Pair-wise Pearson’s correlations were used to examine the similarity between the Combined-PC1 and the regional pattern of both case-control differences and normative aging. Spatial autocorrelation can inflate the false-positive rate when correlating two brain maps [35]. Therefore, a spatial permutation test that creates null models while preserving the spatial contiguity across regions was used to assess the significance of the spatial correlations between different maps [35, 36]. The null models were generated by correlating the spined regional values 10,000 times and then the empirical correlation coefficients were compared against the null distribution to determine the significance level, with p-spin < 0.05 deemed significant.
