Network-based markers of multisystem coordination: A transformative framework to assess physiological function during exercise

Background: Human function during exercise emerges from dynamic network interactions among multiple physiological systems. Cardiovascular, respiratory, and muscular systems must precisely integrate ventilation, oxygen delivery, and muscle activation to meet exercise demands. While decades of research have provided a rich understanding of how these systems function individually, the direct mechanisms regulating their dynamic coupling as a network—and how these networks reorganize with exercise and fatigue—remain unexplored. Objective: We introduce three novel network-based markers quantifying multisystem coordination between (i) cardiovascular and muscular systems (cardio-muscular coupling); (ii) respiratory and muscular systems (respiratory-muscular coupling); and (iii) across multiple muscles (inter-muscular coupling). The objective of this study was to characterize multisystem network coordination during graded exercise, and assess how these networks respond to fatigue. We hypothesized that increasing fatigue would drive a progressive reorganization of multisystem coordination to meet exercise demands. Methodology: Eleven males (21.9 ± 2.7 yrs) and twenty-seven females (22.5 ± 3.9 yrs) performed a graded cycling test to exhaustion, starting at 0 W with 25 W·min - ¹ increments. Continuous synchronous recordings included three-lead electrocardiography (ECG), a respiration waveform via chest belt, and electromyography (EMG) from right and left vastus lateralis (VS) and erector spinae (ES). Multisystem coordination was measured using the amplitude–amplitude cross-frequency coupling (ACFC) method. First, instantaneous heart rate and respiratory rate were extracted from ECG and respiration. Second, EMG signals were decomposed into ten frequency bands F1–F10, representing distinct neuromuscular processes. Last, Pearson correlation coefficients (C) were computed as the ACFC outcome between: (i) heart rate and EMG frequency bands (cardio-muscular); (ii) respiratory rate and EMG bands (respiratory-muscular); and (iii) EMG bands across all pairs of muscles (inter-muscular). To examine fatigue effects, ACFC analyses were performed separately for the Beginning (first third) and End (last third) segments of the test. Results: All network markers showed moderate C values, indicating synchronous activity among cardiovascular, respiratory, and muscular systems (cardio-muscular Formula: see text = 0.45 ± 0.12; respiratory-muscular Formula: see text = 0.56 ± 0.05; inter-muscular Formula: see text = 0.34 ± 0.16). With fatigue accumulation comparing Beginning vs. End segments, Formula: see text values increased for cardio-muscular (0.33 ± 0.12 vs. 0.42 ± 0.19; p < 0.05) and respiratory-muscular coupling (0.15 ± 0.12 vs. 0.50 ± 0.09; p < 0.01). However, inter-muscular analyses revealed a reorganization pattern, with decreased Formula: see text values between the bilateral VL muscle pair (0.35 ± 0.15 vs. 0.25 ± 0.08; p < 0.05), and increased Formula: see text for VL-ES and ES-ES muscle pairs (0.18 ± 0.07 vs. 0.31 ± 0.04; p < 0.05). Conclusions: Fatigue drives a reorganization of multisystem coordination, with increased cardio-muscular and respiratory-muscular coupling and heterogeneous changes in inter-muscular coupling. This differentiated response in multisystem coordination across cardiovascular, respiratory, and muscular systems reflects a compensatory mechanism to cope with the increasing exercise demands. This network-level adaptation shows that physiological responses to exercise arise not only from isolated systems, but also from their dynamic coupling as an integrated network. These network-based markers provide a transformative framework to assess human function beyond traditional physiological measures. This abstract was presented at the American Physiology Summit 2026 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.