Preference reversal: A new look at an old problem

A generalized weak dominance approach is used to test the documented preference reversal (PR) phenomenon. This approach simply models risky choice behavior in PR as a choice between the best possible outcomes or a choice between the worst possible outcomes by equating smaller paired outcome difference between bets. The preference reversals are therefore seen as a consequence of the fact that gamble parameters are designed to encourage individuals to differentiate the difference between the worst possible outcomes of the two bets (i.e., to avoid the worse possible outcome of $ bet) rather than to differentiate the difference between the best possible outcomes of the two bets (i.e., to seek the better outcome of $ bet on which people tend to put a higher price). A "matching" task as well as a "pricing" task was designed to examine whether the knowledge of the value difference of the paired possible outcomes will permit prediction of preferential choice. The overall test results favor the equate-to-differentiate explanation. The present data suggest that the anomaly may be not in individuals' inconsistent preferences but rather in our inadequate knowledge of what it is that is being preferred when a question about preference is posed.