Subject versus experimenter control in the reaction to pain

&NA; Although perceived control has been used as a means of reducing the reaction to pain, conceptually, much is still unclear. The purpose of this study was to clarify the effects of control as a mediating variable in the reaction to pain as a function of the predispositional variables of perceived self‐efficacy and trait anxiety. Self‐efficacy refers to the subject's premeasured, perceived ability to control his pain. The type of control provided either to the subject or the experimenter was varied over 5 independent groups. Predictability of occurrence of the pain stimulation was varied on a within‐subject basis. Results mirrored the complexity of the problem. Overall, when subjects were given decisional control alone the largest reactions to pain were obtained. Subjects given decisional plus behavioral control yielded the lowest reactions to pain. Both self‐efficacy and trait anxiety significantly were pedictive of outcomes. Unexpectedly, experimenter control reduced the reaction to pain among those with high self‐efficacy, but increased the reaction among those with low self‐efficacy. Results of trait anxiety appear to indicate the need to keep anxiety within bounds. Adding responsibility such as control to an individual who is already anxious may increase the reaction to pain. In general, it appears that control that is perceived as inadequate may be worse than not having any control. The theoretical and methodological implications of these results were discussed.