"Sensor data from over 100 participants recorded with the emteqPRO Pico 3 VR headset during empathy-evoking VR experiences"

"The aim of this dataset was to investigate how participants\u2019 empathy is related to changes in their physiological attributes measured by different sensors (inertial measurement unit, photoplethysmogram, electromyography (EMG), and a 3-axis accelerometer). Using a VR headset, 105 participants were immersed in three-dimensional 360\u00b0 videos of actors expressing different emotions (sadness, happiness, anger, and anxiousness). Participants reported their experience of empathy through short questionnaires. Based on their sensor and questionnaire data, different predictive models could be developed to predict a person's empathy score or other affective states using physiological arousal measurements obtained while watching VR videos.For recording participants' sensor data and performing calibration, we relied on the Emteq SDK, while Emteq DAB Tools served as a tool for accessing and reviewing results without the need to send them to the Emteq Emotion AI Engine. OpenFace was used for real-time data observation, data retrieval, and transmission to the Emteq Emotion AI Engine for further processing. Recorded variables included session time, EMG contact impedance, raw and filtered EMG signals, EMG amplitude (root-mean-square) for sensors 0\u20136, raw PPG signals, average heart rate (beats per minute), heart rate variability (HRV) metrics, and IMU measurements along each axis, resulting in a total of 57 features. Following further processing, the Emotion AI Engine applied fusion and ML algorithms to derive 27 additional variables per session (participant), referred to as affective insights. These were organized into nine categories: seven HRV and three respiratory features from PPG; two facial expression features from EMG; four arousal and four valence features from combined EMG and PPG data; one facial activation and one facial valence feature from EMG; one head motion feature from IMU; and four normalized muscle activation metrics from EMG electrodes on the zygomaticus, corrugator, frontalis, and orbicularis muscles."