4 June 2026 in Acta Psychologica
Despite neuroscientific evidence indicating that music training remodels brain structure, behavioral findings concerning its transfer effects on children's executive functions (EF) remain characterized by significant controversy and heterogeneity. To resolve this ambiguity, this study strictly adhered to PRISMA guidelines, conducting a comprehensive systematic search across three major databases (Scopus, Web of Science, and PsycINFO). We synthesized 29 empirical studies (N = 3265) from five continents using Robust Variance Estimation (RVE) to evaluate intervention effects and examine the potential moderating role of macro-cultural context and instructional organization. The analysis revealed a significant, medium-sized facilitative effect of music education on overall EF (g = 0.62). Among subdomains, working memory showed the most substantial benefit, and the preschool period (ages 3-6) was identified as a developmental stage showing larger EF-related gains. Notably, macro-level cultural context (Individualism vs. Collectivism) did not independently moderate outcomes, challenging prior assumptions of a collectivistic advantage. However, at the micro-level, a significant "Culture × Instructional Structure" interaction emerged for working memory, suggesting a possible context-dependent pattern: children in individualistic cultures benefited more from high-structured (skill-oriented) training, whereas those in collectivistic cultures gained significantly more from low-structured (experience-oriented) activities. These findings suggest that the effects of music education on children's EF may vary as a function of instructional organization and broader cultural-educational context. However, the culture × instructional-structure pattern should be interpreted as hypothesis-generating, given the uneven subgroup evidence, variation in control-condition strength, and coarse country-level cultural coding. Future research should test this pattern under stronger active-control designs and more developmentally appropriate EF measurement before translating it into broad instructional recommendations.
15 August 2025 in bioRxiv
Music emotion recognition (MER) is challenging due to the ambiguous nature of audio features. For example, a fast tempo might indicate excitement or anger. To address this issue, we propose the Lyrics-Aware Gate Fusion Network LGFN, a novel cross-modal dynamic fusion approach. This network extracts audio features using ResNet for Mel spectrograms and 1D CNN for waveforms, while leveraging BERT for lyric embeddings. The key innovation of LGFN lies in its attentional alignment layer AAL, which effectively bridges the temporal misalignment between lyrics and audio. Additionally, the dynamic gate fusion module DGFM adaptively adjusts the weight of audio and text features based on their reliability. This allows the model to automatically determine the contribution of each modality, enhancing the accuracy of emotion recognition. Ablation studies on the music emotion recognition datasets MER1101 and DEAM2015 demonstrate the effectiveness of our proposed AAL and DGFM, while visualizing the gate weights offers insights into the model’s decision-making process. We also conduct comprehensive experiments on different human multimodal emotion recognition benchmarks, which show the superiority of our approach over the state-of-the-arts.