tiveness restructuring was more prevalent than importance restructuring during the predecision period. Other studies in other contexts have illustrated predecision differentiation in terms of early rejection of several alternatives. This has been followed by selection of one preliminary choice alternative and processes, which can also be interpreted as reflecting differentiation (e.g., Bettman, Johnson, & Payne, 1991; Montgomery & Svenson , 1989; Russo et al., 1996; Svenson, 1974, 1979).
The preceding presentation has assumed that decision alternatives are represented in alternatives times attribute matrices. However, there are other competing ways of representing decision alternatives. One of these is a scenario-type script version ( Beach, 1990). This means that participants use their imagination to run scenarios into the future representing different alternatives. This way of representing alternatives does not conflict with an alternative times attribute representation, but it may provide a better foundation for a model of decision processes in some cases.
The alternative times attribute representation seems to imply that, in principle, decision makers use each attribute for the evaluation of at least two alternatives. However, it is also possible to assume that some attributes are used for only one alternative and others for another alternative forcing compensatory thinking across attributes. To exemplify, a decision process between the alternatives of a divorce and of continuing a marriage may use the attributes of security and habit supporting the no-divorce alternative weighed against the aspects on the attributes of variety in life and pleasure supporting the divorce alternative. Again, this can be modeled straightforward in an alternative times attribute representation, but perhaps another representation could be intuitively more appealing for modeling the psychological processes taking place. It would be interesting to explore how Diff Con could be adopted to alternative representations of decision alternatives such as those exemplified above. However, as shown in this chapter, there is still a need for a great deal of research within the existing Diff Con representation of decision problems, such as explorations of decision rules in process differentiation and consolidation, using the alternative times attribute representation relied on in the present chapter.
This study was supported by grants from the Swedish Council for Research in the Humanities and Social Sciences and Royal Netherlands Academy of Arts and Sciences through a NIAS scholarship.