アルビノラットにおける学習セット形成と異次元転移効果

平野 信喜

doi:10.4992/jjpsy.50.63

THE FORMATION AND EXTRADIMENSIONAL TRANSFER EFFECT OF LEARNING SETS IN ALBINO RATS

NOBUYOSHI HIRANO

Author information

JOURNAL FREE ACCESS

1979 Volume 50 Issue 2 Pages 63-72

DOI https://doi.org/10.4992/jjpsy.50.63

Details

Published: July 20, 1979 Received: March 29, 1976 Available on J-STAGE: July 16, 2010 Accepted: - Advance online publication: - Revised: -
Correction information
Date of correction: July 16, 2010 Reason for correction: - Correction: ABSTRACT Details: Wrong : The purpose of the present experiments was to show the evidence for both interproblem transfer effect (learning set phenomenon) and extradimensional transfer effect in the albino rats using a series of both pattern and black-white discrimination problems.
In Experiment I, 20 albino rats were divided into two groups of 10 each. The Group LS rats were given 10 independent pattern discrimination problems (Fig. 2) in succession in a Y-maze, 100 trials per problem, 20 trials per day. Each rat in the control group (Group TR) was yoked with each of the Group LS rats regarding the reward or nonreward and the pattern-stimuli exposed on any particular trial. But for this group there was no consistent relationships between the pattern-stimulus and reward or nonreward. Both groups were, then, given a 20 trial black-white discrimination problem as an extradimensional transfer task.
The results were as follows: (1) There was clear evidence of interproblem transfer effect or the LS phenomenon in Group LS (Fig. 3). (2) The Group LS showed evidence of discrimination in the transfer task but the Group TR did not. (3) The Group TR developed a marked position preference during the first stage of training, which was (negatively) transferred to the second or the transfer task and thus did not provide an adequate control level for the Group LS during the transfer training. (4) The evidence for the extradimensional transfer effect was inconclusive for the reasons given in (3) and the other.
In Experiment II, 30 rats were divided into three groups. The Group LS rats were run under approximately identical conditions as in Experiment I. Two control groups were prepared: For the Group SP, forced trials were used so that the number of trials to both left and right goal boxes and to both stimulus patterns presented simultaneously were equalized. For the Group P rats, two gray cards were used instead of pattern cards and the forced trials were used in order to equalize the number of left and right runs. Each of the Group SP and P rats was yoked with each of the Group LS rats and the trial on which reward or nonreward was given was the same for all groups.
All groups were then given the pattern and black-white discrimination tasks, 200 trials for each. Half of each group underwent the pattern discrimination first and the black-white discrimination second, and for the other half, the order was reversed.
The results were as follows: (1) Again there was clear evidence of the LS phenomenon in the Group LS (Fig. 3). (2) The evidence of the extradimensional transfer effect in the Group LS was clear in comparison with the transfer task results of Groups SP and P. (3) The subjects simply exposed to a series of pattern stimulus cards without opportunities for discrimination (Group SP) showed significantly better pattern discrimination learning during the transfer task than the subjects exposed to gray cards (Group P).
The present experiments as a whole clearly indicated that the LS phenomenon commonly observed in higher mammals was ascertained in the albino rats using pattern discrimination tasks with a fix-trial method. Theoretical discussions of the present results were attempted with reference to Harlow's and Reese's theories on LS.
Right : Two experiments were run in order to ascertain the learning set (LS) phenomenon in rats. The task was a series of two-choice pattern discrimination problems in Y-maze and 100 trials were given per problem. In both Experiments I and II, the rats' correct responses gradually increased over ten independent pattern discrimination tasks, thus showing interproblem transfer effect or the LS phenomenon. In Experiment II, there was evidence of extradimensional transfer effect; rats exposed to a series of pattern discrimination problems showed better performance in a subsequent black-white discrimination task. Theoretical discussions of the results were attempted with reference to Harlow's and Reese's theories on LS.

Abstract

Two experiments were run in order to ascertain the learning set (LS) phenomenon in rats. The task was a series of two-choice pattern discrimination problems in Y-maze and 100 trials were given per problem. In both Experiments I and II, the rats' correct responses gradually increased over ten independent pattern discrimination tasks, thus showing interproblem transfer effect or the LS phenomenon. In Experiment II, there was evidence of extradimensional transfer effect; rats exposed to a series of pattern discrimination problems showed better performance in a subsequent black-white discrimination task. Theoretical discussions of the results were attempted with reference to Harlow's and Reese's theories on LS.

Corresponding author

Correction information

Register with J-STAGE for free!