A Bayesian Network Analysis of Exercise Experiences With Audio‐Visual Stimuli

Abstract

Audio-visual stimuli are widely used to enhance the exercise experience, yet the technological devices used to display these stimuli have received limited research attention. The aim of the present investigation was to apply a Bayesian Network (BN) to formally model and quantify the relationships between technological features of exercise-related audio-visual interventions and aspects of the exercise experience, offering a novel approach to understanding these complex interactions. A dataset compiled from the extant literature (k = 6) was used to construct the BN, specifying the network structure and learning the conditional probability distributions of the model. Through this framework, we examined how technological features—viewed through the lens of the Embodiment–Presence–Interactivity Cube—and music influence the exercise experience. The findings indicated that the pairwise combination of high interactivity and presence increased the probability of more positive affective valence. The three-way effect of high embodiment, presence and music increased the probability of high arousal as well as a more external focus of attention. The combination of high affective valence and high arousal was the strongest indicator of exercise enjoyment. Collectively, the findings offer new insights into how technological features of audio-visual interventions can shape the exercise experience, providing guidance for the optimal design of such interventions. These results have important implications for both research and practice, suggesting that practitioners should prioritise interactive audio-visual interventions with music to promote positive exercise experiences.