Does mindfulness mediate the relationship between anxiety and second language performance? A meta-analytic structural...

Anxiety is a well-established affective factor that impairs second language (L2) learning, whereas mindfulness may alleviate anxiety and enhance cognitive processing. However, whether mindfulness serves as a mechanism linking anxiety to L2 performance remains unclear. This study aimed to investigate the mediating role of mindfulness in the relationship between anxiety and L2 performance.

A systematic review was conducted for studies published up to April 2026. Data were extracted from observational studies (Pearson's correlations) and from baseline scores of intervention studies. Random-effects univariate meta-analyses were performed for each bivariate relation, and meta-analytic structural equation modeling (MASEM) was applied to test the mediation model. Moderator analyses were conducted using subgroup comparisons.

The meta-analysis included 13 studies (17 effect sizes) for the anxiety-mindfulness relation, 11 studies (14 effect sizes) for mindfulness-L2 performance, and 35 studies (41 effect sizes) for anxiety-L2 performance. Univariate meta-analyses showed that anxiety was negatively correlated with mindfulness [r = -0.32, 95% CI (-0.36, -0.27)] and with L2 performance [r = -0.34, 95% CI (-0.38, -0.30)], while mindfulness was positively correlated with L2 performance [r = 0.20, 95% CI (0.10, 0.28)]. MASEM revealed a saturated model with the following standardized path coefficients: anxiety → mindfulness β = -0.32 (p < 0.001), mindfulness → L2 performance β = 0.11 (p < 0.001), and anxiety → L2 performance (direct effect) β = -0.30 (p < 0.001). The indirect effect of anxiety on L2 performance through mindfulness was β = -0.03 (p < 0.001), and the total effect was β = -0.34 (p < 0.001).

Mindfulness partially mediates the anxiety-L2 performance relationship, although the indirect effect is very small (β = -0.03). Anxiety is directly associated with lower L2 performance and also indirectly related to it via reduced mindfulness. Because this estimate was derived from pooled correlations across studies, it reflects an aggregate mediation pattern rather than a direct within-sample observation, and thus warrants cautious interpretation. Future research should explore additional mechanisms and use longitudinal designs to establish causality.

https://www.crd.york.ac.uk/PROSPERO/view/CRD420261377122, identifier: CRD420261377122.