This paper reviews theoretical developments in the study of discrimination learning in relation to 2 major contemporary viewpoints-a perceptual orientation as expressed in differentiation theory and a learning orientation as expressed in mediation theory. Early as well as present theory and research are found to have consistently emphasized response processes, particularly those of a mediational nature, in accounts of discrimination learning. This line of development is seen as tending to have precluded proper consideration of the nature and role of the stimulus. When viewed in this historical context, the conceptions of differentiation theory are seen as promoting a desirable theoretical and empirical balance.
Experimental Ss (Ist-grade children) received in succession (a) nonreinforced pretraining in making same-different judgments to stimuli varying in height and brightness, (b) a reinforced-discrimination task in which stimuli varied simultaneously in height and brightness, and (c) either a reversal shift of the initial discrimination (J Ss) or a nonreversal shift (J Ss). Control Ss received the same discrimination task and shifts but preceded by a picture-completion and a picture-arrangement test. Ss in the experimental-reversal group required significantly fewer trials to reach criterion than Ss in the experimental-nonreversal and control-reversal groups. The control-reversal and control-nonreversal groups did not differ. The results are interpreted within a differentiation theory of discrimination learning.
Following training to respond to 1 dimension of a 2-dimensional discrimination, rats and 3-and 4-yr.-old children had the option of learning a second discrimination by executing either a reversal or an extradimensional shift, or by responding nonselectively. Rats received 0, 200, or 300 overtraining (OT) trials in the initial discrimination, and children 0 or 100 OT trials. OT increased the proportion of optional reversal shifts (RS) in children but tended to decrease it in rats. Optional RS was relatively infrequent in rats as compared to children. In both species preference for RS varied with the dimension relevant in the initial discrimination.
The discrimination shift behavior of children was studied within a 2X2X2 factorial design which varied age of S (4 yr. vs. 10 yr.), type of shift (reversal, RS, vs. extradimensional, EDS), and state of the irrelevant dimension during shift (constant vs. variable). The 10-yr. olds did not differ in speed of learning RS and EDS under the condition of a constant irrelevant dimension but learned RS faster than EDS when the irrelevant dimension was variable. The 4-yr. olds accomplished EDS more rapidly than RS regardless of the state of the irrelevant dimension. Both types of shift were learned more slowly by the 10-yr. olds when the irrelevant dimension varied than when it was constant, but the shift performance of the 4-yr. olds was not significantly affected by this variable.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.