THE USE OF CUING IN TRAINING TASKS: PHASE 3
Abstract
The present study investigates three potentially trainable discriminations, the relative pitch, intensity and duration of pairs of signals embedded in noise. These are relevant to the active sonar cues for doppler, echo strength and echo length respectively. It is emphasized that no statement on trainability can be final until all methods reasonably likely to be successful have been investigated. This study compares three methods, (a) cuing, in which the trainee is informed just before each pair of signals what the correct response should be, (b) knowledge of results in which he gets the same information just after hearing the signals and making his responses and (c) a mixed cuing/KR condition in which training begins with cuing and later transfers to KR. The results show that all three methods are effective for pitch and intensity judgments but that none is effective for judgments of relative duration. Training effects, although significant, are generally small and occur early in practice. The results confirm the suggestion that previously found differences between cuing and KR are due to manipulation of response criterion rather than differential effects on sensitivity and they confirm the view that learning in these cases is best described by a simple associative model rather than an S-R reinforcement model.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1967
- Accession Number
- AD0657367
Entities
People
- John Annett
- Laura Paterson
Organizations
- University of Hull