Psychophysiological States of Athletes: A Recurrence Network Approach with Phase Profiles

An increasing body of research suggests that athletes’ psychophysiological states develop out of complex and dynamic processes. However, a notable gap exists between the recognition of these complex dynamics and the application of methodological tools to capture such dynamics. The present study bridges this gap by using a recurrence network analysis strategy renowned for its proficiency in detecting recurring (stable) states and transitions between states of multivariate time series. We collected data from three teams of professional youth football players for up to three consecutive seasons. For every training and match day, we used a tailor-made application and wearable sensors to measure psychological and physiological factors (e.g., self-efficacy, heart rate). Next, we computed multiplex (cumulative) recurrence networks for each athlete and created so-called state profiles. With this technique, patterns such as stability, critical fluctuations, phase transitions, and resilience can be derived from multivariate time series. Our findings show individual-specific and dynamic networks of interacting psychological and physiological variables before, during, and after sustaining an injury. For instance, these networks show that individual athletes transition from relatively stable states with recurring patterns before an injury to new states during an injury, to eventually return to pre-injury stable states after an injury. This analysis strategy can have widespread implications. For instance, a deeper understanding can be gained of the type of state and its stability when performing optimally, or when an athlete loses resilience and is vulnerable to injuries. Such investigations can advance scientific insights into the complex dynamics of health and performance in the sports context and beyond. Accordingly, in practice, insights could be used in daily monitoring and decision-making.