Selection Versus Rejection: The Role of Task Framing in Decision Making

Jing Chen, Purdue University

Abstract

Procedure invariance is a basic assumption of rational theories of choice, however, it has been shown to be violated: Different response modes, or task frames, sometimes reveal opposite preferences. This study focused on selection and rejection task frames, involving a unique type of problem with enriched and impoverished options, which has led to conflicting findings and theoretical explanations. On the one hand, greater preference has been found for the enriched option in the selection task than in the rejection task; this result is explained by a compatibility account, in which the positive features of the enriched option are more compatible with the selection task and the negative features with the rejection task (Shafir, 1993). On the other hand, it has been found that this preference difference in the two tasks interacts with the relative attractiveness of the two options: The enriched option is preferred more (less) often in the selection task than in the rejection task when it is more (less) attractiveness than the impoverished option; this finding is attributed to the accentuation of difference between options in the selection task, as stated in the accentuation account (Wedell, 1997). My dissertation focused on examining the role of task frame in human decision making by distinguishing the compatibility and accentuation accounts, using an information-processing approach. Experiment 1 was conducted online on introductory psychology class students, with a plain statement for the task (either a selection or a rejection task). A large difference between the two task frames (i.e., the task framing effect) was found as predicted by the accentuation account. In Experiment 2, participants were recruited from the same subject pool but were required to verbalize their thoughts while performing the same tasks in a laboratory. No difference between the two task frames in the choice data was found in this experiment, possibly due to the need for verbalization of reasons in Experiment 2 or participants’ confusion about the rejection task in Experiment 1. With a modified version of the questionnaire conducted on both MTurk workers (Experiment 3A) and introductory psychology students (Experiment 3B), Experiment 3 emphasized the tasks in several different ways to reduce the possible confusion regarding the task, and a similar pattern as in Experiment 1 was evident though with a smaller effect size. Thus, it was established that task confusion cannot explain the task framing effect alone. Experiment 4 used a judgment task, in which participants were required to rate the likelihood of selecting or rejecting an option. It was again found that more participants in the negative task did not understand the task correctly before any feedback was provided. The ratings from this experiment were used as direct attractiveness measures, and a similar task framing effect was found with these measures. The finding of task framing effect was supported by the data from an eye-tracking experiment (Experiment 5), in which participants performed the tasks in the lab without being required to verbalize their thoughts. In the last two experiments, whether the task framing effect was influenced by time pressure was tested. Experiment 6 imposed time limits on participants and required them to respond within a short time, whereas Experiment 7 forced participants to wait a certain amount of time before they could respond. Both experiments found a task framing effect that did not differ from that in Experiment 3A, which indicates that this task framing effect was relatively automatic and that it did not take extra time for people to be more discriminating in the selection task than in the rejection task. The current results are not consistent with the compatibility account, which predicts the enriched option always to be preferred more in the selection task. Instead, they are more consistent with the accentuation account, which predicts that the difference between the two task frames would interact with the relative attractiveness of the two options, with people being more discriminating under the selection task frame. Based on the current findings, a modified version of the accentuation account, explaining the difference between the two task frames in terms of availability of cognitive resources, was proposed. The modified accentuation account suggests that people are less discriminating in the rejection task because understanding the task per se is more effortful and occupies more cognitive resources.

Degree

Ph.D.

Advisors

Proctor, Purdue University.

Subject Area

Behavioral psychology|Cognitive psychology

COinS