Motor competence is related to acquisition of error-based but not reinforcement learning in children ages 6 to 12

Abstract

Background: An essential component of childhood development is increasing motor competence. Poor motor learning is often thought to underlie impaired motor competence, but this link is unclear in previous studies. Aims: Our aim was to test the relationship between motor competence and motor learning in the acquisition phase. Both reinforcement learning (RL) and error-based learning (EBL) were tested. We hypothesized that slower RL and slower EBL acquisition rates would relate to lower motor competence. Methods and procedures: Eighty-six participants ages 6–12 performed a target throwing task under RL and EBL conditions. The Movement Assessment Battery for Children – 2nd edition (MABC-2) provided a measure of motor competence. We assessed EBL and RL acquisition rates, baseline variability, and baseline bias from the throwing task. Outcomes and results: In a multiple linear regression model, baseline variability (β = −0.49, p = <0.001) and the EBL acquisition rate (β = −0.24, p = 0.018) significantly explained the MABC-2 score. Participants with higher baseline variability and slower EBL acquisition had lower motor competence scores. The RL acquisition rate was independent of MABC-2 score suggesting that RL may be less of a contributor to poor motor competence. Conclusions and implications: Children with slower EBL acquisition had lower motor competence scores but RL acquisition was unrelated to the level of motor competence. Emphasizing the unrelated reinforcement mechanisms over error-based mechanisms during motor skill interventions may help children with poor motor competence better acquire new motor skills.

Keywords

• Error-based learning
• Reinforcement learning
• Motor coordination
• Motor competence
• MABC-2