Negative associations between corpus callosum midsagittal area and IQ in a representative sample of healthy children and adolescents

Abstract

Documented associations between corpus callosum size and cognitive ability have heretofore been inconsistent potentially owing to differences in sample characteristics, differing methodologies in measuring CC size, or the use of absolute versus relative measures. We investigated the relationship between CC size and intelligence quotient (IQ) in the NIH MRI Study of Normal Brain Development sample, a large cohort of healthy children and adolescents (aged six to 18, n = 198) recruited to be representative of the US population. CC midsagittal area was measured using an automated system that partitioned the CC into 25 subregions. IQ was measured using the Wechsler Abbreviated Scale of Intelligence (WASI). After correcting for total brain volume and age, a significant negative correlation was found between total CC midsagittal area and IQ (r = -0.147; p = 0.040). Post hoc analyses revealed a significant negative correlation in children (age<12) (r = -0.279; p = 0.004) but not in adolescents (age≥12) (r = -0.005; p = 0.962). Partitioning the subjects by gender revealed a negative correlation in males (r = -0.231; p = 0.034) but not in females (r = 0.083; p = 0.389). Results suggest that the association between CC and intelligence is mostly driven by male children. In children, a significant gender difference was observed for FSIQ and PIQ, and in males, a significant age-group difference was observed for FSIQ and PIQ. These findings suggest that the correlation between CC midsagittal area and IQ may be related to age and gender.

Figure 1. The midsagittal corpus callosum divided into 25 segments.

Figure 2. Regression of total corpus callosum area against total brain volume.

Figure 3. Correlations between corpus callosum midsagittal area and IQ: combined sample and sample partitioned by gender.

Partial correlation coefficients and associated significance values for Full-Scale IQ measurements are shown in the first and second columns of callosal maps. Correlation coefficients and associated significant values for Performance IQ are shown in the third and fourth columns, and for Verbal IQ in the fifth and sixth columns. The combined sample is shown in the first row, the males in the second row, and the females in the third row. The upper color bar encodes the r-values that depict the magnitude and direction of correlations, and the lower color bar encodes the level of significance.

Figure 4. Correlations between corpus callosum midsagittal area and IQ: combined sample and sample partitioned by age.

Partial correlation coefficients and associated significance values for Full-Scale IQ measurements are shown in the first and second columns of callosal maps. Correlation coefficients and associated significant values for Performance IQ are shown in the third and fourth columns, and for Verbal IQ in the fifth and sixth columns. The combined sample is shown in the first row, the children in the second row, and the adolescents in the third row. The upper color bar encodes the r-values that depict the magnitude and direction of correlations, and the lower color bar encodes the level of significance.

Figure 5. Correlations between corpus callosum midsagittal area and IQ: combined sample and sample partitioned by age and gender.

A. Partial correlation coefficients and associated significance values for Full-Scale IQ measurements are shown in the first and second columns of callosal maps. Correlation coefficients and associated significant values for Performance IQ are shown in the third and fourth columns, and for Verbal IQ in the fifth and sixth columns. The male children are shown in the first row, the males adolescents in the second row, the female children in the third row, and the female adolescents in the fourth row. B. Gender and age interactions. Gender interaction in children is shown in the first row and age-group interaction is shown in the second row. The upper color bar encodes the r-values that depict the magnitude and direction of correlations, and the lower color bar encodes the level of significance for the correlation coefficients and the interactions.

Figure 6. Direct comparison of VIQ and PIQ correlations in male children.

Figure 7. Age by IQ interaction: combined sample and sample partitioned by gender.

A general linear model was used with an age by IQ interaction term to examine the interaction between IQ and age as a continuous variable. Significance values are displayed as callosal maps with Full-Scale IQ results shown in the first column, Performance IQ results in the second column, and Verbal IQ results in the third column. Values for the combined sample are shown in the first row and results for males and females are shown in the second and third rows respectively. The color bar encodes the level of significance for the interactions.