Toward AI-Enabled Precision Exercise: Neurophysiological Effects of Acute Moderate-Intensity Resistance Exercise on Executive Function in College Students with Different Motor Skill Backgrounds

Abstract

With the rapid advancement of artificial intelligence (AI) and intelligent healthcare technologies, integrating neuroimaging with exercise interventions has emerged as a promising strategy for promoting cognitive health and precision medicine. This study investigated the immediate effects of acute moderate-intensity resistance exercise on executive function in college students with different motor skill backgrounds and explored the underlying neural mechanisms using functional near-infrared spectroscopy (fNIRS). Furthermore, the multimodal behavioral and neurophysiological data generated in this study provide a valuable foundation for future AI-assisted cognitive assessment and precision exercise prescription. A total of 36 healthy male collegiate athletes were recruited and assigned to either an open-skill group (n = 18) or a closed-skill group (n = 18). Participants completed a 30-minute acute moderate-intensity resistance exercise session at 70% of their 10-repetition maximum (10RM). Executive function was assessed before and after exercise using the Stroop task, 2-back task, and More-odd shifting task to evaluate inhibitory control, working memory, and cognitive flexibility, respectively. Simultaneously, changes in oxygenated hemoglobin (ΔHbO 2 ) within the prefrontal cortex were continuously recorded using fNIRS. Data were analyzed using paired-sample t-tests and two-way repeated-measures analysis of variance. The results demonstrated that acute moderate-intensity resistance exercise was associated with significant improvements in selected executive-function outcomes within both groups. In the open-skill group, reaction time decreased in the Stroop task, while accuracy increased and reaction time decreased in the 2-back task. In the closed-skill group, reaction time decreased in the 2-back task. Both groups showed improvements in selected outcomes of the More-odd shifting task. However, no significant Group × Time interactions were observed for any behavioral outcome, indicating that the magnitude of behavioral change did not differ significantly between groups. fNIRS findings revealed task- and channel-specific changes in prefrontal ΔHbO 2 following exercise. Moreover, significant Group × Time interactions in ΔHbO 2 were observed at selected prefrontal channels during the Stroop, 2-back, and More-odd shifting tasks, suggesting that motor skill background may be associated with different patterns of prefrontal hemodynamic response following acute exercise. In conclusion, acute moderate-intensity resistance exercise was followed by improvements in selected executive-function outcomes in young male collegiate athletes. The integration of behavioral assessments with fNIRS-derived neurophysiological data provides additional evidence regarding behavioral and prefrontal hemodynamic responses to acute resistance exercise. Although the behavioral changes did not differ significantly between groups, exploratory fNIRS findings indicated task- and channel-specific differences in ΔHbO 2 changes between the open-skill and closed-skill groups. Further studies incorporating a non-exercise control condition, larger and more diverse samples, and appropriate correction for multiple channel-level comparisons are needed to confirm these findings.