An experimental analysis of maze discrimination.

The non-correction method and the delayed response problem of Blodgett and McCutchan.

“…Other rats had errorless days, but had not reached the criterion when on the 22nd day, after a 4 day plateau, training was discontinued. Thus, the learning manifested by this group was inferior to that of the groups run in Experiment I (Non-correction Method) and Experiment IV (Contrast by Temporal Contiguity) , 6 but it was definitely better than that of the group run in Experiment III (Correction Method with 6 trials a day). Also, the behavior of the rats run in this experiment was quite different from that of the rats run in Experiment III.…”

“…Fio. 6 ment is too difficult, if not impossible, to perform), it is impossible to discover if contrast by temporal contiguity alone is sufficient.…”

“…Thus, an analysis of the differences between the correction and the non-correction methods seems the essential problem for an understanding of the difference between Blodgett and McCutchan's results and those obtained in Experiment I. 5 'Both Seward (6) and Hull and Spence (3) have investigated the non-correction method and its relation to the correction method in the learning of a simple T-maze. However, both of these discussions are irrelevant to the issue in question.…”

Discrimination performance as affected by training procedure, problem difficulty, and shock for the correct response.

“…The isolation of problem difficulty as a dimension influencing shock-right facilitation suggests that previously re-ported differences may relate to a greater difficulty of the correction task as a whole. In line with this interpretation, early studies comparing correction and noncorrection have shown the former method to produce significantly more errors (Hull & Spence, 1938;Kalish, 1946;Seward, 1943).…”

The Methods and Findings of Experiments on the Visual Discrimination of Shape by Animals

“…Unfortunately, there are no systematic studies of the efficacy of these two methods in teaching shape discrimination: little difference in efficacy has been found where the two methods have been compared in situations involving the learning of spatial discriminations, and the direction of the difference varies with the experimental situation used (Hull and Spence, 1938;Kalish, 1946;Seward, 1943). It is possible that in the jumping stand where position habits develop easily the correction procedure might have some advantage since it would tend to break up such habits the rat having responded to the card in the wrong window is forced to go on responding until it selects the other window : it cannot obtain reward 50 per cent, of the time by jumping consistently to one window.…”