This article studied the effectiveness of therapeutic modeling by using models similar or dissimilar to observers in a class of behavior being modified (fear) and a characteristic (age) seemingly irrelevant to the modeled response. Forty second-and third-grade girls who showed behavioral fear of snakes viewed models of one of four types: fearless child, fearful child, fearless adult, fearful adult. In addition, 10 subjects saw no model in a control condition. Premeasures and postmeasures of fear were taken on behavioral and attitude dimensions. With respect to overt avoidance, model similarity on the response dimension (level of fear) was an unimportant variable, but model similarity on the age dimension was important. With respect to attitudes, more similar models produced the greatest change regardless of the dimension on which similarity occurred. An integration of these findings with other research is presented.

No abstract

No abstract