EXTINCTION IN DISCRIMINATION LEARNING: PRESENTATION AND CONTINGENCY VARIABLES AND ASSOCIATED SIDE EFFECTS.
Abstract
The purpose of the study was to assess the effects of two methods of stimulus presentation (fading and constant) and two response contingencies (delay and no-delay) on the course of discrimination learning; and to examine the effects of response histories on various aspects of discrimination performance. In the fading procedure S- was gradually faded along the dimensions of brightness and time, and in the constant procedure S- maintained a constant value throughout training. Subjects trained with the constant procedure made significantly more S- responses in the course of acquisition than subjects trained with the fading procedure. In the constant-delay procedure, S- responses were followed by a delay in the offset of S-. In the constant-no-delay procedure S- responses were followed by an intertrial interval. Subjects trained with the constant-no-delay procedure made significantly more responses to S- than subjects trained with the constant-delay procedure. The amount of extinction which occurred during learning was highly correlated with intertrial responding and the stability of the learned discrimination. Discrimination reversal learning was also a function of original learning history. The results were discussed in terms of the differential processes which underlie discriminative performance when different training procedures are used. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 1968
- Accession Number
- AD0667656
Entities
People
- James G. Holland
- Miriam Cohen
- Robert Glaser
Organizations
- University of Pittsburgh