Feedback in Response Recognition and Production

Abstract

The role of feedback in the development of recognition memory was examined by testing pairs of subjects; one subject performed the movement while the other listened to the movement in an adjacent room. Both groups of subjects developed recognition memory during training with KR, and recognition performance was virtually identical for both groups. In addition, previous experience with the sensory consequences of the movement led to improved response production during initial trials. A possible mechanism to account for this finding is proposed. Appreciation is extended to Richard A. Schmidt and Anne Marie Bird for helpful comments in earlier drafts of this paper.

On The Structure of Motor Recognition Memory

“…Analysis of variance indicated that there was a significant reduction in AE as a function of trials, F(18, 1 026)= 5.5, p < .001, confirming previous work on the learning without KR via error detection (Newell, 1976b;Zelaznik & Spring, 1976). In general, subjects reduced their AE scores by 24 msec over the first 10 no-KR trials.…”

“…These two experiments along with our previous research (Newell, 1976b;Zelaznik & Spring, 1976) demonstrate that subjects who receive listening experience with the criterion MT compared to subjects with no listening experience, and subjects who received variability in listening compared to no variability in listening, exhibited responses with less error on the first movement trial. Since sensory feedback does not play a role in the structure of motor recall memory, the theoretical independence of recall (response production) and recognition predicts that improved recognition processes should not have affected the response production on Trial 1.…”

“…It seems that this is primarily due to the use of the slow positioning response in KR studies where theoretically the subject could match the feedback from an ongoing response with the reference of the correct response so that no updating of the response on future trials was necessary. In fast movements, however, we (Newell, 1976b;Zelaznik & Spring, 1976) have recently supported the contention (Adams, 1971) that subjects can learn in the absence of KR via subjective reinforcement. Although there appears to be little question as to whether feedback is utilized during recognition or whether subjects can learn without KR, the underlying representation of the motor recognition process still is not understood.…”

“…Recently, we (Newell, 1976b;Zelaznik & Spring, 1976) independently developed an alternative paradigm to study motor recognition. Basically, we were interested in whether subjects were capable of developing an error detection mechanism without ever having produced an actual overt response.…”

“…The present investigation examined the nature of the representation in motor recognition memory, and determined if the learning without KR demonstrated by us (Newell, 1976b;Zelaznik & Spring, 1976) was due to a range of responses being used for the development of the response recognition process. As discussed earlier, the Adams (1971) and the Schmidt (1975) positions state that recognition memory is developed by fundamentally different processes.…”