3.4.1 Correlation Analysis 
Table 2 displays the results of the Pearson correlation analysis of the variables. A positive and strong correlation between the amplitudes of MMN and N1 initial adaptation was observed (r35 = .67, P < .001). This indicates that participants with a stronger N1 initial adaptation effect exhibited a larger MMN (see Figure 5a). No other significant correlations related to the MMN amplitude were identified (all P > .008 based on Holm-Bonferroni correction).
3.5.2 Regression Analysis
The relationship between adaptation and MMN was further examined through a backward stepwise regression to elucidate which adaptation effect contributes to the MMN and to what extent. All four adaptation variables measured by amplitudes were included in the regression model, with the MMN peak amplitude as the DV. Eventually, only N1 initial adaptation (β (standardized beta) = 0.72; P < .001) and P2 subsequent adaptation (β = 0.33; P = .009) emerged as significant predictors in the final model (F2,34 = 20.50; P < .001; see Figure 5b for the correlation between the P2 subsequent adaptation effect and MMN). The squared semipartial correlation, indicating the unique contribution of each independent variable to the dependent variable (Aloe & Becker, 2010), revealed that these two variables collectively explained over 50% of the MMN variance (49.98% and 10.43%, respectively). The standardized coefficients of each variable and the adjusted R2 of the initial and final models of the adaptation and MMN are presented in Table 3.
To ensure that the initial and subsequent adaptation effects did not influence each other, we conducted partial correlations. Specifically, we examined the partial correlation between the MMN and the N1 adaptation amplitudes while controlling for the N1 subsequent adaptation effect, as well as the partial correlation between the MMN and the P2 subsequent adaptation amplitudes while controlling for the P2 initial adaptation effect. These analyses yielded consistent results with the corresponding correlations reported earlier, indicating that the initial and subsequent adaptation effects relevant to the regression analysis were not confounded. Detailed results are presented in Table S2 in Supplementary Material II. 
3.5.3 Additional Analyses
Although efforts were made to balance the trial numbers by including only the final standards (but not all standards) in the analyses, the trial numbers remained unequal between the deviant, 2nd tone, and final tone conditions. To address this issue, we conducted the same analyses with an equal number of trials in these three conditions, based on the trial number of the final tone condition (the condition with the least trial numbers) in each participant. The results were consistent with our previous findings, showing that N1 initial adaptation and P2 subsequent adaptation amplitudes predicted the MMN amplitude in the regression model. However, N1 subsequent adaptation also emerged as a significant predictor. Further details can be found in Supplementary Material III.