Observational learning: A tool for the acquisition of new motor skills

It is generally agreed that the first determinant of motor learning is physical practice. However, physical practice is not always a suitable first step or is not always possible. This is the case, for example, when the task involves a certain degree of danger or when an injury requires that certain skills be relearned. In such instances, observation of a model performing the task may prove to be beneficial for learning or to reduce the amount of physical practice needed to reach proficiency (Ferrari, 1996; McCullagh, 1994). In fact, research indicates that observation facilitates learning of a large variety of tasks such as action pattern production (Carroll et Bandura, 1982), coincidence anticipation (Blandin et Protean, 2000; Weeks, 1992) and gross (Southard et Higgins, 1987) as well as fine motor skills (Pollock et Lee, 1992). Sheffield (1961) and more recently Bandura (1986) proposed that observation of a model results in the observer developing a standard of reference of the task to be learned (named a much less than perceptual blueprint much greater than by Sheffield) as well as of mechanisms for the detection and correction of errors. These hypothesized mechanisms are not totally estranged to mechanisms believed to develop during physical practice-and to be responsible for motor learning. In fact, it has been hypothesized that observational learning and learning through physical practice might be mediated by similar cognitive processes (Adams, 1986). The goal of this contribution is to present the current research on observational learning in order to test this hypothesis and to determine the best conditions to favor observational learning.