Scheduling observational and physical practice: Influence on the coding of simple motor sequences

Abstract

The main purpose of the present experiment was to determine the coordinate system used in the development of movement codes when observational and physical practice are scheduled across practice sessions. The task was to reproduce a 1,300-ms spatial–temporal pattern of elbow flexions and extensions. An intermanual transfer paradigm with a retention test and two effector (contralateral limb) transfer tests was used. The mirror effector transfer test required the same pattern of homologous muscle activation and sequence of limb joint angles as that performed or observed during practice, and the nonmirror effector transfer test required the same spatial pattern movements as that performed or observed. The test results following the first acquisition session replicated the findings of Gruetzmacher, Panzer, Blandin, and Shea (2011). The results following the second acquisition session indicated a strong advantage for participants who received physical practice in both practice sessions or received observational practice followed by physical practice. This advantage was found on both the retention and the mirror transfer tests compared to the nonmirror transfer test. These results demonstrate that codes based in motor coordinates can be developed relatively quickly and effectively for a simple spatial–temporal movement sequence when participants are provided with physical practice or observation followed by physical practice, but physical practice followed by observational practice or observational practice alone limits the development of codes based in motor coordinates.

Keywords:

• Observational learning
• Motor learning
• Effector transfer
• Movement representation

Acknowledgments

This work was supported by a grant from the German Research Foundation (PA 774/8-1) and the Agence Nationale de la Recherche (ANR-08-FASHS-14).

Notes

¹ Note that the movement goal was to produce the pattern of movement depicted in the display in 1,300 ms. Thus, in order to determine the discrepancy between the time series comprising the participant's waveform and the goal waveform, the differences between the goal waveform and the subject-produced waveform were compared at each of the 585 points (450-Hz sampling rate for 1,300 ms) comprising the two time series in the initial calculation of root mean square error (RMSE). The additional values (1,301–2,000 ms) were not used because the goal time series did not extend beyond 1,300 ms.