Investigating cognitive-physical task interaction during self-paced cycling: A Granger causality study

Self-pacing physical exercise is thought to rely on high-order cognitive processing (e.g., attentional control to monitor afferent cardiovascular feedback for exercise goals). Therefore, performing a cognitive task during a self-paced exercise could lead to cognitive-physical interactions. We explored cognitive-physical interactions by applying time-domain Granger Causality (a correlation analysis that uses a temporal series of one variable to improve the prediction of values in a temporal series of another variable given its past values) to data that combined 20 min of indoor self-paced high-intensity cycling and a Sustained Attention to Response cognitive task, and to data that combined 30 min of indoor self-paced high-intensity cycling and a stimulus-response conflict task. Moreover, we explored whether greater experience in cycling would reduce the need for exerting cognitive attentional control and therefore dual-task effects. The results showed that the experienced cycling group (i.e., at least 4 days of weekly cycling training in the last 3 years) demonstrated better overall physical performance than the non-experienced cycling group (i.e., at least 4 days of weekly training in another endurance sport different to cycling in the last 3 years), while no evidence of differences in cognitive performance was obtained. The results also showed that reaction times and power output interacted bidirectionally in a reduced sample of experienced cyclists and non-experienced cyclists. Hence, cognitive-physical interactions may not be excluded for every single high-fit athlete, irrespective of their particular exercise experience. Our study highlights the value of GC to investigate cognitive-physical interactions during self-paced exercise at the individual level.