Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)




This study assessed the impact of selected task, instructional and personological variables as they contribute to relaxation during frontalis EMG biofeedback. Subjects (_n = 60) divided equally by sex and high versus low absorption were randomly assigned to one of five groups. Four groups (n = 48) comprised a Feedback x Instruction factorial and were provided either contingent or noncontingent feedback information and pre-training instructions that either emphasized a passive attention approach to acquired self-relaxation or were nonspecific. Care was taken to avoid task instructions for noncontingent feedback subjects which would be deceiving or countertherapeutic. A fifth no-treatment (resting) control group was also included to assess the affects of adaptation on frontalis EMG. Relaxation was operationally defined by changes in frontalis EMG as observed across three training sessions.

The results indicated that subjects receiving contingent EMG biofeedback were significantly lower in frontalis muscle tension across sessions. However, subjects in general did not demonstrate a significant negative linear trend in EMG during training and treatment conditions did not differ significantly from no-treatment controls in rate of within-session decline of EMG. Pre-training instructions, gender and absorption main effects were nonsignificant but a significant Instruction x Sex x Absorption interaction was obtained. Relaxation performance during Session 1 was found to significantly predict EMG levels for Session 3. The findings were interpreted to provide only partial support for the efficacy of EMG biofeedback. In addition, learned control of relaxation was viewed as a complex process involving interactions among instructional, task and individual difference factors.