Fig. 2: Statistical analysis outline for identifying and summarizing the common and unique brain abnormalities of SZ and ASD. | Communications Biology

Fig. 2: Statistical analysis outline for identifying and summarizing the common and unique brain abnormalities of SZ and ASD.

From: Evidence of shared and distinct functional and structural brain signatures in schizophrenia and autism spectrum disorder

Fig. 2

The statistical analysis procedure is consistent across all neuroimaging measures mentioned in Fig. 1. Regarding each measure, analysis of variance (ANOVA) and two-tailed two-sample t-tests were first performed, resulting in group differences for HC vs. SZ, HC vs. ASD, and SZ vs. ASD. For the measures passing ANOVA, different types of changes were then summarized, including the disorder-common decrease compared to HC (the voxels with T-values > 0 in both HC vs. SZ and HC vs. ASD), the disorder-common increase compared to HC (the voxels with T-values < 0 in both HC vs. SZ and HC vs. ASD), the SZ-unique decrease (the voxels with both T-values < 0 in HC vs. ASD, and T-values > 0 in HC vs. SZ), and the ASD-unique decrease (the voxels with both T-values < 0 in HC vs. SZ and T-values > 0 in HC vs. ASD). For each of the four types of changes, the percentage was calculated as the number of neuroimaging measures relating to the change divided by the number of measures passing ANOVA. Among the measures with the disorder-common decrease (or increase), the percentage of ASD-weaker decrease (or increase) than SZ, which showed T-values < 0 in SZ vs. ASD in the disorder-common decrease (or showed T-values > 0 in SZ vs. ASD in the disorder-common increase), were further summarized.

Back to article page