Jack Capehart

An application of adaptation-level theory to transposition responses in a conditional discrimination

Adult human Ss were required to learn a conditional discrimination in which the choice response of “left” or “right” was made to a series of two lifted-weights stimuli. The correct response to the lighter member (100 g) was left while right was correct for the heavier (200 g). Upon completion of the learning of this discrimination Ss were presented a test series of seven weights which ranged from 160 to 450 g. The training stimulus of 200 g was one member of the seven test series. The A-L theory prediction that Ss would respond left to weights lighter than the calculated A-L, and right to weights heavier than A-L, was strongly supported by the data. The prediction that Ss would shift their choice response from right to left for the 200 g weight when it was a member of the test series was also confirmed.

Generalization of a voluntary response as a function of presentation frequency of the training stimulus in testing

Frequency of presentation of the original training stimulus in the test phase of a study of human voluntary generalization was found to affect the form of the generalization gradient on a weight dimension. That is, in a situation in which all of the additional test stimuli are larger than the original training stimulus, and all stimuli are presented with equal frequency, a tendency to choose stimuli toward the middle of the test range as the original was observed However, if the original stimulus was presented more frequently than any of the other stimuli in testing, Ss tended to choose the original “correctly,” yielding a “typical” unidirectional generalization gradient.

Decision speed for the choosing of the larger or the smaller of two digits

A recent published report suggests that numbers may be processed internally as analog magnitudes when Ss are required to select the larger of two digits. If this is the case then it is suggested that Ss should be able to select larger digits more quickly than smaller digits. Experimental investigation confirms that the choosing of larger digits is faster than the choosing of smaller digits.

Attentional processes in overtraining.

The effects of overtraining were examined using two dependent variables, choice response and eye fixations. Although an overtraining reversal effect was nor found using choice-response, examinations of a number of eye fixations both before and after reversal suggest that the introduction of reversal training produces greater shifts in attention in the overtrained (n = 6) as opposed to the criterion trained group (n = 6). Results are interpreted as supporting a theory of discrimination learning which proposes that S is developing a plan in which inputs are tested or compared against some centrally organized process.

Human successive discrimination reversal: Effects of overtraining and reinforcement

The overtraining reversal effect (ORE) was found in a human successive discrimination task. Two types of reinforcement were used, with three levels of training on the original problem. The ORE appeared with the first level of overtraining and then became practically asymptotic with additional overtraining.

The overtraining reversal effect under two types of discrimination

The effects of overtraining were examined under both simultaneous and successive discrimination training. A significant overtraining reversal effect (ORE) was found. Further examination revealed a trend toward an ORE in the simultaneous condition which failed to reach significance and a significant ORE in the successive condition. Ss were 40, six to seven year old children. The ORE is examined in light of Lovejoy’s (1966) mathematical model.

INFORMATION PROCESSING VS VISUAL ATTENTION DURING INCIDENTAL LEARNING

The relationship between attention and incidental learning during discrimination training was studied in 30 children, aged 10 to 11. A polymetric eye-movement recorder measured direct visual attention. Consistent with previous findings, recall of incidental stimuli was greatest during the initial and terminal stages of intentional learning. Contrary to previous explanations, however, visual attention to incidental stimuli was not related to training. While individual differences in attention to incidental stimuli were predictive of recall, attention to incidental stimuli was not related to level of training. Results suggested that changes in higher order information processing rather than direct visual attention were responsible for the curvilinear learning of incidental stimuli during intentional training.

A TRANSFER APPROACH TO THE ORE

The present study was conducted to test Lovejoys attentional hypothesis concerning the effect of difficulty in a discrimination task on the overtraining reversal effect (ORE). Results indicated that either criterion or overtraining on a difficult problem produced efficient reversal acquisition. However, significantly rapid reversal performance only occurred when overtraining was combined with a difficult discrimination. Lovejoys approach to the ORE was supported.

POSTCRITERION INSERTION OF A SECOND RELEVANT DIMENSION IN A SIMPLE DISCRIMINATION PROBLEM

Thirty-one human Ss participated in a discrimination problem having three stages. Stage I involved one relevant dimension and one irrelevant dimension. Ss were trained to a criterion of eight successive correct responses. After reaching criterion on Stage I, Ss began Stage II, in which both the relevant and irrelevant cues from Stage I were maintained, and one more relevant and one more irrelevant dimension were added. At this point, it was possible for Ss to solve the problem by responding to the “old” relevant cue, the new cue, or both. Certain theories predict a sampling of the inserted cues; others predict they will not be sampled. In Stage III, the relevant and irrelevant cues from Stage I were dropped out. Solution of the problem could be gained only by responding to the cues inserted in Stage II. A control group was overtrained on Stage I and then went directly to Stage III. The data supported theories predicting sampling of all cues in Stage II.

Spence’s Theory of Discrimination Learning: A Re-Evaluation

Spence’s theory of discrimination learning (1936, 1937) which was orginally proposed to account for the phenomenon of transposition, is, when combined with certain postulates of the Hullian system, capable of dealing with an impressive range of empirical data from the related areas of stimulus generalization, transposition, and discrimination learning. A new class of study introduced by Hanson, (1957, 1959) based upon the comparison of generalization gradients obtained with and without prior discrimination training has, in the authors’ view, been a recent source of embarrassment for the Hull-Spence approach. The present paper reviews a number of these and similar studies, and concludes that the Hull-Spence formulation is now confronted with such grave negative evidence that it requires extensive modification in order to remain a viable theory.