Psychology ETDs


E. Kim Smith

Publication Date



Two experiments on the role of stimulus recoding in the A-B, A-C paradigm are described. In Experiment I, complexity (number of points) was held constant and number of relevant cues was varied. Subjects were divided into two groups: one learned a paired associate task involving 1-cue, 24-point high similarity stimuli and the other learned a paired associate task involving 2-cue, 24-point high similarity stimuli. Both groups were then given a transfer task in which the number of correct responses was recorded. It was found that the subjects in 2-cue stimulus condition performed much better than the subjects in the 1-cue condition. These results support (but do not demand) a switching hypothesis which explains the transfer by positing that the subjects switch encodings from one of the cues to the other. Experiment II examined the recording process directly. Subjects were divided into three groups and given paired-associate pretraining on 24-point, high similarity, random shape stimuli. Of the two groups that trained on one stimulus set, one was given a probe at the end of task one learning in order to locate the subjects’ encoding of each stimulus. The other group received no probe and acted as a control over the probing operation. The third group was trained on a comparable set of 24-point stimuli and acted as the C-D control. All groups were given the second task and the number of correct responses was recorded. The probe was regiven at the end of transfer and the encodings from the first and second task were compared. Massive positive transfer resulted from the two A-C conditions when compared to the C-D condition. However, it was found that within the same subject, recording (or switching codes) resulted in much poorer performance that did staying with the same encoding. The results suggest that recording is not the mechanism responsible for positive transfer as Martin (1968) had suggested. This finding is integrated with other findings to show that Martin and Walter’s (1970) revised hypothesis is also not sufficient. No theoretical explanation is offered in the body of the paper. However, the author does offer the beginnings of a theoretical explanation of his own in the Appendix.

Degree Name


Level of Degree


Department Name


First Committee Member (Chair)

Henry Carleton Ellis

Second Committee Member

Peder Jack Johnson

Third Committee Member

David Theodore Benedetti



Document Type


Included in

Psychology Commons