Examining the Effects of Mixed-Models and Self-Observation on Motor Skill Acquisition Within a Gymnastics Environment

Abstract

Watching oneself on video (self-observation) compared to self-observation coupled with a skilled model video (mixed-models) was examined in a gymnastics environment to determine whether combining two model types would be better than just one. Twenty-one gymnasts learned one gymnastics skill with mixed-models and a second skill with self-observation across pre-test, three learning sessions, and post-test. Physical performance, scored by two evaluators, revealed a significant condition by session interaction (F(3,51) = 3.329, p = .027). At session 3 and post-test, scores obtained with mixed-models were significantly higher than those with self-observation. Cognitive representation of the skills was measured at pre-test and post-test via error detection and recognition tests, analyzed using signal detection. Participants had significantly higher response sensitivity scores with mixed-models (F(1,14) = 10.810, p = .005) compared to self-observation. The conclusion drawn is that it is better to incorporate self and skilled models in a gymnastics setting than self-observation alone.