How do our emotions express themselves at a physiological level?
To answer this question, we must look at the activity of the Sympathetic and Parasympathetic nervous systems.
In a very stressful situation, for example, our heart starts to beat faster, while our blood pressure increases. Moreover we sweat and our muscles stretch. In other words, the body prepares itself physiologically to fight a danger, and it is the Autonomous Nervous System (ANS) that sets out to give our body the extra energy. This phenomenon is made possible by the activation of a branch of the ANS, the Sympathetic Nervous System (SNS).
However, in a relaxed situation, the ANS allows our muscles to relax, our heart rate to decrease and generally our physiological activity to decrease in intensity. This happens when another branch of the ANS called the Parasympathetic Nervous System (PNS) becomes active. This activation corresponds to changes in the pattern of cardiac activation (heart rate variability) and in various indicators of the electrical and thermal properties of the skin.

Data collection, analysis and emotional classification

From four, non-invasive, on-board sensors, our range of connected bracelets collect physiological indicators of SNS and PNS activities to provide input data to the Emotion Sense™ technology.

Heart rate, and heart rate variability (HRV), is measured by a photoplethysmographic sensor (PPG). It is the slight variations in heart rate that tell us about the emotional experience.
The sweat gland activity is also analyzed from the electro-dermal activity sensor (GSR). Electro-dermal activity has been widely shown to be a reliable indicator of the emotional state.
The temperature sensor provides information on temperature changes at the skin’s surface. Numerous studies have shown that this temperature varies according to the emotion felt.
Finally, the accelerometer sensor measures motion and activity to improve the ability of the Emotion Sense™ technology to detect emotional states.
Signal processing algorithms make it possible to prepare and optimize the signals collected by the sensors (noise reduction, etc.) before extracting the variables allowing the classification of emotional states (machine learning).
Emotion Sense ™ analyses and combines these variables in real time to identify the emotional experience (stress, fear, joy, relaxation, boredom, etc.). This experience depends on the level of pleasure or displeasure, as well as the degree of physiological excitement.
For example, an emotion such as joy implies feelings of pleasure and a high level of excitement while boredom implies feelings of displeasure and a relatively low level of excitement. Thus, Emotion Sense ™ reliably understands whether the emotional state is more likely to be in a positive or negative zone, while determining the level of excitement. The status is then communicated to any system (PC, Tablet, Console, Smartphone…).

Scientific background

  • Ali, M., Al Machot, F., Mosa, A. H., Jdeed, M., Al Machot, E., & Kyamakya, K. (2018). A Globally Generalized Emotion Recognition System Involving Different Physiological Signals. Sensors (Basel, Switzerland),18(6).
  • Ali, M., Al Machot, F., Mosa, A. H., & Kyamakya, K. (2016). CNN Based Subject-Independent Driver Emotion Recognition System Involving Physiological Signals for ADAS. In Advanced Microsystems for Automotive Applications 2016(p. 125–138). Springer.
  • Jiang, M., Mieronkoski, R., Rahmani, A. M., Hagelberg, N., Salantera, S., & Liljeberg, P. (2017). Ultra-short-term analysis of heart rate variability for real-time acute pain monitoring with wearable electronics. In 2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) (p. 1025–1032). IEEE.
  • Picard, R. W., Vyzas, E., & Healey, J. (2001). Toward machine emotional intelligence: Analysis of affective physiological state. IEEE transactions on pattern analysis and machine intelligence, 23(10), 1175–1191.
  • Schumm, J., Bachlin, M., Setz, C., Arnrich, B., Roggen, D., & Troster, G. (2008). Effect of movements on the electrodermal response after a startle event. In Pervasive Computing Technologies for Healthcare, 2008. PervasiveHealth 2008. Second International Conference on(p. 315–318). IEEE.
  • Shi, H., Yang, L., Zhao, L., Su, Z., Mao, X., Zhang, L., & Liu, C. (2017). Differences of heart rate variability between happiness and sadness emotion states: a pilot study. Journal of Medical and Biological Engineering, 37(4), 527–539.
  • Shu, L., Xie, J., Yang, M., Li, Z., Li, Z., Liao, D., … Yang, X. (2018). A Review of Emotion Recognition Using Physiological Signals. Sensors (Basel, Switzerland), 18(7).
  • Valenza, G., Citi, L., Lanatá, A., Scilingo, E. P., & Barbieri, R. (2014). Revealing real-time emotional responses: a personalized assessment based on heartbeat dynamics. Scientific reports, 4, 4998.