Learning a pressure under conditions of immediate and delayed knowledge of results

The results of a series of experiments on learning to reproduce a precise pressure are described and discussed in relation to a tentative hypothesis about the importance of timing of feedback signals. The evidence supports the contention that a careful analysis of the perceptual aspects of performance is necessary to account for the acquisition and retention of a precise motor response. Whilst the original hypothesis is sustained, other factors leading to the same result are clearly important. New evidence is produced which strongly suggests that cues presented simultaneously in two modalities interact. A possible distinction between sensory feedback which is used to control a response and feedback which cannot so be used, and the relation of this distinction to the learning of the response is also discussed.