Applying a New Methodological Approach to a Problem of Information Processing Model Validation

摘要

The paper introduces the notion of the characteristic function of a binary decision graph as a tool for analyzing the validation of complex information processing models. The technique sheds light on the specific validation question how one can discern which of a set of process models with similar output is more relevant with regard to the particular set of decision subprocesses it contains. The approach is illustrated by applying the characteristic function to information processing models of consumer decision processes. After a brief description of these models, the characteristic function is defined and the notion of a conditional decision net is developed. The notion is then utilized to analyze the consumer models. Some implications about the form and character of decision processes are also drawn. (Author)