What I see and what I feel: the influence of deceptive visual cues and interoceptive accuracy on affective valence and sense of effort during virtual reality cycling

BACKGROUND: How we feel during exercise is influenced by exteroceptive (e.g., vision) and interoceptive (i.e., internal body signals) sensory information, and by our prior experiences and expectations. Deceptive visual cues about one’s performance during exercise can increase work rate, without negatively impacting affective valence (good/bad responses) or perceived exertion. However, what is less understood is whether the perception of the exercise experience itself can be shifted, if work rate is held constant. Here we aimed to investigate whether deceptive vision—via illusory hills in a virtual reality (VR) cycling experience—alters affective valence and perceived exertion when physical effort is controlled. We also evaluated whether the accuracy with which one detects interoceptive cues influences the extent to which deceptive visual information can shift exercise experiences. METHODS: A total of 20 participants (10 female; 30.2 ± 11.2 yrs) completed three VR cycling conditions each of 10-min duration, in a randomised, counterbalanced order. Pedal resistance/cadence were individualised (to exercise intensity around ventilatory threshold) and held constant across conditions; only visual cues varied. Two conditions provided deceptive visual cues about the terrain (illusory uphill, illusory downhill; resistance did not change); one condition provided accurate visual cues (flat terrain). Ratings of affective valence (Feeling Scale) and of perceived exertion (Borg’s RPE) were obtained at standardised timepoints in each VR condition. Interoceptive accuracy was measured via a heartbeat detection test. RESULTS: Linear mixed effects models revealed that deceptive visual cues altered affective valence (f(2) = 0.0198). Relative to flat terrain, illusory downhill reduced affective valence (Est = −0.21, p = 0.003), but illusory uphill did not significantly improve affective valence (Est = 0.107, p = 0.14). Deceptive visual cues altered perceived exertion, and this was moderated by the level of interoceptive accuracy (Condition-Interoception interaction, p = 0.00000024, f(2) = 0.0307). Higher levels of interoceptive accuracy resulted in higher perceived exertion in the illusory downhill condition (vs flat), while lower interoceptive accuracy resulted in lower perceived exertion in both illusory hill conditions (vs flat) and shifts of greater magnitude. CONCLUSIONS: Deceptive visual cues influence perceptual responses during exercise when physical effort does not vary, and for perceived exertion, the weighting given to visual exteroceptive cues is determined by accuracy with which interoceptive cues are detected. Contrary to our hypotheses, deceptive visual cues did not improve affective valence. Our findings suggest that those with lower levels of interoceptive accuracy experience most benefit from deceptive visual cues, providing preliminary insight into individualised exercise prescription to promote positive (and avoid negative) exercise experiences.