Arts education is widely believed to foster creativity and innovation, alongside subject-specific skills. Many children participate in extracurricular music and visual arts, supplementing formal art classes. In Japan, all students receive art classes, and many participate in extracurricular arts clubs, often for several hours per week. Existing research shows mixed results regarding the relationship between music education and academic performance; while some studies indicate a positive association, others show no significant relationship. Similarly, the link between visual arts education and academic performance is debated. Some research suggests a positive correlation, while others highlight the correlational nature of these studies and lack of control for confounding variables such as IQ and socioeconomic status (SES). This study aimed to investigate (1) whether music and visual arts education (both in- and out-of-school) is associated with general academic performance; and (2) whether this association is mediated by improvements in music and visual arts skills. A two-year longitudinal study was designed to analyze these relationships, accounting for factors such as baseline academic performance, SES, and participation in other extracurricular activities (particularly sports). The study tracked students' art-related scores and general academic performance across two years, allowing for the examination of changes and the mediating effects of art-related scores. The model also considered direct and indirect effects: a direct effect of extracurricular participation on later academic performance and indirect effects mediated by improvement in arts-related scores in both the first and third years of junior high school.

A substantial body of research explores the impact of music education on academic performance. Studies have shown a positive correlation between extracurricular music lessons and academic outcomes. For example, Wetter et al. (2009) found a significant positive association, while Southgate and Roscigno (2009) demonstrated a relationship even after controlling for SES and family background. However, some studies show no significant relationship, and others have utilized quasi-experimental designs indicating higher IQ scores in musically trained groups. The relationship between visual arts education and academic performance is similarly contested. Vaughn and Winner (2000) reported higher SAT scores among students selecting visual arts courses, while studies on Visual Thinking Strategies (VTS) have shown positive impacts on academic test scores and critical thinking. However, these studies have been critiqued for their correlational nature and lack of control variables. Winner et al. (2013) proposed a model of transfer effects in arts education, suggesting that arts education's impact on non-artistic outcomes is a form of transfer, potentially mediated by the development of artistic skills. This study aimed to test Winner's model by examining both direct and indirect effects, mediated by changes in arts-related subject scores.

This two-year longitudinal study involved 488 seventh-grade students (259 boys, 229 girls) from six public junior high schools in Sapporo, Japan. Data were collected at the end of the first and third years of junior high school. The measures included: 1. **Grade Point Average (GPA):** GPAs in music (MUGP), visual arts (VAGP), and general academics (ACGP) were obtained from school records. The ACGP was calculated using scores from five core academic subjects (Japanese, mathematics, social studies, science, and English). 2. **Socioeconomic Status (SES):** Assessed using parental household income and maternal education level. 3. **Learning Habits:** Measured by self-reported study time on weekdays and weekends. 4. **Extracurricular Activities (ECA):** Participation in music (brass band, choir), visual arts, sports, and other cultural clubs was recorded. Structural Equation Modeling (SEM) was employed to test the hypothesized model. The model included direct paths from ECA participation to ACGP in the third year and indirect paths mediated by changes in MUGP and VAGP scores between the first and third years. Control variables included SES, learning habits, and sex. The study also conducted a secondary analysis excluding students participating in sports-related ECAs to assess the influence of sports on the results. Data analysis was performed using R.

The SEM showed an adequate fit for the primary model (N=488). Regarding music ECAs, while the indirect effect (mediated by improved MUGP scores) was significantly positive in both years, the direct effect was significantly negative in both years, resulting in non-significant total effects. For visual arts ECAs, the indirect effect (mediated by VAGP scores) was significantly positive in both years, with the total effect marginally insignificant. The analysis excluding sports ECAs (N=147) yielded similar indirect effects for both music and visual arts but showed a significant positive total effect for visual arts ECAs only. This suggests that while improvement in arts-related scores mediate the positive relationship between extracurricular arts participation and academic performance, other factors might also influence the total effect, particularly the direct effect of music participation. Figure 2 and 3 illustrates the model and indirect effects, showing the significant paths for MUGP and VAGP changes. Table 2 displays the demographic characteristics of the participating children, while Table 3 details the standardized direct, indirect and total effects of music and visual arts ECAs. Supplementary Tables 1, 2, and 3, and Figure 1 provide additional details.

The results generally support the hypothesis that participation in music and visual arts ECAs positively impacts academic performance, mediated by the development of artistic skills. The mediating role of improved arts scores highlights the importance of arts education in enhancing cognitive and other skills transferable to academic domains. The inconsistent findings regarding the direct effects of music ECAs warrant further investigation. The negative direct effect in the primary analysis might reflect the initial burden of intensive music practice, potentially impacting time available for core subjects. However, the positive indirect effects highlight the long-term benefit of such involvement. The significant total effect of visual arts ECAs in the analysis excluding sports suggests that visual arts involvement might have a more consistent and independent impact on academic performance. This study contributes to the literature by providing longitudinal evidence, controlling for important confounders, and examining the mediating effects of arts skills development. The study's findings emphasize the value of arts education and its potential contribution to overall academic success.

This study demonstrates a positive association between extracurricular participation in music and visual arts and improved academic performance, particularly when mediated by improvements in arts-related skills. While the direct effects of music involvement were unexpectedly negative, the indirect positive effects highlight the long-term benefits. Further research should explore the nature of arts education programs and activities to enhance their positive impact on academic achievement. Investigating specific aspects of music and visual arts training and exploring diverse cultural contexts would greatly enhance understanding of this relationship.

The study's correlational design limits the ability to definitively establish causality. The sample size, particularly for students participating in music and visual arts ECAs, was relatively small, potentially affecting the generalizability of the results. The unbalanced design impacted comparison of the total effect. The self-reported nature of some measures (e.g., learning habits) could introduce bias. Further, the cultural context specific to Japan needs to be considered when interpreting the results and its applicability to other cultural contexts needs to be studied further. Future studies should utilize experimental designs and address these limitations.