Vicarious reinforcement and imitative learning.

VioLit summary: OBJECTIVE: The intent of this article by Bandura et al. was to explore the effects of vicarious reinforcement on conditioned emotional responses among nursery school children with a particular focus on modeling aggression. METHODOLOGY: The authors employed an experimental design by randomly assigning 40 boys and 40 girls from the Stanford University Nursery School into one of 2 experimental groups or one of 2 control groups. The mean age of the children was 51 months, with a range between 38 and 63 months. The children witnessed a 5 minute televised program which was either: 1) The Aggressive Model-Rewarded condition, in which a man's aggressive verbal and physical behavior in a play room were accompanied by a sense of victory and mastery over another man and his environment. 2) The Aggressive Model-Punished condition, in which a man's aggressive verbal and physical behavior in a play room (identical to the scenario in the previous model) were accompanied by severe punishment. 3) The Nonaggressive Model-Control group, in which the objects in the play environment were unchanged, but they were utilized by both characters in vigorous but nonaggressive play. 4) The Nonexposure Control group received no exposure to any of the models. The children were then tested for the incidence of postexposure imitative and nonimitative aggressive responses, while observed in an experimental play room. Several objects were available in the play room that could be used to reenact the aggressive behavior (baton, balls, Bobo dolls, hoola hoop, lasso, dart guns, cars, plastic farm animals) as well as several objects that tended to elicit predominantly nonaggressive responses (blackboard, doll house, cotton-stuffed dolls, building blocks). Each child spent 20 minutes in the play room, with only a noninteractive adult present (to ensure the child's comfort in a strange environment), and was rated by a hidden evaluator. Scores were obtained in predetermined categories of aggression which matched either those in the experimental aggressive models (kicking, lassoing or striking the Bobo doll with a ball or baton, shooting darts at cars or plastic farm animals, or repeating the verbal aggression which was displayed in the models) or those in predetermined categories of nonmatching aggression (punching or slapping the Bobo dolls, crashing the automobiles, acting out physical attacks toward the dolls or animals, etc.). A second rater witnessed 11 of the 80 children to provide an estimate of interrater reliability, and neither rater had knowledge of the treatment conditions to which the children had been exposed. The interscorer reliabilities were high, with product-moment coefficients in the .90s. After the rating each child was asked to discuss the behavior of the characters in the program and to identify the characters he or she preferred to emulate. FINDINGS/DISCUSSION: Many of the children, especially the girls, received low imitation scores and therefore nonparametric techniques were employed to evaluate the significance of the differences which were obtained. Results from the Kruskal-Wallis analysis of variance indicated that imitation was significantly influenced by response-consequences to the model (rewarded versus punished p AUTHORS' RECOMMENDATIONS: Additional research, argued the authors, should be implemented in which the difference between performance and learning can be better assessed, in order to identify the depth of impact from the vicarious transmission, rather than the just behavioral manifestation of the vicarious transmission. (CSPV Abstract - Copyright © 1992-2007 by the Center for the Study and Prevention of Violence, Institute of Behavioral Science, Regents of the University of Colorado) KW - California KW - Early Childhood KW - Child Aggression KW - Preschool Student KW - Reward KW - Punishment KW - Social Learning KW - Aggression Causes KW - Imitation Language: en