THE THEORY OF COGNITIVE FIT - Human-Computer Interaction and Management Information Systems: Foundations - page 151

An ER model with geospatio-temporal data represented as annotations on the traditional concep-
tual schema, in which fact and context are stored together, therefore matches the way the problem
solver structures the task in internal memory, while the representation of contextual geospatio-
temporal data separate from the associated fact (i.e., entities and relationships) does not. The
researchers' thesis that problem solvers using the annotated representation of geospatio-temporal
data on conceptual schema would outperform those using the separate representation of geospatio-
temporal data was supported in the experiment.
Evaluation of Cognitive Fit Based on Traditional Match/Mismatch
Using New Dimensions of Fit
With respect to the study by Hubona et al., on declarative and procedural dimensions of language-
conveyed spatial information, the theory is supported for survey tasks on the basis of time, and for
route tasks on the basis of accuracy, while the finding for speed was the opposite of that hypothe-
sized. What is apparent here is that: (1) we need additional information to be able to assess the sit-
uation in more detail; (2) we need to ensure that tasks are comparable in all ways; and (3) we need
to ensure that we use analyses that fully control for the possibility of accuracy/time trade-offs.
Beckman's study is quite a traditional study of the theory of cognitive fit in a significantly dif-
ferent domain, that is, tasks involving physical control of ground vehicles. The meaningful results
from this study fully support the theory of cognitive fit. The study by Khatri et al. (2006) is also
a traditional study of cognitive fit that recognizes the fact that the internal representation of the
task itself can be matched to the problem representation. The results for accuracy in this experi-
ment support the theory of cognitive fit.
Hence the analyses of the studies presented here suggest that the theory of cognitive fit can be
extended to other dimensions of fit.
COGNITIVE FIT BASED ON THE EXTENT OF FIT BETWEEN
PROBLEM-SOLVING TASK AND PROBLEM REPRESENTATION
We identified two studies in which the extent of fit between the problem-solving task and prob-
lem representation varied. Table 8.8 presents details of these studies. We first present the studies
and then evaluate the findings.
Studies of Cognitive Fit Based on the Extent of Fit Between Problem-Solving
Task and Problem Representation
The first study is by Borthick et al. (2001) who investigated “the effects of information request
ambiguity and construct incongruence on query development.” Although the researchers con-
ducted a single experiment, their design of the ambiguity and construct congruence treatments and
the results of the statistical analyses conducted can be conceived as testing what may be regarded
as two cognitive fit models. What characterizes both of these studies is the notion of close and far
matches between pairs of variables in the cognitive fit model. We do not further consider the first
part of the study because it involves matching the problem-solving task to the problem-solving
tool, which is outside the scope of this analysis.
The second set of results presented examines the effect of high and low congruence between
problem-solving tasks (information requests) and the problem representation (in this case, an