Perceived psychological safety, state social connection, and student flourishing in collaborative music activities: a three-wave longitudinal study.

Collaborative music activities can support relationships and well-being, but the present study did not directly test music-specific mechanisms such as synchrony or collective expression. Instead, it examined a general interpersonal process within higher music education: whether students' individually perceived psychological safety was longitudinally associated with flourishing through state social connection. A three-wave survey was completed by 341 university music students from 23 groups at three Chinese universities, with approximately 6 weeks between waves. Psychological safety was analyzed as an individual perception rather than as a shared team construct. Robust maximum likelihood estimation, full information maximum likelihood for partial missing data, cluster-robust standard errors, and 5,000 bootstrap samples were used. After baseline social connection and autoregressive effects were controlled, T1 perceived psychological safety was associated with higher T2 state social connection (β = 0.350, p < 0.001), which was associated with higher T3 flourishing after prior flourishing was controlled (β = 0.274, p < 0.001). The indirect effect was significant (B = 0.0949, β = 0.096, 95% BCa CI [0.054, 0.142]); the direct effect was not (β = -0.004, p = 0.941). Model comparisons did not uniquely support a one-way sequence: AIC favored a reciprocal model, BIC favored the more parsimonious hypothesized model, and prior social connection significantly predicted later perceived psychological safety. The findings therefore support a longitudinal indirect association alongside a plausible reciprocal process between perceived safety and connection. They should be understood as evidence for a general group-process model examined in a heterogeneous music-education context, not as evidence for a uniquely musical or causal mechanism.