Biased evaluations of randomly determined gambling outcomes

We investigated whether people are biased in their evaluations of gambling outcomes even when the outcomes are chance determined. In Experiment 1, subjects exhibited a bias in their evaluations of the outcomes of a computerized bingo game. Manipulating whether the outcome of the first round was determined by a series of anomalous or “fluke” events had a greater impact on the subsequent bets of those who had lost their bets than on those who had won. Losers appeared to use these fluke events to explain away the outcome, whereas winners discounted their significance. Experiment 2 tested the limiting conditions of these findings by manipulating whether or not subjects were induced to perceive an “illusion of control” over the outcomes by allowing some of them to make a series of incidental choices during the contest. Only those subjects in the illusion of control conditions exhibited a bias in their evaluations of the outcome of the first round. The connection between this research and earlier behavioral theories of the psychology of gambling is discussed.