Competition, Superstition and the Illusion of Control.(Statistical Data Included)

The effects of reinforcement schedule and competition on generating superstitious behaviors and beliefs were examined in 72 people. Superstition was induced by having participants respond to turn on a tone under a concurrent 2-lever Variable Interval (V1) Extinction (EXT) schedule. During the session, stimulus lights would occasionally be illuminated, although they did not signal any change in contingency. Attributing importance to the inactive lever, a pattern of switching between levers, or to the illumination of the lights were considered to be superstitious beliefs. Participants were either run alone or in pairs, and manipulation of the reinforcement schedule resulted in groups which were matched in probability of reinforcement, as well as in groups which were mismatched. Reinforcement schedule (VI 30" versus VI 60") and competitive situation did not affect degree of superstitious belief, except when people were placed in a "winning" condition. However, Superstition was associated with participants' belief in improved future performance and with participants' perceived skill relative to their opponents. Results are discussed in terms of relationships between superstition, the illusion of control, and self-efficacy. Differences between experimentally-induced and commonly held superstitions are also discussed.

In operant behavior, an instrumental response is strengthened when it is followed by a desired consequence. Occasionally, the desired outcomes occur independently of actual responding, yet people still attribute a causal relationship between the two. Such an attribution can be called a superstitious belief. The resultant behavior, superstitious responding, can be defined as those instances in which responses are maintained through accidental correlation with reinforcement.

In the first thorough investigation of this phenomenon, Skinner (1948) provided food reinforcement to pigeons every 15 seconds, completely independently of their behavior. After only a few minutes the pigeons developed distinctive stereotyped rituals; one walked around in circles, another rubbed his face on the floor, etc. Skinner explained these idiosyncratic behaviors as resulting from the adventitious reinforcement of responses that were temporally contiguous with the presentation of the grain. Superstitious rituals resulting from similar schedules has also been reported in children (Wagner and Morris, 1987) and adults (Ono, 1987).

Superstitions can also arise in response-dependent situations, and are often seen when reinforcement is dependent upon behavior with the accidental correlation of one response for subsequent reinforcements for other responses (Catania & Cutts, 1963). In Catania and Cutts' study, superstitious responding

in humans was examined using a two-button situation. Presses on one button were reinforced with points on a Variable Interval (VI) 30" schedule (i.e., on the average of every 30," a reinforcer was scheduled and produced by the subsequent response) while presses on the other button had no consequences. This second button can be thought of as presenting an extinction (EXT) contingency. Despite the fact that reinforcement was independent of any responding on the second button, responding on it was observed throughout sessions. Moreover, participants' responding often revealed stereotyped patterns in their responses across both buttons. The authors reported that this was likely due to presses on the inactive button frequently being followed by reinforced presses on the active button.

A qualitatively different type of superstition, termed a sensory superstition, involves giving special importance to some feature of our surroundings (Vyse, 1997). What appears to be a salient aspect of the environment may turn out to be irrelevant, but gain associative meaning due to its temporal contiguity with reinforcement. For example, Morse & Skinner (1957) rewarded pigeons on a very lean reinforcement schedule (i.e., VI 30'). At the start of the session, a blue light was illuminated. Occasionally, the blue light was changed to orange for 4-minute periods. Even though the reinforcement schedule was the same under the two colored lights, the pigeon treated them differently, often altering its rate of responding when the orange light was illuminated. Starr & Staddon (1982) extended these results using several different colored lights and different reinforcement schedules. Once again, the pigeons responded with different rates in the presence of different colored lights, even though the lights did not set the occasion for a difference in the contingency. Similar findings with human participants have also been reported. Ono (1987) had participants respond under Fixed--or Variable-Time schedules (unlike Interval schedules, reinforcement is not contingent upon a response in Time schedules and reinfocers are delivered when scheduled). During the session, three different signal-light colors were presented in a random sequence, but matched for total number of presentations and duration. Changing of the colors had no bearing on the reinforcement schedule. Ono reported that 30 percent of his participants significantly altered their responding in a systematic fashion under these different stimulus conditions.

While many studies have shown that superstitions can develop through adventitious reinforcement, fewer studies have examined various factors that may make it more or less likely that superstitious behavior or beliefs will develop. One variable that has received some attention is reinforcement schedule. For example, Vyse (1991) reported that superstitious behavior increased when response requirements changed from a Fixed Ratio 1 to a Random Ratio 2 reinforcement schedule. Ono (1987) showed that more superstitious behavior occurred when people received points under a fixed time 30" schedule than under a fixed time 60' schedule. Finally, Rudski et al. (1999) demonstrated that the likelihood of people in generating fallacious rules linking specific responses to random letter-digit combinations was positively related to probability of a desired outcome.