A comparison of the contribution of perceptual modeling and knowledge of results to coincident-timing skill acquisition : an honors thesis [(HONRS 499)]

Abstract

While a few studies have compared the relative contribution of knowledge of results (KR) and modeling to learning of closed motor skills (e.g., McCullagh & Little, 1989; Ross, Bird, Doody, & Zoeller, 1985), the strength of each in contributing to open skill acquisition, specifically coincident-timing skill acquisition, is largely unexplored. Weeks (1991) has demonstrated that coincident-timing skill acquisition is accelerated if subjects receive pre-practice modeling with the perceptual demands of the task. Specifically, subjects passively viewing stimulus runway lights on a Bassin anticipation timer were more accurate on initiation of active practice than subjects simply initiating practice with KR. In these experiments, the groups receiving perceptual modeling also received KR; therefore, the relative contribution of modeling independent of KR is unknown. This study compared the contribution of perceptual modeling independent of KR in learning a coincident-timing task. Subjects (n=48) were randomly assigned to one of four groups: a modeling+KR group, a modeling only group, a KR only group, or a no modeling/no KR group. A Bassin timing apparatus was used to provide perceptual modeling and support measurement ofcoincident-timing ability. The task consisted of a 60 cm right-toleft arm motion to knock over a barrier coincident with the lighting of the final light on the stimulus runway. Groups receiving modeling viewed the lights ten times prior to the initiation of twenty active practice trials, while groups given KR (in ms early or late in displacing the barrier) received it after each trial. Results indicated that groups receiving perceptual modeling initiated practice with less absolute constant error (ACE) than the KR only group. However, the KR only group was more accurate later in practice than the group receiving only perceptual modeling. The no KR/no modeling group performed with significantly less accuracy throughout acquisition than any other group. It was concluded that perceptual modeling is relatively more important than KR early in practice, but was not sufficient to encourage improvement in mid to late acquisition. Instead, KR appears to be relatively more important than modeling in this stage of acquisition. Overall, optimal conditions for skill acquisition seem to arise when perceptual modeling is used in combination with KR.