The Functional Value of Realistic Attributions

A research program is described that revolves around the assumption that it is functional to make realistic attributions. First, the application of an attributional framework to questions of behavior modification is described. Second, it is argued that this application reveals limitations of basic attribution research. Theories about the antecedents and consequences of causal attributions have been treated as unrelated. Third, a theoretical model and empirical studies that unite theories about attributional antecedents and consequences is introduced. Fourth, it is demonstrated that a comprehensive psychotherapeutic system can be deduced from this integrative model, which is compatible with Beck's and Ellis' cognitive behavior modification. From this perspective, research on attributional biases is readdressed. Fifth, it is shown that Kelley's covariation model does not specify all attributionally relevant information, and we describe an extension of the Kelley Model (the ANOVA Model) that includes this ‘missing’ information. It is then, sixth, tested whether attributional biases still occur when subjects are provided with complete information.