Does flow state predict heart rate during self-paced cycling exercise in low and high immersive virtual environments?

Flow is an intrinsically rewarding state where individuals feel fully absorbed in an activity, experiencing high levels of control and ease, even during demanding exercise tasks. Virtual reality (VR) offers an immersive environment and when combined with exercise is recognized to improve the overall experience and enjoyment. However, it is not known if "ow impacts the level of exercise intensity an individual may self-select using different types of VR. Therefore, the aim of the study was to explore the association between sub-components of the Flow State Scale (FSS) and heart rate (HR) during self-selected exercise in low and high immersive VR conditions. Nineteen participants (n = 10 male, n = 9 female; age: 32 ± 9 years; height: 169 ± 10 cm; weight: 68 ± 13 kg) completed 15-min of self-selected cycling exercise in four conditions - No VR and VR using a computer screen (Low Screen), room-projector (Low Room), and head-mounted display (High HMD). HR was recorded during exercise. After each session, participants completed the FSS, which measured nine dimensions: challenge-skills balance; merging of action and awareness; clear goals; unambiguous feedback; concentration on the task at hand; sense of control; loss of self-consciousness; transformation of time and autotelic experience. Separate ANOVAs with repeated measures assessed the effect of VR on HR, while linear regression analyzed the relationship between FSS sub-components and HR.A one-way ANOVA with repeated measures revealed no signi!cant main effect of condition on HR (mean ± SD, range 125 – 130 ± 9 - 16 bpm; F(3,54) = .83, p = .48). However, separate regression analyses revealed that different sub-components of the FSS explained variance in HR depending on condition. In the No VR condition, transformation of time was positively associated with HR (r = .63, R2 = 0.36, B = 7.41, p =.004). In the low-immersive VR conditions loss of self-consciousness was negatively associated with HR (Low Screen, r = -0.510, R2 = .26, B = 12.34, p = .03; Low Room, r = -.56, R2 = .31, B = 8.61, p = .014). In the High HMD condition, action and awareness was positively associated with HR (r = .65, R2 = .42, B = 5.68, p = .003). The !ndings indicate that "ow states varied across VR conditions, in"uencing the level of selfselected exercise intensity. In the No VR condition, participants who felt a greater alteration in their perception of time worked harder. While, in the High HMD condition, those with a high awareness showed an increase of effort. In the low VR environments, the more self-conscious an individual was, the harder they worked. The !ndings indicate that speci!c sub-components of "ow (a) are in"uenced by the level of VR immersion, and (b) contribute to how hard an individual chooses to work during self-selected exercise. Findings are important for exercise promotion, showing how VR can enhance engagement and effort by targeting speci!c "ow states, improving adherence to exercise.