The Proportion of Model's Critical Responses in Vicarious Verbal Conditioning

The effects of vicarious reinforcement and proportion of model' s critical responses upon verbal conditioning were investigated by a Taffel-type task. It was expected that the effect of vicarious reinforcement would be greater under the higher proportion(60%) of the model' s critical responses than the lower proportion(30%). Forty male and thirty-nine female undergraduate students served as subjects. Four different pronouns, I, YOU, HE, and SHE, and three verbs with different impression values were written on each of the 140 indexcards. The subjects were requested to pick out one of the pronouns and a verb on the card and to construct a sentense with them. Completing 20 operant trials, each subject was assigned to one of the four conditions of the model tape; 60% VR, 60% NVR, 30% VR, and 30% NVR. 60% VR and 60% NVR tapesinvolved 60% critical responses during 40 trials, and 30% VR and 30% NVR tapes involved 30% critical responses during 40 trials. Under these conditions subjects listened to the model's responses for ten trials at a time and then took his own turn for the next ten trials. The alternation of model and subject was repeated four times. After these acquisition trials each subject was run forty extinction trials without listening to the model's responses. The results showed that neither the effect of vicarious reinforcement nor that of the proportion of the model' s critical responses was significant. Dividing the subjects into higher and lower performers on the basis of their operant performance level, the data was reanalysed. It was found that the higher-operant subjects decreased their critical responses on the acquisition trials, while the lower-operant subjects increased them. These results were discussed in terms of the operant level and the degree of the difficulty of the task employed