THE THEORY OF COGNITIVE FIT - Human-Computer Interaction and Management Information Systems: Foundations - page 147

to more complex tasks as presented in Vessey (1994). 1 The types of tasks investigated are not sim-
ple information acquisition and evaluation tasks and the simple version of cognitive fit does not
apply in these circumstances. The symbolic containment task will be subject to time/accuracy
trade-off as the complexity of the task increases. In the intermediate stages of complexity (i.e.,
prior to the level at which decision makers can no longer address the problem using analytical
processes), accuracy may still be better with symbolic representations and speed will be faster
with maps, as was found in this study.
Evaluation of Cognitive Fit in Multi-Criteria Decision-Making Tasks
Unlike the multi-criteria decision-making systems used in Speier and Morris (2003), GIS facili-
tate a broader range of possible tasks, some of which may be spatial in nature, while others may
be symbolic. The two studies presented here address both spatial (proximity and adjacency) tasks
and symbolic (containment) tasks. Proximity and adjacency tasks are best supported with maps
rather than with tables of data (Smelcer and Carmel, 1997; Dennis and Carte, 1998). Smelcer and
Carmel found the effect in speed of decision making, while Dennis and Carte found it by analyz-
ing accuracy and time simultaneously. For Smelcer and Carmel's containment task, it appears
likely that the spatial and symbolic problem representations were not sufficiently difficult to result
in time differences, their measure of choice.
All of these findings support the theory of cognitive fit.
EXTENDING THE THEORY OF COGNITIVE FIT TO NEW DIMENSIONS
Certain studies of cognitive fit address dimensions of fit that are qualitatively different from those
associated with spatial and symbolic processing. We characterize the first set of studies as using new
dimensions in the context of the traditional match-mismatch form of fit and the second set of stud-
ies in terms of the extent of match between the problem-solving task and the problem representation.
Cognitive Fit Based on Traditional Match/Mismatch Using New Dimensions of Fit
We identified three studies that addressed the match of problem representation to task using new
dimensions of fit, each of which is unique in its approach. One of the studies introduces a new dimen-
sion into an already-established domain of investigation; the second study is based in a new domain;
and the third study establishes a new dimension of the task in the context of cognitive fit. Table 8.7
presents details of these studies. We first present the studies and then evaluate the findings.
Studies of Cognitive Fit Based on Traditional Match/Mismatch Using New Dimensions of Fit
The first study that extends the basic theory of cognitive fit to new dimensions is that of Hubona,
Everett, March, and Wauchope (1998) . The domain of investigation was that of “language-conveyed
representations of spatial information.” Their study used: (1) two types of spatial problem represen-
tations known as “survey” and “route” presented using natural language; and (2) two types of tasks
that also emphasized the same types of information. The survey representation described the layout
of a town as if one were picking out features on a map, while the route representation described the
layout in terms of what one would observe on driving through it. Survey and route representations
therefore emphasized declarative and procedural knowledge, respectively (Anderson, 1982, 1996).
Hence, these dimensions represented a more in-depth differentiation of spatial representations than