Abstract
It is proposed that the decision making process is intrinsically formulative in nature; i.e., for the individual the crux of the process is in the development of a frame, a knowledge base, that can be used to make the decision, rather than the evaluation of the alternatives ’per se’. A two stage model of decision making is forwarded. In the first stage the individual develops his frame. This is the critical, formulative, stage and dependent on the socio‐cultural environment. In the second stage the individual makes a decision by evaluating his frame. This stage is mechanical, evaluative, and determined by the frame formulated during the first stage. A consequence of the basic thesis is that the social and cognitive processes can not be decoupled if we are to understand decision making behavior. However, there are few methods currently available that allow the researcher to look at the relationship between the social and the cognitive process. Herein, a set of methods that permit the researcher to look at this relationship are described and then applied to a particular case study. On the social side these methods are based on a network representation of social structure. On the cognitive side these methods are based on a set of semantic models of human knowledge. A set of cognitive models for representing information are presented — these include frames, maps, and knowledge bases. A set of programs, Frame Technology, for coding, manipulating, and analyzing frames, based on these models, are briefly described. The relationship between social structure and individual cognitive formulations for a group of undergraduates at MIT formulating the problem of selecting a new tutor is then examined using these programs. This work is an overview of a new approach to studying decision making, one that tries to relate social structure to cognitive structure at an empirical level. It is not a detailed analysis of a particular decision: the point is not what decision was made, but the role that information played in this decision. Nor is a detailed discussion of the set of new methods provided: the point is not how the methods work, but the way in which they can be used.