Beneficial and detrimental effects of social observational modeling on generalized creative response production
Date
1977
Authors
Mueller, Ludwig Karl
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Typically, "creativity" in the person has been viewed as an individual trait, manifesting itself in various forms and measurable behaviors. However, there exist literature reports analysing influences capable of enhancing creative production through various applied, external forces. Specifically, these are attempts to "train" the individual to yield greater and more unique creative products through various diverse methods. As such, these attempts all strive to alter creative production in the direction of enhancement, through the application of educational or direct training procedures congruent with the various existing notions of creativity.
The present study attempted to demonstrate the effects of social observational modeling on creative production, without resort to any deliberate "strategy-taught" educational or direct reinforcement procedures. It was postulated that such effects may not be only beneficial to the subsequent productive tendencies of the subject where creative responses are concerned, but that effects detrimental to creative behaviors may also occur in the presence of highly uncreative models.
Experiment 1 assessed these modeling effects concurrently across two different figural and verbal tasks attempting to elicit creative responses. One hundred and twenty female introductory university students were randomly divided into 3 groups. Group 1 observed a model (experimenter) producing highly creative response objects on identical figural items (Repeated Figures tasks of the Torrance Tests of Creative Thinking, 1966) which Ss had performed on in an initial pre-test condition. Group 2 observed similar model responses, but which were highly uncreative in nature. A third control group did not observe model performances. In a completely counterbalanced design, alternate forms of these tasks were presented to Ss in a subsequent post-test condition.
Results showed significant modeling effects generalized over the two different figural tasks. Relative to control conditions, Ss' post-task responses were found to be congruent in direction with the model's creative (or uncreative) levels of response. These results were analysed in terms of total creativity and the component~ creativity elements of fluency, flexibility, originality, and elaboration as outlined by the Torrance tests.
To assess whether the modeling effects in the figural modality were generalizable to the verbal modality, the verbal (Unusual Uses) tasks of the Torrance Tests were also presented concurently in the identical counterbalanced conditions that the tasks of the figural modality had been presented. However, no model performances were presented between the pre-task and post-task conditions. Generalization from the figural to the verbal modality was not found to have occurred.
It was concluded that observational learning had an effect on creativity in the figural modality, exerting both beneficial and detrimental influences on creative production. Failure to generalize to the verbal modality indicated some specificity of modeling influence in creative production. It was suggested that where educational goals are aimed at increasing creativity in many forms and across many modalities, the model or educator must himself display all his behaviors in creative and unique manners across many different modalities to be effective.
Experiment 2 attempted to assess whether true modeling effects had taken place, or whether simple "mimicry" of model responses had occurred. Four peer judges were found to have independently labelled, about equally, responses of Ss who had and who had not seen prior model performances, as having simply been copied to the model displays of the study. This indicated that model displays resembled in nature responses Ss would normally make without model influence, varying only in degree of creativity. Simple copying was therefore not judged to have occurred to any greater degree in responses of Ss after model exposure than to responses where indeed no model had as yet been presented. A subsequent measure of inter-judge reliabilities was found to be very low, indicating little inter-judge agreement as to what constituted similar or copied creative or uncreative acts, and reflecting the complexity of what is to be defined as creative.