Lack of correlation between frontalis electromyography and self-ratings of either frontalis tension or state anxiety

Since training in deep muscular relaxation is common to most behavioral treatments for anxiety and tension-related disorders, muscle tension measures, such as electromyography (EMG), would provide valuable information as to the efficacy of these treatments. In the past, however, the majority of clinical reports on behavioral treatments employing muscular relaxation have based results primarily on self-rated outcome measures, despite evidence in the literature of a lack of correspondence between self-ratings and physiological measures of muscle tension. An experiment was performed on 20 normal subjects comparing self-ratings of anxiety and frontalis muscle tension with frontalis EMG before and after a 15-min. relaxation period accompanied by frontalis EMG biofeedback. Analysis showed that the self-ratings of anxiety and frontalis tension correlated significantly both before and after the relaxation period. However, no correlation was observed between self-rated anxiety or frontalis tension and frontalis EMG either before or after the relaxation period. The relaxation period had no significant effect on the dependent measures. It is suggested that, if people are unable to report accurately on their somatic states, then physiological outcome measures, such as EMG, would be preferable to self-rated measures in assessments of relaxation training procedures.