Bull's eye-hand coordination: visual and motor contributions to observational learning

Watching other people move affords observers many benefits such as presenting opportunities for social interactions, deciphering other’s intentions and emotional states, and learning new motor skills. Observational motor learning is the process of learning to perform a novel motor skill by watching others execute that skill. Common coding theories of the visual and motor representations of actions allow for a relationship between action observation and action execution that makes observational motor learning possible. More specific theories related to observational learning conflict in terms of whether observed actions are first represented at the kinematic level or at the action goal level. The overarching goal of this series of experiments was to increase our understanding of how the visual system and the motor system work together to enhance action learning. Experiments 1 and 2 examined the influence of model expertise on observational learning of dart throwing. Dart throwing was selected as the motor task of interest because it has been used previously in perception-action coupling research and because it represents an ecologically valid complex motor task. This differs from past research on action learning which has tended to rely on simple, contrived motor skills. In Experiments 1 and 2, participants threw darts before and after they watched an expert or a novice dart throwing model. To rule out the possibility that the observation of any dart throwing actions might improve an observer’s ability to throw darts, Experiment 2 included an additional control condition involving a model playing basketball. No significant differences were found in participants’ dart throwing abilities after the observation of either dart thrower or the basketball player. Instead, physical practice effects were sufficient to account for all improvements in dart throwing performance. Experiments 3 and 4 used measures of visual sensitivity, rather than motor performance, to assess the motor system’s contributions to visual learning of other people’s actions. Action observation is thought to involve an action simulation process that impacts an observer’s ability to predict the outcomes of other people’s actions. Thus, if an observer’s motor system is otherwise engaged, that observer should be compromised in his or her ability to simulate another person’s actions and as a result, should demonstrate deficits in predicting action outcomes. In Experiment 3, participants completed a dart throwing prediction task before and after the observation of an expert dart thrower. The outcomes of this dart thrower’s actions (i.e., where darts landed) were only visible in the observation phase. Importantly, during this action observation phase, participants’ motor systems were engaged to allow for the determination of core characteristics of the action simulation process. The results suggest that some types of motor system engagement reduce action prediction capabilities. Interestingly, significant inverse correlations were found between physical effort during motor engagement and action prediction accuracy. Experiment 4 investigated the theoretical common coding between the visual and motor systems by assessing the relative impacts of visual and nonvisual motor training on action outcome prediction. Participants completed the action prediction task from Experiment 3 before and after performing visual or nonvisual motor training. In the nonvisual training condition, participants physically performed dart throwing while their vision of their throwing arm was occluded. In the visual training condition, participants physically performed dart throwing with full vision of their throwing arm. Lastly, in a control condition, participants played basketball. Participants in the nonvisual motor training condition demonstrated the largest gain in visual sensitivity in the action prediction task. In conclusion, while the results of these experiments lend partial support for the common coding theory in general, they do not differentiate between specific perception-action coupling theories. Nonetheless, the current results do raise important questions about the generalizability of simple motor action studies to more complex, real world actions. Additionally, exciting future directions are revealed by the results of these experiments.Ph. D.Includes bibliographical referencesIncludes vitaby Ashley Blanchar