The effects of a computer program (Health Imagery Technology Systems, HITS) designed to promote attitude and cognitive changes through elicitation of evoked response potentials were evaluated with chronic pain patients. A treatment and control group were used for comparison (52 patients, 22 females, 32 males, mean ages 47). Wechsler Adult Intelligence Scale-Revised subtests, a Semantic Differential scale, the Health Attribution Test, an imagery protocol, the McCoy-Lawlis Pain Drawing, and the Zung Depression scale were used at admission and discharge to measure change. A pre- post-mood thermometer was used with the treatment group. The hypotheses that the treatment group would show significant changes on these measures were tested with a two group repeated measures analysis of variance design. No significant changes were noted for either group on the intellectual measures, on health attitudes, or reports of pain. The similarities subscale showed significant within group variance (F = 5.46, p < .023). One bipolar adjective pair indicated significant differences (F = 4.79, p < .035), possibly a result of chance. One of seven imagery measures suggested a significant improvement in strength of imagery for the treatment group (F = 18.2, p < .00008). Both groups showed significantly improved imagery of body defenses …