Psychology ETDs

Publication Date

5-22-1970

Abstract

In Experiment 1, 32 rats were pretrained to press a lever for a .045 gm. Noyes food pellet either in the presence of a light or a tone. For half of the Ss within each group, responses were reinforced 100% or 25% of the time. After 16 days of pretraining, a novel stimulus (light or tone) was simultaneously compounded with the pretraining stimulus for another 16 days. Probability of reinforcement continued at the pretraining value for half of the Ss within each group. The remaining Ss had their reinforcement probability increased (from 25% to 100%) or decreased (from 100% to 25%). Two additional groups of Ss, without pretraining experience, receive compound training with a 100% or 25% reinforcement probability in effect. Following compound training, two days of testing were given in which the components of the compound were presented individually. All responses made in the presence of the components were nonreinforced. Test results indicated that while all pretrained Ss responded in the presence of the pretrained component, only Ss who had their reinforcement probabilities changed at the onset of compound training also responded in the presence of the added component. The Ss with no pretraining experience responded in the presence of both components. In Experiment 2, 24 rats receive pretraining experience equivalent to that in Experiment 1 except that responses in the presence of the pretraining stimulus were reinforced 50% of the time. After ten days of pretraining, each group of Ss was divided into three subgroups and given ten days of transfer training. For all Ss, the stimulus not present during pretraining was simultaneously compounded with the pretraining stimulus on those trials in which responses were reinforced during pretraining. On those pretraining stimulus trials in which responses were never reinforced, the added stimulus was compounded with the pretraining stimulus either 0%, 25%, or 75% of the time. Thus, the added stimulus was a more reliable predictor of reinforcement than the pretraining stimulus. An additional 24 Ss, without pretraining experience, received transfer training experience identical to that of the pretrained Ss. Terminal transfer training results indicated that all Ss made a greater percentage of responses in the presence of the compound than in the presence of the pretraining stimulus presented alone. The rate at which this discrimination was formed was an increasing function of the reliability of the added stimulus. For all groups, Ss without pretraining experience learned this discrimination significantly faster than pretrained Ss. A conceptual account of the results of Experiment 1 and 2 is offered. The major assumptions are: (a) increments in signal value strength to a stimulus are a function of the degree to which other stimuli predict the goal event; and, (b) increments are a function of the degree to which a particular goal event is predicted.

Degree Name

Psychology

Level of Degree

Doctoral

Department Name

Psychology

First Committee Member (Chair)

Frank Anderson Logan

Second Committee Member

G. Robert Grice

Third Committee Member

Douglas Peter Ferraro

Fourth Committee Member

Peder Jack Johnson

Language

English

Document Type

Dissertation

Download

Included in

Psychology Commons

Share

COinS