The present study investigates the effectiveness of Erikson hypnosis and Jacobson relaxation for the reduction of osteoarthritis pain. Participants reporting pain from hip or knee osteoarthritis were randomly assigned to one of the following conditions: (a) hypnosis (i.e. standardized eight-session hypnosis treatment); (b) relaxation (i.e. standardized eight sessions of Jacobson's relaxation treatment); (c) control (i.e. waiting list). Overall, results show that the two experimental groups had a lower level of subjective pain than the control group and that the level of subjective pain decreased with time. An interaction effect between group treatment and time measurement was also observed in which beneficial effects of treatment appeared more rapidly for the hypnosis group. Results also show that hypnosis and relaxation are effective in reducing the amount of analgesic medication taken by participants. Finally, the present results suggest that individual differences in imagery moderate the effect of the psychological treatment at the 6 month follow-up but not at previous times of measurement (i.e. after 4 weeks of treatment, after 8 weeks of treatment and at the 3 month follow-up). The results are interpreted in terms of psychological processes underlying hypnosis, and their implications for the psychological treatment of pain are discussed.