Abstract

The purpose of this study was to test the relative validity of the dominance structure search model in comparison with the two-stage models which had been proposed to explain decision process.. The search for a dominance structure is assumed to go through four phases, viz. pre-editing, finding a promising alternative, dominance testing, and dominance structuring. Experiment 1 tested whether or not the number of attributes searched in the initial stage of decision making were determined by the limited capacity of information processing. The results indicate that the number of attribute information searched in the initial stage of decision making is determined within the limited capacity. Experiment 2 was conducted to test the assumption of the dominance search model, that a certain promising alternative is selected tentatively after the pre-editing stage and this alternative has high likelihood to be selected later finally. The results showed that regardless of the number of dominant alternatives, the selection rate of the promising alternative was about 75% of all trials and of the alternative was selected finally 85% of the times even in competitive situation. In Experiment 3, the evidences for the dominance testing process and the dominance structuring process were investigated. The results revealed that both important and unimportant attributes were searched before the final choice and all conflicting attributes became the objects of conflict-resolution operation, regardless of the attribute importance. Experiment 4 tested whether or not the given information (e.g., attribute importance and attractiveness of attribute values) was restructured in the direction that a selected alternative might be dominant than competitive ones by means of the de-emphasizing and the bolstering operation in the case of interalternatives conflict. The results showed that the importance of dominant attributes in the selected alternative was bolstered more strongly in a post-decision than in a pre-decision, however, the importance of the dominant attributes in the competitive ones was de-emphasized much more low. The result on the attractiveness evaluation of attribute values showed that the evauation of attractiveness of attribute values in the competitive ones was not changed in the decision process, however, the attractiveness evaluation of attribute values in the selected one was bolstered more strongly. In summary, the results suggest that the dominance search model can explain the decision process very well in comparison with the two-stages models, a decision is made generally via the stages assumed in the dominance search model.