Hypotheses about the involvement of physiological and personality mechanisms in the relation between stress and headache have not yet been tested in a pretreatment-post-treatment treatment design though many behavioral techniques are based on these notions. Effects of four treatments (stress-coping, stress-coping including finger temperature biofeedback, stress-coping including temporal vasoconstriction biofeedback and waiting list-control) were analysed in migraine patients with regard to headache measures, psychophysiological functioning and personality variables. In tension headache patients three treatments were compared (stress-coping, stress-coping including frontalis EMG biofeedback, and waiting list-control). Changes neither in physiological nor in personality measures showed concurrency with changes in headache variables. Our results do not confirm the hypothesis of a mediating role between stress and headache for temporal artery and frontal muscle stress responses in migraine patients and tension headache patients, respectively. Neither was the notion of involvement of personality change in headache improvement sustained.