The Department of Psychology
Abstracts of Posters presented at the Convention of the Association for Behavior Analysis
GENERALIZED SELELCTION Of INTRA-STIMULUS RELATIONS
AMY STIRMAN (& Barry Lowenkron)
Department of Psychology
The first goal of this study was to gather evidence showing that there are matching to sample tasks in which the correct selection of a comparison requires some response to the comparison besides the selection response (e.g., pointing). This evidence is important because current conceptions of stimulus control in comparison selection tasks have the comparison stimulus functioning as an SD that evokes no behavior aside from the comparison selection response(e.g., pointing) itself.
But, as has been shown elsewhere (Lowenkron, 1984, 1988, 1989, 1991, 1992) such an account cannot explain the origin of generalized responding based on consistent relations between stimuli. On the other hand, the notion of comparison selection under joint control predicts that where generalized responding occurs, there are other, verbal responses to the comparison stimulus besides selection. Thus, the first point of the study is to collect data showing that other, supplementary responses to the comparison, besides pointing, must occur before generalized responding can appear, while, the second point of the study is to examine the nature of the generalized responding so produced.
THE EMERGENCE OF BIDIRECTIONALITY
SHARON LEE (& Barry Lowenkron)
Department of Psychology
In the course of learning word-object selection during a matching task, when does the bi-directional object-word relation develop? One possibility is that it is a result of some automatic process of symmetry. Another possibility is that bidirectionality is the result of the covert rehearsal or other operant behaviors.. Thus, when the subject emits a correct response to a verbal stimulus, they may repeat the two of them by way of rehearsing the response to the stimulus. If the emergence of bidirectional relations is indeed the result of such rehearsal, then bidirectionality should be inhibited to the extent this rehearsal is inhibited.
As a preliminary study here, subjects were trained to emit each of a series of consonent-vowel-consonent trigrams in response to other such trigrams with which they were paired. For one group of subjects a double-function list was constructed so that all trigrams functioned both as stimuli for some trigrams, and as responses for others. As a result, rehearsal to form backward associations within each pair would interfere with forward learning. After subjects learned to emit appropriate responses to each of the stimulus trigrams in this list, they were shown the response member of each pair and asked to produce the stimulus member of the pair. The procedure was then repeated with a single-function list comprised of other, novel trigrams. In this list each trigram appeared once: either as a stimulus member or as a response member of a single pair. Again, after learning to emit the response member of each pair when shown the stimulus member, the subjects were shown each of the response members and asked to provide the stimulus member. A control group was only given training and tests with the single-function list.
The results showed that the control group emitted more accurate responses in the test than the experimental group with its prior exposure to a double-function list. These data suggest that prior exposure to the double function list inhibited the rehearsal of backward associations in the single-function list thereby indicating that symmetry may be the result of covert operant rehearsal rather than an innate process.