Breathing Emotion: Assessing the Respiratory Cues in HCI Emotional Dynamics

Abstract

This study looks into the complex dynamics of emotional and physiological responses evoked as individuals engage with various web-based tasks. The emphasis is on understanding how the emotions elicited by one task can influence both self-evaluated emotion and breathing patterns during the following task. The aim of the research is to show the intricate ways task-induced emotions, particularly on the valence and arousal dimensions, affect subsequent user experiences within a digital environment. 19 participants interacted with a controlled environment designed as a clothing store and completed a series of tasks designed to evoke different levels of emotions. Throughout these tasks, breathing patterns were measured using a respiration belt alongside participants’ self-reporting of their emotional states at the end of the study. The findings show evidence that emotions carry over between tasks, particularly in tasks designed to have a greater emotional impact and those that are temporally close to each other. Breathing patterns varied across tasks and showed correlations with self-assessed emotions suggesting a link between the two. The research emphasizes the importance of considering both physiological data and self-reported emotions when understanding web-based interactions. Self-reported arousal showed predictive links across tasks while valence did not. Self-reported valence was however better explained by the breathing patterns over the course of the experiment. This exploratory study offers the field of HCI insights into emotional carry-over in web-based tasks and how breathing patterns can be used as markers of emotion. It calls for further research to deepen the understanding of the interplay between emotional responses, task characteristics, and respiratory markers in a digital environment.