Observational and Accelerometer Analysis of Head Movement Patterns in Psychotherapeutic Dialogue

Abstract

Psychotherapists, who use their communicative skills to assist people, review their dialogue practices and improve their skills from their experiences. However, technology has not been fully exploited for this purpose. In this study, we analyze the use of head movements during actual psychotherapeutic dialogues between two participants—therapist and client—using video recordings and head-mounted accelerometers. Accelerometers have been utilized in the mental health domain but not for analyzing mental health related communications. We examined the relationship between the state of the interaction and temporally varying head nod and movement patterns in psychological counseling sessions. Head nods were manually annotated and the head movements were measured using accelerometers. Head nod counts were analyzed based on annotations taken from video data. We conducted cross-correlation analysis of the head movements of the two participants using the accelerometer data. The results of two case studies suggest that upward and downward head nod count patterns may reflect stage transitions in counseling dialogues and that peaks of head movement synchrony may be related to emphasis in the interaction.

A randomised controlled test of emotional attributes of a virtual coach within a virtual reality (VR) mental health treatment

“…However, when warm facial expressions were combined with affirmative head nods, there was a trend towards higher credibility ratings. This result might be due to the head nods giving the impression that the therapist was attentively listening and acknowledging participant responses 36 . Such an impression could have then enhanced the potential positive effects of warm facial expressions on treatment credibility when they were displayed simultaneously.…”

Using Inertial Sensors to Determine Head Motion—A Review

“…However, this range frequency was rarely used in comparison with 48-100 Hz. Existing studies utilizing this acquisition range are mostly focused on medical fields, such as the study of palsy as a human disorder or to improve the users' physical and mental state [35,38,56,67].…”

“…Regarding the age of the volunteers participating in the experiments, there was a wide variation: from 20 years [56] to 68 years old [34]. Most of the values of data acquisition frequency were as follows: 4 Hz [53], 10 Hz [28], 20 Hz [40], 48 Hz [20,36,54,71], 100 Hz [6,36,41,50,55], 125 Hz [38], 128 Hz [35], 200 Hz [47,74], 3.2 kHz [67], and 48 kHz [70]. The maximum sampling rate was found in a paper which proposed an inertial device which enabled psychotherapists to analyze mental-health-related communications [70].…”