Psychophysiological parameters for emotion recognition – Conception and first evaluation of a measurement environment

Abstract

Emotions are of great importance for human interaction. The recognition of emotions is an essential basis for the social interaction of people, relevant at all levels of human interaction. However, the ability varies greatly among people, even to the point of a pathological deficit in autism (e.g., symptoms of "social blindness"). Furthermore, it is also an interesting topic in the current COVID-19 pandemic-situation with its restrictions regarding personal contacts and the resulting increase of digital contacts through online communication. For this reason, the measurement of emotions is of great interest for a wide variety of applications and, in particular, for product design or the development of trainings (e.g., for children with autism, leaders and managers, employees working in virtual teams etc.). The presented paper focuses on a concept and first evaluation of a measurement environment. Based on the objective emotional assessment (OEA) of Boucsein and his colleagues, a multidimensional approach was chosen for the developed concept. The studied psychophysiological measures were: cardiovascular (ECG) and electrodermal activity (EDA) as well as the forearm and neck electromyogram (EMG). In addition, the facial expression was recorded via camera. The results of a first evaluation (different tasks like a short game, a backward counting task, or thinking of different emotional contents) show, that the combination of the different measures was successful and promising for the use in further research. The study was able to show a need for improvement with regard to the derivation of the ECG (chest strap vs. adhesive electrodes). The design of the measuring environment can be seen as successful regarding the combination of the different psychophysiological measures and devices. In the future, an expansion with additional parameters such as pulse volume amplitude (PVA) or eye tracking will be considered.

Keywords: Emotion Recognition, Psychophysiological Measures, Measurement Environment

DOI: 10.54941/ahfe100958