, levels of these growth factors were not measured in this study and so that hypothesis is merely speculative.

4.2 Learning effect

An incidental finding of this study was detection of a significant learning effect for the flanker task within and across sessions. In our perusal of the literature this appears to be poorly characterized. We found one study of the psychometric properties of a novel choice reaction time task that demonstrated learning effects across repeated testing [65]; one study that found a learning effect of a choice reaction time task from the first to the second day [66], though not across time points on the first day; and one study that reported flanker task improvement from an initial to a final session [67], though it was not certain whether this was due to a test-retest learning effect or a result of the interceding tasks that the participants performed. Our results support these limited findings and further characterize the learning effect of the flanker task. We found that performance improves both across sessions as well as across time points during the same session, in particular on the first day.

The methodological implications of this substantial learning effect are broad. Very often when using psychometric tests, particularly choice reaction time tests, papers (including ours) state that training was given to participants before administering the test in order to eliminate learning effects from the data [66]. However, it is not often demonstrated that this training actually leads to a plateau in performance; in our case, it was demonstrable that our training block of 50 Flanker trials performed at the start of each session absolutely did not do this. In our case, this was not a huge issue as we had counterbalanced our conditions meaning that our results were not biased by the learning effect. However, there may well be situations in which researchers judge it difficult to use strategies such as counterbalancing or control groups that would address this issue and may subsequently justify not doing those things by leaning on the assumption that the task training given to participants rendered the learning effect negligible. Our results show that for the Flanker task this is likely to be an untenable assumption which is something that must be considered when designing studies using this task.

4.3 Limitations

The most significant limitation in this study is that physiological arousal was evaluated using a very coarse measure and therefore we cannot confidently know whether or not a stress response occurred with the anticipation of exercise. As a result, we cannot conclusively determine whether or not arousal occurred due to the anticipation of exercise. Other limitations include a relatively small sample size drawn from a homogenous group and the use of just one measure of cognitive function. Our sample was quite small at only 31 participants. This may have limited our ability to detect a small but true effect of exercise anticipation, However, our results indicate a very small anticipatory effect (0.03 between rest and deception and 0.04 between rest and exercise at time 1, necessitating a sample size over 3800 to detect). Our participants were all young healthy adults, comprised entirely of undergraduate and graduate university students, therefore limiting generalizability. It is not prudent to assume that the benefits seen in this study necessarily generalize to other age groups or clinical populations. Additionally, our only measure of cognitive function was a modified Flanker task. This task only probes attentional control and cognitive inhibition meaning that the improvements seen following exercise in this study may only apply to those aspects of cognitive control and not others, such as memory or cognitive flexibility.

4.4 Future directions

This study provides evidence for the benefits of exercise to cognitive function. Future studies should continue to probe these benefits are, who they occur to, and how they arise. We also provide evidence that cognitive function does not benefit from the anticipation of exercise, at least for measures of attentional control and cognitive inhibition. However, open questions still include whether anticipation generates a measurable stress response, for whom anticipation generates a stress response, what activities generate an anticipatory stress response, and whether the stress response benefits cognitive function in those in whom it occurs. Finally, our incidental finding of a learning effect across times and sessions for the modified Flanker task carries important methodological implications. Given that a learning effect is present, it is necessary to take this into account when designing future experiments, for example by counterbalancing conditions or including control groups. There appears to be a serious gap in the literature when it comes to characterizing the learning effect of the flanker, and so this too is an area that needs to be further explored.

Summary results of all statistical analyses.

(PDF)

Click here for additional data file.

19 Aug 2020

PONE-D-20-21829

Moderate Aerobic Exercise, but not Anticipation of Exercise, Improves Cognitive Control

PLOS ONE

Dear Dr. Maximilian Bergelt,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Two experts in the field evaluated the manuscript. Although they feel that the manuscript has merit, both of them have identified methodological issues. In particular, the second reviewer is concerned with the even distribution of congruent and incongruent trials in the modified Flanker task used. The academic editor agrees that this is a crucial issue that should be clarified by the authors. Moreover, please carefully address all comments made by each reviewer.

Please submit your revised manuscript by November 20th, 2020. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Samuel Penna Wanner, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments:

In addition to the comments made by the two reviewers, please consider:

1. To run experiments to address if an even distribution of congruent and incongruent trials affects the modified Flanker task outcomes.

2. To present the heart rate and rating of perceiving exertion data (both the means and SD) without decimal places. Also, please describe how these two intensity parameters were measured (e.g., equipment and scale used).

3. To present the mechanical power output during the exercise trial.

4. To provide more details in the legend of figures, particularly of figures 2 and 3. Please remember that a figure must stand alone.

5. To replace “didn’t” with “did not” in lines 272 and 294.

6. To briefly indicate, in the Results section, whether the score used (i.e., RTLISAS) was more influenced by the response time or accuracy. In general, the literature suggests that an acute bout of physical exercise improves response time but does not change accuracy. Please also explain why a lower score means better performance (lower response time and/or lower accuracy?).

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please provide further details regarding how participants were recruited, including the participant recruitment date.

3. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The concept of the study is good and interesting. Methodology could be more detailed. However, the results can be further explored as well as their relationship with the existing literature.

.Abstract needs some review. The experimental situations are not included as well the deception concept.

.Introduction is clear and concise.

Line #46 The authors needs to clarify if "brain changes" are better explored, would they be positive or negative?

.Methods

The cardiorrespiratory fitness or mechanical power of participants could be included to observe the fitness level of participantes.This is important considering that different effects of acute aerobic exercise on cognition are observed both at different intensities of exercise (as mentioned by the author in the introduction) as well as at different levels of physical fitness.

The sample size is not adequate. A posteriori sample size calculation is stimulated.

Effect size calculation is also stimulated to provide the magnitude of the main effects observed in the results.

Line #49: Considering that learning effect is an important topic of this study, the authors should explain or refer to a protocol already published in literature that tested if learning effect is minmized after only one week.

Line #153: It should be informed whether the use of several criteria to analyze the parameters of cognitive task is arbitrarily proposed by authors or is already in the literature. If so, please cite this original work.

.Results

The results are too much descriptive and data are repeated at some points data.

Line #184: Regarding missing data of some participants concerning cognitive task parameters, were the remmaining data considered for analysis or were the participants excluded? Was there any criteria for this decision?

Line #189: The data could be reported according to relative values (i.e. %HRmáx) to increase the individuality of measure. The proximity of anaerobic thresold influences their cognitive control response, for example.

.Discussion

The discussion is superficial. A very simple relationship is made with the findings of the literature.

An important finding, the "learning effect" is poorly explorated. The authors can explore and relate these data with literature in conclusion, not just in the future directions.

Line #291: Here the authors need to reinforce that catecholamines and cortisol were not assessed in the present study.

Lines #291 and 298: unpadronized citation. These citations do not follow the journal's instructions

Line #316: the term 'strong' appears to be inappropriately used. The authors reinforce multiple limitations of the present study. Also, they did not measure the magnitude of statistical difference (i.e. effect size).

Reviewer #2: The authors presented a well-written manuscript on the effects of moderate exercise, and its anticipation effect on cognitive control. The authors also used a relatively new measure of the Flanker test.

The results are interesting and bring novelty to the literature. However, some methodological problems were found.

Major comments:

The authors adopted an even distribution of congruent and incongruent trials. However, Lehle and Hubner [1] pointed out the need to adopt a high frequency of congruent stimulus compared with incongruent, once with an even distribution, the participants can adapt to them. In that way, we cannot know if anticipation exercise did not affect cognition due to the low difficulty level of the task.

The sample was composed of 55% of females. The literature indicates that cognitive control is modulated by estrogen levels [2-4]. However, the authors did not control participants’ menstrual cycle, and we did not know at which extend this can be influencing the results.

Minor comments:

Line 40: The authors introduce the paragraph talking about exercise benefits to cognition. Then, they talk about the physiological and metabolic benefits of exercise. I believe this second sentence is not necessary.

Line 110: It seems that the sentence has a typing error.

Line 130: The author affirmed that they recorded the resistance during exercise condition, but they did not report these data in the results section.

Table 1. It not common to use this grid format in tables. I believe, for future submissions, the authors should fix it.

1. Lehle C, Hubner R. On-the-fly adaptation of selectivity in the flanker task. Psychonomic Bulletin & Review 2008;15(4):814-8

2. Colzato LS, Hertsig G, van den Wildenberg WP, Hommel B. Estrogen modulates inhibitory control in healthy human females: evidence from the stop-signal paradigm. Neuroscience 2010;167(3):709-15 doi: 10.1016/j.neuroscience.2010.02.029[published Online First: Epub Date]|.

3. Hodgetts S, Weis S, Hausmann M. Estradiol-related variations in top-down and bottom-up processes of cerebral lateralization. Neuropsychology 2017;31(3):319-27 doi: 10.1037/neu0000338[published Online First: Epub Date]|.

4. Jacobs E, D'Esposito M. Estrogen shapes dopamine-dependent cognitive processes: implications for women's health. J Neurosci 2011;31(14):5286-93 doi: 10.1523/jneurosci.6394-10.2011[published Online First: Epub Date]|.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: João Gabriel Silveira-Rodrigues

Reviewer #2: Yes: Larissa Oliveira Faria

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

15 Oct 2020

Dear Dr. Samuel Penna Wanner,

Thank you for the opportunity to revise our manuscript as well as the thoughtful comments and suggestions. In our revised manuscript, we have carefully considered your and the reviewers’ suggestions and have updated the manuscript accordingly. In the rest of the letter we respond to each of the comments raised by you and the reviewers. Reviewer comments are above, while our responses are below, preceded by '>'. Within the manuscript new additions or changes that were made to satisfy reviewer concerns are highlighted in yellow and tracked changes is turned on. In addition to changes to address reviewer comments, we have also made changes to address some grammatical, spelling, and stylistic errors that we found in our own review of the manuscript. Overall, we feel that the quality of our paper is substantially improved thanks to the changes we have made to address the issues raised in this review.

Cheers,

Max Bergelt

Additional Editor Comments:

In addition to the comments made by the two reviewers, please consider:

1. To run experiments to address if an even distribution of congruent and incongruent trials affects the modified Flanker task outcomes.

> Most prior studies of exercise in relation to the Flanker task have used an even distribution of congruent and incongruent trials (recent examples: (1–5)), in line with our design. Unfortunately, in-person research is not possible at our University at this time so it is impossible to systematically examine whether a harder Flanker congruency distribution would have been more sensitive to anticipation of exercise. However, we had discussion regarding this point, and it will certainly be something we look into for any future studies we do in this area.

2. To present the heart rate and rating of perceiving exertion data (both the means and SD) without decimal places. Also, please describe how these two intensity parameters were measured (e.g., equipment and scale used).

>Done

3. To present the mechanical power output during the exercise trial.

>Done

4. To provide more details in the legend of figures, particularly of figures 2 and 3. Please remember that a figure must stand alone.

>Done

5. To replace “didn’t” with “did not” in lines 272 and 294.

>Done

6. To briefly indicate, in the Results section, whether the score used (i.e., RTLISAS) was more influenced by the response time or accuracy. In general, the literature suggests that an acute bout of physical exercise improves response time but does not change accuracy. Please also explain why a lower score means better performance (lower response time and/or lower accuracy?).

>We have added in an explanation for why a lower score means improved performance to the Methods section of our paper. In cases of high accuracy (as in this study), RTLISAS can be considered an accuracy-adjusted response time, so the primary driver of changes in RTLISAS is changes in reaction time. A study comparing methods to combine accuracy and RT considered RTLISAS as the method adding the most value (6).

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

>We have updated the manuscript and our files to ensure that titles, references, figures, and supplementary files are formatted and named correctly.

2. Please provide further details regarding how participants were recruited, including the participant recruitment date.

>We have added more detail as to how participants were recruited (by email and posters) and the dates recruitment was active (July 2018 to April 2019).

3. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

>Done.

Response to reviewers

Reviewer #1: The concept of the study is good and interesting. Methodology could be more detailed. However, the results can be further explored as well as their relationship with the existing literature.

.Abstract needs some review. The experimental situations are not included as well the deception concept.

>We have clarified the experimental design in the abstract, within the limitations of the word limit.

.Introduction is clear and concise.

>Thank you.

Line #46 The authors needs to clarify if "brain changes" are better explored, would they be positive or negative?

>This is a good point, ‘brain changes’ is vague. We have changed the wording specify increased volume of areas of the brain (e.g. hippocampus) and connectivity in some brain networks.

.Methods

The cardiorrespiratory fitness or mechanical power of participants could be included to observe the fitness level of participants. This is important considering that different effects of acute aerobic exercise on cognition are observed both at different intensities of exercise (as mentioned by the author in the introduction) as well as at different levels of physical fitness.

>Thank you for catching this. We did calculate the mechanical power of participants during the exercise session, though we overlooked including it in the manuscript. We have now added it to the section 3.2 Exercise Characteristics.

The sample size is not adequate. A posteriori sample size calculation is stimulated.

>We agree that the sample size is not adequate to confidently assert that exercise anticipation does not have an effect, and we have added a few sentences to the limitations section of our paper to underscore this for readers. Based on our effect sizes (acknowledging they may be misleading with our small sample), an unreasonable >3800 people would need to be recruited to detect anticipatory effects.

Effect size calculation is also stimulated to provide the magnitude of the main effects observed in the results.

> We have now added standardized effect sizes (standardized mean differences) for all of our pairwise comparisons.

Line #49: Considering that learning effect is an important topic of this study, the authors should explain or refer to a protocol already published in literature that tested if learning effect is minmized after only one week.

> We now made reference to what we could find in the existing literature on learning effects in our discussion. However, there is relatively little research investigating learning effects with the Flanker. By randomizing the order of sessions and controlling for session number as well as session type, however, we adjust for the impact of learning effects across sessions.

Line #153: It should be informed whether the use of several criteria to analyze the parameters of cognitive task is arbitrarily proposed by authors or is already in the literature. If so, please cite this original work.

>Thank you for this suggestion, we have added an explanation for why chose to use an integrated speed-accuracy score vs the more typical speed score. Though the measure we used is relatively new, it has already been used a fair amount within the literature. We have added a few example references that use it in the domain of cognition.

Results

The results are too much descriptive and data are repeated at some points data.

>We have reworked our results section and removed the repeated data.

Line #184: Regarding missing data of some participants concerning cognitive task parameters, were the remaining data considered for analysis or were the participants excluded? Was there any criteria for this decision?

>All remaining data was considered for analysis, that is, no data that we possessed was excluded. This is because things like pairwise deletion will often result in biased estimates (as the mechanism of missing data is far more likely to be Missing at Random rather than Missing Completely at Random) and was the main reason for our use of linear mixed effects models as opposed to a standard repeated measures ANOVA. We have reworded the sentence to make it more clear that all data was used in the analysis.

Line #189: The data could be reported according to relative values (i.e. %HRmáx) to increase the individuality of measure. The proximity of anaerobic thresold influences their cognitive control response, for example.

>This is a good point and so we’ve now reported all heart rate data as %HR max.

.Discussion

The discussion is superficial. A very simple relationship is made with the findings of the literature.

An important finding, the "learning effect" is poorly explorated. The authors can explore and relate these data with literature in conclusion, not just in the future directions.

> The learning effects were certainly of interest to us. It surprised us that there was almost no mention of learning effects with reference to the Flanker task in the literature, despite an obvious learning effect across sessions. This has import implications for future studies, where learning effects should either be minimized (difficult in a restricted time frame) or adjusted for within analyses. We have added a more thorough discussion.

Line #291: Here the authors need to reinforce that catecholamines and cortisol were not assessed in the present study.

>We have added to the sentence to reinforce this.

Lines #291 and 298: unpadronized citation. These citations do not follow the journal's instructions

>Thanks for catching this, we have now fixed them.

Line #316: the term 'strong' appears to be inappropriately used. The authors reinforce multiple limitations of the present study. Also, they did not measure the magnitude of statistical difference (i.e. effect size).

>This is a great point, ‘strong’ is often used in relation to magnitude of effects so we have removed the word ‘strong’. We have also added effect sizes to our results.

Reviewer #2: The authors presented a well-written manuscript on the effects of moderate exercise, and its anticipation effect on cognitive control. The authors also used a relatively new measure of the Flanker test.

The results are interesting and bring novelty to the literature. However, some methodological problems were found.

Major comments:

The authors adopted an even distribution of congruent and incongruent trials. However, Lehle and Hubner [1] pointed out the need to adopt a high frequency of congruent stimulus compared with incongruent, once with an even distribution, the participants can adapt to them. In that way, we cannot know if anticipation exercise did not affect cognition due to the low difficulty level of the task.

>Indeed, a higher frequency of congruent trials will increase the Flanker Congruency Effect. It is possible that Flanker parameters that elicit an increased Flanker Congruency Effect may be more sensitive to anticipatory effects, though that is not certain. In the case of exercise-related literature, use of an even distribution of congruent and incongruent trials has been common (recent examples: (1–5)), as we did here. Also, positive effects of exercise have also been observed in choice response time tasks. However, it remains possible that anticipation of exercise may be more strongly observed in very difficult tasks (which are more likely to suffer with stress) and we have added this point to our discussion.

The sample was composed of 55% of females. The literature indicates that cognitive control is modulated by estrogen levels [2-4]. However, the authors did not control participants’ menstrual cycle, and we did not know at which extend this can be influencing the results.

> We acknowledge the reviewers point that the female participants’ Flanker performance may have been influenced by their menstrual cycle. However, given the randomization across conditions, it seems unlikely that this would result in a systematic bias across conditions, though it likely adds noise.

Minor comments:

Line 40: The authors introduce the paragraph talking about exercise benefits to cognition. Then, they talk about the physiological and metabolic benefits of exercise. I believe this second sentence is not necessary.

>We agree and have deleted this sentence and rewrote the surrounding sentences.

Line 110: It seems that the sentence has a typing error.

>Thanks for catching this. Upon revisiting, the sentence did seem slightly confusing, so we have changed it.

Line 130: The author affirmed that they recorded the resistance during exercise condition, but they did not report these data in the results section.

>Thank you for catching this, Reviewer 1 mentioned this as well. We have now added it to the section 3.2 Exercise Characteristics.

Table 1. It not common to use this grid format in tables. I believe, for future submissions, the authors should fix it.

>We have changed the styling.

1. Lehle C, Hubner R. On-the-fly adaptation of selectivity in the flanker task. Psychonomic Bulletin & Review 2008;15(4):814-8

2. Colzato LS, Hertsig G, van den Wildenberg WP, Hommel B. Estrogen modulates inhibitory control in healthy human females: evidence from the stop-signal paradigm. Neuroscience 2010;167(3):709-15 doi: 10.1016/j.neuroscience.2010.02.029[published Online First: Epub Date]|.

3. Hodgetts S, Weis S, Hausmann M. Estradiol-related variations in top-down and bottom-up processes of cerebral lateralization. Neuropsychology 2017;31(3):319-27 doi: 10.1037/neu0000338[published Online First: Epub Date]|.

4. Jacobs E, D'Esposito M. Estrogen shapes dopamine-dependent cognitive processes: implications for women's health. J Neurosci 2011;31(14):5286-93 doi: 10.1523/jneurosci.6394-10.2011[published Online First: Epub Date]|.

________________________________________

References

1. Lefferts WK, DeBlois JP, White CN, Heffernan KS. Effects of Acute Aerobic Exercise on Cognition and Constructs of Decision-Making in Adults With and Without Hypertension. Front Aging Neurosci. 2019;

2. Won J, Alfini AJ, Weiss LR, Callow DD, Smith JC. Brain activation during executive control after acute exercise in older adults. Int J Psychophysiol. 2019;

3. Du Rietz E, Barker AR, Michelini G, Rommel AS, Vainieri I, Asherson P, et al. Beneficial effects of acute high-intensity exercise on electrophysiological indices of attention processes in young adult men. Behav Brain Res. 2019;

4. Tsai CL, Ukropec J, Ukropcová B, Pai MC. An acute bout of aerobic or strength exercise specifically modifies circulating exerkine levels and neurocognitive functions in elderly individuals with mild cognitive impairment. NeuroImage Clin. 2018;

5. Beyer KB, Sage MD, Staines WR, Middleton LE, McIlroy WE. A single aerobic exercise session accelerates movement execution but not central processing. Neuroscience. 2017;

6. Vandierendonck A. Further Tests of the Utility of Integrated Speed-Accuracy Measures in Task Switching. J Cogn. 2018;

Submitted filename: PLOSONE_Rebuttal_Letter.docx

Click here for additional data file.

26 Oct 2020

PONE-D-20-21829R1

Moderate aerobic exercise, but not anticipation of exercise, improves cognitive control

PLOS ONE

Dear Dr. Maximilian Bergelt,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The academic editor and the two reviewers believe that the revised manuscript has merit and that it has been dramatically improved compared to the previous version submitted to PLOS One. Indeed, the authors were highly responsive to all comments. Congratulations! Although both reviewers judged that the manuscript is acceptable for publication in the present form, I still think that the authors should explain how they have calculated the standardized mean differences (as required by the first reviewer) in the material and methods section. If the authors adequately address the issue mentioned above, this academic editor will handle the manuscript by himself and not send it again for external reviewer evaluation.

Please submit your revised manuscript by November 25th, 2020. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Samuel Penna Wanner, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

Please consider:

1) To define the meaning of the abbreviation SMD and explain how the standardized mean differences were calculated. Please also indicate the threshold values used to classify the effects sizes based on the SMD calculation.

2) To present the heart rate data as absolute values and indicate the percentage of maximum heart rate for specific time points, particularly after the exercise.

The editor is looking forward to receiving a revised and improved version of the manuscript.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: João Gabriel Silveira-Rodrigues

Reviewer #2: Yes: Larissa Oliveira Faria

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

27 Oct 2020

Rebuttal Letter

PONE-D-20-21829

Moderate aerobic exercise, but not anticipation of exercise, improves cognitive control

PLOS ONE

Dear Dr. Wanner,

Thank you for the opportunity to revise our manuscript as well as the thoughtful comments and suggestions. In our revised manuscript, we have addressed your additional comments with our responses detailed below. Alterations to the manuscript addressing your comments are highlighted in yellow.

Thank you again for your suggestions, we believe that they significantly improve the clarity of this manuscript.

Best regards,

Max Bergelt

Additional Editor Comments:

1) To define the meaning of the abbreviation SMD and explain how the standardized mean differences were calculated. Please also indicate the threshold values used to classify the effects sizes based on the SMD calculation.

>We have added the meaning of SMD and explained how it was calculated in section ‘2.5 Statistical analysis’. We have also added thresholds for small, medium, and large effects in order to clarify interpretation.

2) To present the heart rate data as absolute values and indicate the percentage of maximum heart rate for specific time points, particularly after the exercise.

>We now present both the absolute heart rates at each time point, as well as percentage of maximum heart rate. The absolute heart rate is presented in Table 1. The %HR max during and after the intervention are presented in the text of ‘3.2 Exercise characteristics’.

Submitted filename: PLOSONE_Rebuttal_Letter_Rev_2.docx

Click here for additional data file.

30 Oct 2020

Moderate aerobic exercise, but not anticipation of exercise, improves cognitive control

PONE-D-20-21829R2

Dear Dr. Maximilian Bergelt,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Samuel Penna Wanner, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

The authors have adequately addressed my comments. Thank you.

Reviewers' comments:

5 Nov 2020

PONE-D-20-21829R2

Moderate aerobic exercise, but not anticipation of exercise, improves cognitive control

Dear Dr. Bergelt:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Samuel Penna Wanner

Academic Editor

PLOS ONE

Table 1

Exercise characteristics by condition and time.

Time	Rest	Deception	Exercise	P-Value
Early HR	78 (11)	81 (11)	81 (12)	0.287
Pre-Intervention HR	78 (10)	79 (11)	78 (11)	0.784
During Intervention HR*	75 (9)	76 (9)	132 (20)	< .001
Post-Intervention HR	76 (9)	77 (12)	91 (17)	< .001
During Intervention RPE†	-	-	13 ± 1	-
During Intervention Watts†	-	-	66 ± 29	-

All values are represented by mean ± SD unless otherwise specified.

* Mean of the median values of the ten time points collected two minutes apart during the intervention

† Queried only during the exercise condition