THE THEORY OF COGNITIVE FIT - Human-Computer Interaction and Management Information Systems: Foundations - page 124

INTRODUCTION
The theory of cognitive fit was introduced in 1991 to explain the inconsistent results of numerous
studies in the area of information presentation using graphs and tables (Vessey, 1991). Although
Vessey referred to cognitive fit as a “paradigm,” the research community has consistently accorded it
the status of a theory, referring to it since its first appearance as “the theory of cognitive fit” or “cog-
nitive fit theory.” In keeping with these subsequent studies, we refer to cognitive fit as the “theory of
cognitive fit.”
Vessey introduced cognitive fit in 1991 as one element of a general theory of problem solving.
This statement was intended to convey that the notions of cognitive fit applied across many
domains and could be applied to many dimensions of fit. The notion of a general theory of prob-
lem solving was conceived early in the development process: Vessey developed the theory after
observing similar problem-solving situations in the systems development domain (see Vessey and
Weber, 1986). It was therefore clear when the theory first appeared that it had sufficient explana-
tory power to be applicable across domains.
The focus in the original paper on cognitive fit, in which Vessey analyzed prior studies as a basis
for formulating the basic theory, was on the simple information acquisition and information evalua-
tion tasks that almost exclusively made up the information presentation studies published at that
time. Then, in 1994, Vessey published the extension to the basic theory that addressed problem solv-
ing in more complex tasks. That paper used cognitive cost-benefit theory as a unifying theory to
explain cognitive fit in both simple and more complex tasks. The basic theory of cognitive fit is a
special case of cost-benefit theory. Cost-benefit theory as applied to more complex tasks explains
contingent decision-making behavior in which decision makers change strategy to accommodate
even minor variations in the task or the task environment; for example, they trade off a slight loss in
accuracy for a significant reduction in effort. Numerous factors influence the choice of strategy,
among them the way in which information is presented.
Vessey and Galletta (1991) further extended the theory to include matching specific skills to
problem representation and task, while Sinha and Vessey (1992) extended it to include matching
a problem-solving tool to a problem representation and task. These studies revealed that, although
matches of this type affect problem-solving performance, the key match remains that of matching
a problem representation to a task.
Since their appearance in print, these papers on cognitive fit have been cited in numerous pub-
lications. As of April 28, 2004, for example, the original paper (Vessey, 1991) had been cited 102
times, as recorded by the ISI's Web of Science, and the empirical study based on it (Vessey and
Galletta, 1991) had been cited sixty-eight times. Further, the extension of the theory to more com-
plex tasks (Vessey, 1994) had been cited fourteen times, while the extension of CF to problem-
solving tool (Sinha and Vessey, 1992) had been cited eleven times.
Our objective in this paper is to revisit the theory of cognitive fit at this point in time. We aim
to determine whether the theory has stood the test of time, and, if so, examine the ways in which
the theory has developed over time, that is, how it has been used and extended. At the same time,
we attempt to trace a path for further evolution of the theory. For the purposes of this paper, we
examine the core concepts of the theory, that is, the match between task and problem representa-
tion, in those articles that have examined cognitive fit in the domain of decision making. We do
not examine, therefore, research on cognitive fit that appears in the software arena.
In the next section, we first present cost-benefit theory, which forms the overall theoretical
framework for the theory of cognitive fit, the types of tasks to which it applies and the types of
representations that support them, followed by the theory itself. It seems particularly important to