Evidence for an error detection mechanism in motor skills

One of the fundamental predictions of the Adams (1971) closed-loop theory is that S should develop over practice a perceptual trace which forms the basis of an error detection mechanism. If knowledge of results (KR) is withdrawn, S can use the error detection mechanism to guide performance in its place. To test this notion, 10 Ss practiced 170 trials of a discrete ballistic response, attempting to produce a 9.5-in. manual slider movement in exactly 150 msec., with KR present except on Trials 11-20 and 141-170. There was an increasing correspondence between the actual errors and Ss' estimation of them as practice continued, and ratings of confidence increased, both of which were indicative of a developing error detection mechanism. During KR withdrawal trials, Ss displayed no worsening of performance, and even continued to learn slightly, suggesting that the error detection mechanism acted as a substitute for KR. The results generally supported Adams' theory.