Illusory Personal Control as a Determinant of Bet Size and Type in Casino Craps Games1

Langer's (1974, 1975) theory regarding the conditions under which subjects performing a chance task will suffer from an illusion of control over the outcome has specified 6 conditions proposed to enhance the illusion ofcontrol in chance tasks. A number of studies have applied her theory and predictions to gambling, a real-world arena of chance tasks where participants have been observed to act as if the outcomes are controllable by the use of various “logical” (acting on the “gambler's fallacy”) and superstitious strategies (e. g., blowing on dice). Thus far, however, these studies have not been conducted in naturalistic gaming environments and have yielded mixed results regarding the operation of the illusion of control. The present research offers the first study of the operation of the illusion of control in such a real-world context. In order to examine the effects of active vs. passive task participation (a variable hypothesized by Langer to affect the illusion of control), patrons of Reno casinos were observed placing craps bets on their own and another yoked patron's dice rolls. It was hypothesized that subjects would (a) place higher bets and (b) place more “difficult” bets (e. g., where only one specific number, as opposed to any of several numbers, may win) on their own rolls (when they would experience the illusion of control over the outcome) than on other patrons' roles (when they would not experience such an illusion). That is, players were expected to generally adopt riskier betting strategies when throwing the dice. Results supported the hypotheses.