Reinforcement Effects on Participation in Group Discussion

Abstract

Research studies in verbal operant conditioning have recently been reviewed by Krasner (1958) and Salzinger (1959). These reviews indicate that numerous classes of verbal behavior have been found to be sensitive to a variety of reinforcing stimuli, both verbal and non-verbal. Verbal reinforcers used have, for the most part, been simple utterances, such as right, wrong, good, rnrn-hmm, see, fine, and other such terms denoting agreement or disagreement. Non-verbal reinforcers used have ranged from gestural cues such as head nodding, smiling, attentive looks, and postural changes, to mechanical cues including poker chips, lights, buzzers, and bell-tones. The response or class of verbal behavior reinforced has been equally varied. Many studies have used the technique first employed by Greenspoon (1955). In this procedure S is told to say and a certain class of words, such as plural nouns, verbs, etc., is reinforced. The largest group of studies has employed a technique devised by Taffel ( 1955 ). In this technique, S is presented with a series of 3-X 5-in. cards, on each of which is written a verb and several pronouns, and is instructed to make up a sentence using the verb and beginning with one of the pronouns. The class of behavior reinforced has usually been sentences beginning with I or we. Other investigations have systematically reinforced certain classes of response in test-like situations. These include human responses (Fahmy, 1953) and movement responses (Wickes, 1956) on ink-blot tests, attitudes toward general education on a questionnaire (Hildum & Brown, 1956), acceptance of self responses on a personality test (Nuthrnan, 1957), emotional words in imaginative stories given to TAT-like drawings (Krasner, 1959), and Si scale items from the MMPI (Oakes & Droge, 1960). While the reinforcers and the classes of behavior investigated have varied greatly in these studies, investigators have with few exceptions employed rather simple laboratory situations involving one S and an E. Few studies have been reported extending the verbal operant conditioning paradigm to group discussion situations. Further, many of the techniques used resemble the controlled operant type, in that the experimental design necessitates individual trials, as for example in the Taffel procedure (Taffel, 1955). Two recent studies have utilized the free operant conditioning paradigm in group situations. 'This investigation was supported by Research Grant No. M3212 from the National Institute of Mental Health, Public Health Service.