Abstract The results of this study show that decision makers (DM) do not simply react to the dimensions presented by the experimenter only, but also Introduce new information from memory when constructing a cognitive representation of a decision task. In Experiments 1 and 2 subjects chose among candidates for a manager-post, who were described on two or three dimensions. On the common dimension information was available for both candidates, on the single dimensions for only one of them. The results showed that if a relevant aspect was missing and the DM has a knowledge base to predict the missing value, the missing value is substituted. If no prediction is possible, most DMs base their choice on considerations whether the missing aspect on the single dimension D s is better or worse than the available aspect on this dimension. In Experiment 3 subject were confronted with a familiar decision task (student jobs) and a new one (research projects). As may be predicted DMs produced more new information from memory in the familiar decision task than in the new one. In the new task the less information available the more information the DMs introduced from memory. Consequences of these results for methodology and theory are finally discussed in the paper.
[1]
J. Fleiss.
Measuring nominal scale agreement among many raters.
,
1971
.
[2]
Daniel G Bobrow,et al.
On data-limited and resource-limited processes
,
1975,
Cognitive Psychology.
[3]
O. Huber.
Nontransitive multidimensional preferences: Theoretical analysis of a model
,
1979
.
[4]
O. Svenson.
Process descriptions of decision making.
,
1979
.
[5]
Lars-Göran Nilsson,et al.
Perspectives on memory research : essays in honor of Uppsala University's 500th anniversary
,
1979
.
[6]
Paul Slovic,et al.
Dimensional Commensurability and Cue Utilization in Comparative Judgment.
,
1974
.