THE THEORY OF COGNITIVE FIT - Human-Computer Interaction and Management Information Systems: Foundations - page 128

for the accuracy of the outcome and/or the effort involved. More complex tasks, for example, may
vary substantially in both the amount and type of evaluation.
In general, more complex spatial tasks—forecasting, for example—involve perceptual evalua-
tion. Because perceptual processes require less effort than analytical processes, the decision
maker will have little incentive to use analytical evaluation processes to solve a spatial problem.
In general, more complex symbolic tasks involve analytical evaluation. The decision maker
may, at some point, be induced to use perceptual rather than analytical evaluation processes to
reduce the effort involved. Complex symbolic tasks may require a substantial amount of analyti-
cal processing; for example, they may require computations that are not difficult, but are tedious
and time-consuming, and therefore error-prone. They may therefore place significant strain on the
cognitive resources of the decision maker (Campbell, 1988). A complex symbolic task with even
higher analytical demands may be a limiting task, that is, one that cannot be solved analytically
without assistance of some kind. It may involve, for example, calculating trends from extremely
complex functions. Such a task has further constraints placed on it from the viewpoint of its suc-
cessful completion. Johnson and Payne (1985) speculate on the existence of this type of complex
decision-making task in discussing the process of expected utility maximization. They state:
“Such processes, however, could well require inordinate amounts of time, and, in practice, be
impossible for the unaided decision maker. Processing constraints, therefore, may impose severe
limitations for the feasible region in which accuracy-effort trade-offs could be made.”
The Theory of Cognitive Fit
The theory of cognitive fit applies cost-benefit theory to decision making using graphs and tables
in two ways. First, the simple form of the theory, which addresses information acquisition and
well-defined evaluation, is a special case of cost-benefit theory. Second, the more traditional view
of cost-benefit theory involving strategy shift applies to decision making on more complex tasks
where a number of appropriate strategies may be available. We first present the theory of cogni-
tive fit followed by its application to both fairly simple as well as cognitively complex tasks. We
also present the propositions that underlie the theory.
Fundamentals of Cognitive Fit
Figure 8.1 presents the model of decision making on which the theory of cognitive fit is based. When
the types of information emphasized in the decision-making elements (problem representation and
problem-solving task) match, the problem solver is able to use processes (and therefore formulate a
mental representation) that also emphasize the same type of information. Consequently, the processes
the problem solver uses to act both on the problem representation and on the task will match. The
resultant, consistent mental representation will facilitate the decision-making process. Hence, cogni-
tive fit leads to an effective (i.e., accurate or precise) and efficient (i.e., fast) problem solution.
Theoretical support for the relationships in the general problem-solving model comes from a vari-
ety of sources. The literature provides substantial support for problem solvers' use of processes that
match the problem representation (Bettman and Kakkar, 1977; Kotovsky, Hayes, and Simon, 1985;
Russo, 1977; Tversky and Kahneman, 1971, 1973, 1974). There is also substantial evidence that prob-
lem solvers use different processes for different types of tasks (Einhorn and Hogarth, 1981; Slovic and
Lichtenstein, 1983; Tversky, Sattath, and Slovic, 1988; Vessey and Weber, 1986). Finally, there is evi-
dence that matching the problem representation directly to the task has significant effects on problem-
solving performance (Bettman and Zins, 1979; Simkin and Hastie, 1987; Wright, 1995).