Assessed the utility of alpha biofeedback training in the treatment of psychiatric patients (N = 66). Biofeedback and placebo biofeedback groups were given alpha or mock-alpha training sessions, and their improvement on 54 variables was compared to that of no-treatment controls. Only a chance number of significant differential changes appeared among the three groups. A comparison of these negative results with an earlier study from our laboratory that had yielded positive findings suggested that clinical improvement in alpha biofeedback treatment may be the result of the relaxation training or special verbal instructions often incorporated into alpha training.