Place, U. T. (1992b). Is there an operant analysis of animal problem-solving? [Conference presentation, presented at 18th Annual Convention of the Association for Behavior Analysis - May 25-28 1992 - San Francisco, California]. Association for Behavior Analysis. Abstract published in Proceedings of 18th Annual Convention of the Association for Behavior Analysis - May 25-28 1992 - San Francisco, California (p. 155). Society for the Advancement of Behavior Analysis.
[Abstract]In 'An operant analysis of problem-solving', Skinner (1966/1969/1988) develops an account of problem-solving based on the distinction between two different ways in which an organism can learn to adapt to environmental contingencies: (1) contingency-shaped behavior in which the behavior of an organism is progressively shaped by repeated exposure to the contingency itself, and (2) rule-governed behavior in which a verbally competent human being adapts to a contingency by constructing a verbal formula or rule which is said to "specify" the contingency in question. A rule may be constructed, as in the case of contingency-shaped behavior, in the light of repeated exposure to the contingency itself. It may equally well be based on information about the contingency supplied by another speaker, on information derived from a written text, or on an inference from other rules derived from any or all these sources. It is this case where the agent infers a new rule tailor-made for the problem with which he/she is confronted that Skinner has in mind in offering an analysis of problem-solving in these terms.
There is a growing body of empirical evidence (Hayes 1989) which confirms the accuracy of Skinner's description of problem-solving as it occurs in the case of verbally competent human beings. But animals also solve problems; and so do pre-verbal human infants. This kind of problem-solving cannot simply be a matter of contingency-shaping, though previous contingency-shaped behavior is the only resource from which a pre-verbal organism can draw in selecting an appropriate problem-solving strategy. It requires some mechanism like that which Köhler (1925) refers to as "insight" whereby the stimulus class which currently controls a particular response class is somehow stretched so as to include the current stimulus situation.
The case for postulating such a behavior mediating mechanism within the conceptual framework of radical behaviorism is argued by appealing
(a) to the analogy between attending behavior and thinking by talking to oneself, and
(b) to the process whose existence is implied by Skinner's (1938) account of "stimulus class" whereby an organism learns to break up its stimulus environment into stimulus classes "along the natural lines of fracture."
Keywords: the natural lines of fracture, stimulus class, rule-governed behaviour, problem-solving, Skinner
[References] [Talks] [1 referring publications by Place]
Download: 1992b Is There an Operant Analysis of Animal Problem-Solving.pdf