Selecting optimal wearables for measuring physiological arousal in robot-delivered mindfulness-based exercises

Area of Science:

• Robotics
• Psychophysiology
• Human-Computer Interaction

Background:

• Social robots offer therapeutic potential, but accurate affective outcome assessment is crucial.
• Selecting appropriate devices for recording autonomic responses during interventions is essential for data quality and participant comfort.

Purpose of the Study:

• To compare the sensitivity and social validity of two electrodermal activity (EDA) devices: Shimmer and Empatica E4.
• To select the most suitable device for a larger study involving robot-delivered mindfulness-based relaxation exercises.

Main Methods:

• Assessed EDA sensitivity to stress-inducing and relaxation exercises using a Shimmer device (sock) and an Empatica E4 (wrist-worn).
• Evaluated device comfort and agreement using Bland-Altman analysis in 28 healthy adult volunteers.
• Corroborated physiological data with self-report measures.

Main Results:

• Both devices detected EDA variations corresponding to stress (increase) and relaxation (decrease) induced by robot-delivered exercises.
• The Shimmer device demonstrated higher sensitivity to EDA changes compared to the Empatica E4.
• No significant differences in wearing comfort were found between the devices, but Bland-Altman analysis indicated low agreement.

Conclusions:

• While both devices show promise for measuring psychophysiological responses in robot-assisted interventions, the Shimmer device appears more sensitive.
• Device selection should consider the specific demands of the intervention and the need for accurate EDA measurement.
• Further research can refine the use of wearable sensors in therapeutic social robot applications.