Abstract
The present study was designed to investigate the effect of amount of original training upon conceptual shift learning in kindergarten children.
A 2 (shift types)×3 (amount of original training)×2 (number of instances item condition) factorial design was used. The Ss were 120 kindergarteners with a mean age of 5 years and 9 months. The Ss were trained on a 2-conceptual category (e. g. animals and fruits), successive sorting problem to reach one of the 3 following criterions: 4, 8 successive correct responses, and 8 successive correct responses+24 responses. The Ss were given verbal informations,“correct” or wrong about their responses. After training, half of the Ss were given a full-reversal shift problem, and the remaining Ss half-reversal shift problem to reach a criterion of 8 successive correct responses.
The Ss in full-reversal shift condition were required to change their mode of responding from training to shift entirely, while those in half-reversal shift condition required to change only half of previous mode of responding in shift problem.
The main findings were that (a) positive effect of amount of original training was found in fullreversal shift, but negative effect in half-reversal shift,(b) conceptual organization was increased as a function of amount of original training, and (c) as for full-reversal shift, 8-item-condition was more difficult than 4-item-condition in 4 correct criterion, but performances in both item conditions didn't differ significantly in 8 and 8+24 correct criterions.
The results were interpreted as supporting the hypothesis that overtraining strengthened conceptual mediating response and were discussed with reference to related studies.
