The use of time-series notation as an" objective approach to a refined analysis of nonverbal behavior in depression is described with behavioral samples taken from doctor-patient interviews. The movement patterns of 13 depressed patients were studied twice, first when the patients were diagnosed to be in a severely depressed state, and second, when they were judged to be nearly recovered. The movements of head, trunk, shoulders, upper arms, hands, upper legs, and feet were transcribed as a series of positions over time for the first 3 minutes of each of 26 doctorpatient interviews. Codings were obtained from video recordings, at half-second intervals, for 55 coding dimensions. Three parameters, the mobility, the complexity, and the dynamic activation of body movement, were defined and quantified on the basis of these data matrixes. The findings demonstrate that the timeseries notation of nonverbal interaction offers new possibilities in the quantitative study of behavior, especially in the assessment of behavioral features of potential importance in determining diagnostic subgroup and therapeutic response.The assessment of a patient's motor activity has always been considered an important part of the psychiatric examination. From the early textbooks in psychiatry (e.g., Griesinger, 1876;Kraepelin, 1913) to the most recent edition of the Diagnostic and Statistical Manual of Mental Disorders, DSM-III (American Psychiatric Association, 1980), it has been repeatedly emphasized that "objective psychomotor assessments may improve classification, longitudinal monitoring, treatment selection, and prediction of outcome for patients with depression" (Greden & Carroll, 1981, p. 1441. The recent rise of interest in nonverbal communication has given additional impetus to the study of psychomotor function. Mahl (1968) suggested that "some This joint endeavor of the present authors was supported by a research grant from the Fritz Hoffmann-La Roche Foundation for Interdisciplinary Research. Development of the methodology used in this research would not have been possible without the extensive support received from Swiss National Science Foundation Grants 1,913-0.73, 1.467-0,76, and 1.070-0.79.We want to thank Robert Crawford for his constant collaborative support and especially for his critical contributions to this article.Requests for reprints should be sent to
