Effects of adaptive, random, and self-controlled practice schedules on acquisition, retention, and transfer in highly skilled table tennis players: an experimental study

Evidence on the comparative effectiveness of random, learner-adaptive, and self-controlled practice schedules among skilled athletes remains limited. Their effects on motor learning and performance in applied table tennis settings require further investigation. The present study examined the motor performance and learning effects of random, learner-adaptive, and self-controlled practice schedules on table tennis stroke accuracy in highly skilled players. Forty-eight highly skilled table tennis players were assigned to one of four groups: random practice, learner-adaptive practice, self-controlled practice, and a no-practice control group. The self-controlled group regulated task difficulty and serving parameters during practice, whereas these variables were externally controlled in the random and learner-adaptive practice groups. During the acquisition phase, participants completed a total of 630 practice trials distributed across seven practice sessions. Performance was assessed during acquisition, immediate retention, delayed retention, and transfer tests using linear mixed-effects models. The results demonstrated significant improvements in performance across acquisition sessions for all practice groups. During the final acquisition session, the self-controlled group demonstrated lower performance compared with the random and learner-adaptive groups. During immediate retention, all practice groups outperformed the control group; however, no statistically significant differences were observed between the three practice groups. Similarly, no differences were found between the practice groups during delayed retention. During the transfer test, the random practice group demonstrated significantly superior performance relative to the self-controlled group, whereas the learner-adaptive group outperformed the control group. No statistically significant differences were observed between the random and learner-adaptive practice groups during transfer performance. These findings suggest that random practice remains an effective practice schedule for promoting transfer performance in highly skilled table tennis players, whereas the specific rule-based Win-Shift/Lose-Stay learner-adaptive procedure used in this study did not provide additional learning benefits beyond random practice. Given the additional monitoring and adjustment required to implement this adaptive procedure, random practice may represent a simpler and equally effective option in practice. Moreover, the additional choices available in the self-controlled condition may have contributed to increased task difficulty, which could have limited transfer performance.