The effects of deliberation and cue-weight uniformity on judgment accuracy.

Abstract

The view that people should carefully consider their options forms the cornerstone of rational decision making. However, researchers have recently demonstrated that under certain conditions, this type of careful thinking may actually be detrimental. This dissertation extends that work by exploring when careful deliberation will and will not be useful. Our primary hypotheses are that: (1) when many items of information "should" affect one's judgment or decision, thinking carefully will be beneficial, and (2) when one item of information is of particular importance, deliberating about other items will be detrimental. The justification for these hypotheses is that deliberation calls attention to information that was not considered originally, and away from information that initially would have dominated the judgment or decision. This change can be either beneficial or detrimental, depending on the validity of the information to which attention is being called. We performed three experiments to test these hypotheses. In these experiments, subjects judged some criterion on the basis of a number of items of information, or cues. In Experiments 1 and 2, half the subjects were given a judgment aid where, prior to each judgment, they evaluated the utility of each cue for predicting the criterion. Additionally, half the subjects were given equally predictive cues, and half were given cues such that one was much more predictive of the criterion. In Experiment 3, a third condition was constructed where subjects evaluated the relationship between the predictors and the criterion prior to the beginning of the experiment. As predicted, the judgment aid decreased the accuracy of subjects presented with non-uniform cues in Experiment 1, and increased the accuracy of subjects presented with uniform cues in Experiment 2. There were no effects of the judgment aid in any of the remaining conditions. Additionally, the two aids used in Experiment 3 did not differentially affect accuracy. It is concluded that when deliberation does affect people's judgments or decisions, it does so in the hypothesized manner. However, deliberation sometimes has no effect at all, and this is conjectured to occur when people are unable to construct plausible scenarios for why cues should affect the criterion.