Audio-Visual and Heart Signals for Attention and Emotion Analysis
Abstract
The paper proposes a theoretical architecture model to analyze human emotion and attention. This model is basedon the concept of dissonances defined as conflicts between individual, collective, or organizational knowledgeas attention or emotion. The detection or the control of such abnormalities can produce negative emotions asdiscomfort, overload, stress, or misunderstanding, but both detection and control can be disrupted by inattention.Two studies motivated the proposed architecture. The first one concerns the Reverse Comic Strip to analyze feelingfeedback regarding emotional facial pictures and by reproducing thoughts with words. The second one concernsthe impact of the synchronization of alarms with heartbeats on attention. The theoretical model consists then inidentifying negative emotions due to a dissonance occurrence by being inspired by the Reverse Comic Strip supportto treat audio-visual signals from a subject, and takes into account the synchronization of dynamic events with theirheart rate that can affect attention and their human perception abilities