Weekday Exercise Breaks Linked to Improved Cognitive Performance in Student Trial

A twelve-week randomized controlled trial conducted at Tianjin University of Commerce demonstrated that incorporating three brief physical activity bouts into the daily academic routine of sedentary university students yielded measurable improvements across multiple domains of cognitive task performance, self-reported emotional affect, and perceived psychological stress. The parallel-group investigation evaluated one hundred and twenty undergraduate students aged eighteen to twenty-five years who were randomly assigned in equal numbers to either an active exercise-break protocol or a usual-routine control condition. Participants assigned to the intervention completed three five-minute movement sessions across each weekday, accumulating approximately fifteen minutes of daily structured activity during intervals between academic study periods. Repeated-measures intention-to-treat analyses encompassing the entire randomized cohort revealed statistically significant group-by-time interaction effects across all six primary outcomes examined over the twelve-week intervention period. Concurrently, a per-protocol sensitivity analysis of the one hundred and sixteen students who finished all study requirements confirmed directionally consistent performance patterns across all six evaluated outcomes.

Sedentary behavior is formally defined in public health literature as any waking behavior characterized by an energy expenditure of 1.5 metabolic equivalents or fewer while in a sitting, reclining, or lying posture. Because sedentary behavior does not simply equate to an absence of physical activity, university students often accumulate high volumes of prolonged sitting during mandatory classroom lectures, computer-based coursework, library research, and recreational screen time, even if they occasionally participate in sports. Public health authorities widely encourage replacing continuous seated periods with light or moderate physical movement, yet university campuses frequently face structural timetabling constraints that prevent the implementation of extended exercise sessions during the academic day. While prior feasibility trials have explored whether inserting brief movement pauses into higher education environments could alleviate physical fatigue, heighten alertness, and enhance concentration, earlier published findings remained non-uniform and equivocal. Furthermore, previous investigations frequently relied on single acute exercise sessions or short-duration crossover experiments, leaving longitudinal questions unanswered regarding whether repeating brief, low-volume exercise bouts across several consecutive academic months produces sustained cognitive and emotional enhancements.

Recruitment for the trial took place between September 2025 and January 2026 through the university health promotion program, screening a total of one hundred and fifty-six interested undergraduate volunteers. Following preliminary screening, thirty-six students were excluded for failing to satisfy the protocol eligibility criteria, which required participants to be full-time university attendees aged eighteen to twenty-five who reported sitting for at least six hours daily and engaged in structured physical exercise no more than twice weekly during the prior three months. Eligible candidates were also required to possess normal or corrected vision and have no history of psychiatric or neurological disorders, cardiovascular complications, or medications known to alter cognitive or affective functioning. The remaining one hundred and twenty participants were allocated by an independent researcher using a computer-generated randomization sequence into two balanced arms of sixty individuals each. At baseline, the exercise-break cohort exhibited a mean age of 20.84 years, an average body mass index of 21.86, and daily sedentary exposure averaging 8.21 hours, while the control cohort demonstrated comparable baseline characteristics with a mean age of 20.71 years, a body mass index of 21.73, and daily sitting time averaging 8.35 hours.

The experimental intervention was structured around three scheduled five-minute exercise bouts executed every weekday over twelve consecutive weeks, distributed systematically across morning study periods, afternoon academic sessions, and evening independent study intervals. The physical activity routine comprised a standardized circuit of body-weight squats, marching in place, dynamic stretching, upper-body mobility drills, and light aerobic movements designed to elicit a moderate perceived exertion rating between 11 and 13 on the Borg Rating of Perceived Exertion Scale. Intervention adherence was facilitated through a dedicated university monitoring application where participants recorded completed sessions, supplemented by weekly compliance checks from the research staff. In contrast, participants allocated to the control group maintained their customary daily routines without incorporating structured activity breaks, receiving minimal researcher interaction, fewer compliance reminders, and no exercise prompts. The investigators established an operational adherence threshold requiring participants to complete at least eighty percent of scheduled sessions to qualify for the per-protocol sensitivity cohort, providing a practical operational criterion to accommodate occasional missed sessions while maintaining a rigorous evaluation of the intervention.

Cognitive performance was evaluated at baseline, week six, and week twelve under the supervision of research personnel using three computerized and paper-based instruments targeting distinct aspects of executive function. Inhibitory control and selective visual attention were measured using the Stroop Color-Word Test, where participants identified font colors while suppressing semantic word interference, with reaction time recorded in milliseconds as the primary performance metric. Working memory capacity and executive updating were assessed using a computerized two-back continuous recognition paradigm, in which participants indicated whether the current stimulus matched the item presented two steps earlier, quantified by overall response accuracy percentages. Cognitive flexibility, visual searching speed, and mental task switching were appraised through the timed Trail Making Test Part B format, which required students to rapidly alternate between sequential numbers and letters on a paper form. These three complementary measures captured distinct cognitive dimensions, as task-switching agility does not directly correlate with simple working memory storage or inhibitory response latency.

Emotional wellbeing was simultaneously assessed at baseline, mid-intervention, and study conclusion using two standardized self-report psychometric questionnaires administered online. The Positive and Negative Affect Schedule assessed subjective emotional states over the preceding week using twenty distinct emotional descriptors divided equally between positive affect and negative affect subscales, each scored from ten to fifty points where higher values denote greater intensity of the corresponding affective dimension. Perceived psychological stress was measured using the ten-item Perceived Stress Scale, which assessed how much a person felt stressed over the preceding month, generating aggregate scores ranging from zero to forty points. Chinese-language versions of both psychometric instruments were administered online to the student cohort. The study design intentionally focused on self-reported psychological states rather than biological indices of emotional regulation, acknowledging that questionnaire assessments capture subjective perceived wellbeing rather than autonomic or neuroendocrine biomarkers.

The study statistical architecture adhered to an intention-to-treat framework incorporating all one hundred and twenty randomized participants into two-way repeated-measures analyses of variance examining group, time, and group-by-time interaction effects. During the twelve-week trial, four students discontinued active protocol participation due to personal scheduling conflicts—two within the exercise-break arm and two within the control arm—leaving one hundred and sixteen individuals who completed all assigned procedures and formed the per-protocol sensitivity cohort. Because the primary omnibus repeated-measures analyses yielded unadjusted interaction p-values reported below the 0.001 threshold for all six outcome metrics, Holm-adjusted familywise significance values were conservatively designated as p less than 0.006 to account for multiple simultaneous comparisons across the unified outcome family. Bonferroni corrections were applied when contrasting adjacent time points within measures, while unadjusted between-group mean differences and corresponding ninety-five percent confidence intervals at week twelve were estimated directly from completer data summaries to provide descriptive per-protocol contrasts rather than model-adjusted intention-to-treat estimates.

Cognitive testing demonstrated marked, progressive performance improvements within the exercise-break cohort compared to relatively stable trajectories in the control condition across the twelve-week experimental period. Mean Stroop response times among exercise-break participants decreased from 684.52 milliseconds at baseline to 642.18 milliseconds at week six and 598.63 milliseconds at week twelve, whereas control group latencies changed modestly from 679.84 milliseconds to 671.36 milliseconds and 663.42 milliseconds, yielding a statistically significant interaction, F(2, 236) = 32.84, p < 0.001, partial eta-squared = 0.22, and a week-twelve unadjusted mean difference of -64.79 milliseconds. On the two-back working memory task, exercise-break accuracy rose from 72.46 percent to 78.93 percent and reached 84.57 percent, while control accuracy shifted from 71.93 percent to 73.42 percent and 74.65 percent, demonstrating an interaction effect, F(2, 236) = 41.26, p < 0.001, partial eta-squared = 0.26, with an unadjusted completer difference of +9.92 percentage points. Similarly, Trail Making Test completion times in the exercise group dropped from 76.42 seconds to 68.25 seconds and 61.37 seconds, compared to minor reductions in controls from 75.81 seconds to 74.63 seconds and 72.86 seconds, confirming a significant interaction, F(2, 236) = 36.71, p < 0.001, partial eta-squared = 0.24, and a final between-group difference of -11.49 seconds.

Parallel improvements emerged across all affective and psychological stress dimensions over the course of the twelve-week study. Positive affect ratings among exercise-break participants advanced from a baseline mean of 27.84 to 31.26 at week six and 35.42 at week twelve, whereas control participants registered scores of 28.16, 28.74, and 29.31 across the same intervals, reflecting an interaction, F(2, 236) = 38.45, p < 0.001, partial eta-squared = 0.25, and a week-twelve difference of +6.11 points. Negative affect scores in the exercise group declined substantially from 21.37 at baseline to 18.42 at mid-intervention and 15.86 at study completion, compared to minimal changes in controls from 21.65 to 21.02 and 20.74, generating an interaction, F(2, 236) = 29.63, p < 0.001, partial eta-squared = 0.20, with a final difference of -4.88 points. Perceived stress scores mirrored this downward pattern, decreasing in the intervention arm from 21.26 to 18.36 and 15.42, while control scores moved from 21.54 to 21.01 and 20.65, demonstrating an interaction, F(2, 236) = 35.17, p < 0.001, partial eta-squared = 0.23, and a week-twelve contrast of -5.23 points. Exploratory pooled Pearson correlations across all one hundred and sixteen completers indicated that changes in Stroop latency correlated with changes in positive affect (r = -0.54) and perceived stress (r = 0.49), while two-back accuracy changes correlated inversely with negative affect changes (r = -0.46), though these unadjusted associations cannot establish direct mediation or causal relationships between cognitive gains and emotional relief.

Notwithstanding the consistent empirical differences documented across all measured outcomes, the study authors identified multiple design limitations that restrict broader generalizability and causal interpretation. Because participants were recruited from a single educational institution without post-intervention longitudinal follow-up, the findings may not generalize across diverse academic environments or demographic profiles. Furthermore, the trial did not record an a priori power calculation, allocation concealment procedures, outcome assessor blinding, or objective biometric verification of the exercise intensity achieved during unmonitored sessions. Crucially, the protocol did not repeatedly measure total daily sitting duration, sitting-bout patterns, or overall physical activity, precluding any empirical conclusion that the intervention successfully reduced total sedentary volume or that physical movement alone produced the observed benefits. Disparities in researcher contact and compliance tracking between study arms introduce potential behavioral expectancy effects, while candidate physiological mechanisms—including cerebral blood flow, neurotrophic factors, cortisol regulation, and autonomic tone—remained unmeasured hypotheses. Consequently, the investigators emphasized that while brief activity breaks represent a feasible campus health strategy, future research must incorporate objective physical activity tracking, standardized testing environments, prospective trial registration, and cost-effectiveness evaluations before institutional implementation.

Subject of Research: Effects of repeated short exercise breaks on cognitive function and emotional wellbeing in sedentary university students

Article Title: Effects of repeated short exercise breaks on cognitive task performance, affect, and perceived stress in sedentary university students: a 12-week randomized controlled trial

Article References: Sun, C. (2026). Effects of repeated short exercise breaks on cognitive task performance, affect, and perceived stress in sedentary university students: a 12-week randomized controlled trial. Frontiers in Public Health, 14, Article 1961715. https://doi.org/10.3389/fpubh.2026.1961715

Image Credits: AI Generated

DOI: 10.3389/fpubh.2026.1961715

Keywords: randomized controlled trial, sedentary behavior, university students, exercise breaks, executive function, Stroop Color-Word Test, 2-back task, Trail Making Test, Positive and Negative Affect Schedule, Perceived Stress Scale, working memory, cognitive performance, positive affect, negative affect, perceived stress, physical activity, Tianjin University of Commerce, intention-to-treat analysis