Field studies using conventional univariate methods fail to detect physiological rhythms (melatonin, cortisol, sleep architecture) that are consistently documented under laboratory-controlled conditions, suggesting that temporal factors operate on systemic organization rather than isolated metrics. Network analysis quantifies conditional dependencies among variables simultaneously, enabling the detection of organizational changes independent of mean-level shifts. As an exploratory proof-of-concept, this study examines whether network comparison methods detect temporal reorganization of psychophysiological integration invisible to conventional repeated-measures analysis in elite athletes; no confirmatory causal inference is drawn. Twelve elite male athletes completed psychophysiological assessments (mood states, subjective well-being, neurocognitive performance, postural control, intermittent endurance) at three lunar phases (New Moon, First Quarter, Full Moon) in a counterbalanced design. Conventional repeated-measures ANOVA detected no temporal effects on isolated variables (all F p > 0.05, η2 p M = 0.31, padj = 0.008) and connectivity (S = 8.42, padj = 0.029). Centrality systematically redistributed from capacity and well-being constructs (New Moon: vigour z = 1.82) to recovery markers (Full Moon: stress z = 2.13). Twelve connections differed significantly, including enhanced stress-confusion coupling (Δr = +0.39, p = 0.003). This proof-of-concept demonstrates that temporal measurements can exhibit systemic reorganization despite stable component means, resolving apparent contradictions between laboratory-detected physiological rhythms and field-based null findings. Temporal control measurements and multi-cycle replication are required to definitively attribute observed reorganization to lunar periodicity rather than generic temporal variation.
Yousfi et al. (2026) studied this question.