The moderating roles of a "growth mindset" and psychological flexibility on performance anxiety in violin learners

Music performance anxiety (MPA), a common situational anxiety subtype, impairs motor execution and stress response in high-stakes violin learning, a context with rigorous technical and evaluative demands. This cross-sectional non-interventional human subject research strictly followed standardized ethical review protocols.

This study explored the correlations of growth mindset (GM, cognitive) and psychological flexibility (PF, emotional regulatory) with MPA, examining PF's moderating effect and gender differences in their anxiety-mitigating pathways.

A cross-sectional survey of 306 Chinese university violin learners used validated scales to assess GM, PF, and MPA. Data analyses included correlation and hierarchical regression modeling, with gender as a stratification variable. Confirmatory factor analysis (CFA) was additionally conducted to verify construct validity; average variance extracted (AVE), composite reliability (CR), and discriminant validity indices were reported to supplement exploratory factor analysis (EFA) results. Harman's single-factor test was performed to statistically detect common method bias.

GM and PF both negatively correlated with MPA (r = -0.574, p < 0.01; r = -0.515, p < 0.01). PF significantly moderated the GM-MPA relationship, amplifying GM's anxiety-mitigating effect, and this effect was more pronounced in female learners. After recoding simple slopes and rechecking interaction term centralization procedures, the inconsistent surface coefficient of high PF subgroup was clarified with complete simple slope interpretation and visualized interaction diagram.

GM and PF are key protective factors for situational performance anxiety in skill learning, with a synergistic cognitive-emotional interaction. Findings extend anxiety coping research and provide actionable gender-responsive cognitive-emotional intervention strategies for skill-learning populations. All interpretations are limited to correlational associations rather than causal causal inferences given the cross-sectional design.