Fourteen musicians who reported a history of pain in the upper limb associated with the playing of their instruments were compared with a sample of pain-free musicians, matched for age, sex and musical instrument. Four tasks were presented in random order and included neutral, general stressor, personal stressor and pain stressor tasks. Ratings of stressfulness and recordings of skin conductance level confirmed the effectiveness of the experimental manipulations for both subject groups. No differences were found between groups or tasks for frontalis surface electromyograph (EMG) activity. Evidence was found, however, of EMG elevation in flexor and trapezius muscles on the pain side for the pain subjects, in response to the task involving recall of a pain experience. This elevation was not found for the pain-free controls or for other stressor tasks, although some elevation in response to the pain stressor task was found for pain subjects in the trapezius muscles of the non-pain side. The duration of return to baseline of EMG following the pain stressor task was found to be extended in pain subjects for the trapezius, but not for the flexor muscles of the pain side. The findings suggest that site-specific muscle hyper-reactivity may play a role in the development and maintenance of occupational upper limb pain in musicians.