The corpus callosum as anatomical marker of intelligence? A critical examination in a large-scale developmental study

Abstract

Intellectual abilities are supported by a large-scale fronto-parietal brain network distributed across both cerebral hemispheres. This bihemispheric network suggests a functional relevance of inter-hemispheric coordination, a notion which is supported by a series of recent structural magnetic resonance imaging (MRI) studies demonstrating correlations between intelligence scores (IQ) and corpus-callosum anatomy. However, these studies also reveal an age-related dissociation: mostly positive associations are reported in adult samples, while negative associations are found in developing samples. In the present study, we re-examine the association between corpus callosum and intelligence measures in a large (734 datasets from 495 participants) developmental mixed cross-sectional and longitudinal sample (6.4-21.9 years) using raw test scores rather than deviation IQ measures to account for the ongoing cognitive development in this age period. Analyzing mid-sagittal measures of regional callosal thickness, a positive association in the splenium of the corpus callosum was found for both verbal and performance raw test scores. This association was not present when the participants' age was considered in the analysis. Thus, we did not reveal any association that cannot be explained by a temporal co-occurrence of overall developmental trends in intellectual abilities and corpus callosum maturation in the present developing sample.

Keywords: Corpus callosum; Development; Intelligence; Longitudinal studies; MRI; Structure–function association.

Fig. 1

Intellectual development. Spaghetti plot illustrating the development of a raw verbal (v-RS) and b raw performance (p-RS) test scores for 495 participants of which 239 provided measures for two time points (connected with gray lines). Solid black line represents the locally weighted scatterplot smoothing (LOWESS, smoothing factor, f = 0.25)

Fig. 2

Association raw verbal test score (v-RS) and regional callosal thickness (left column) and the interaction of v-RS with Sex (right column). The rows represent the three analysis steps (full models described in “Method” section), with the statistical design of step 2 compared to step 1 additionally including TIV as covariate, and step 3 compared to step 2 additionally including age-related variables (i.e., Age linear, Age squared, and the interaction of v-RS and Age). At each of the 60 segments of the corpus callosum, the direction and magnitude of the association is visualized by a circle, whereby the size of the circle is proportional to empirical t value and the color, red vs blue, codes positive and negative associations, respectively. Lighter red and lighter blue indicate significant associations, with the significance level being adjusted to a false-discovery-rate (FDR) of 0.05. The anterior corpus callosum is on the left side of each panel

Fig. 3

Association of raw performance test score (p-RS) and of regional callosal thickness (left column) and the interaction of p-RS with Sex (right column). As in Fig. 2, rows represent the three analysis steps. At each callosal segment, the direction and magnitude of the association is visualized by a circle. The size of the circle is proportional to empirical t value and positive and negative associations are coded red and blue respectively. Lighter red and lighter blue indicate significant associations [significance level adjusted to a false-discovery-rate (FDR) of 0.05]. The anterior corpus callosum is on the left side of each panel

Fig. 4

Spaghetti plots at location of maximum effect of analysis step 1. Plot a shows the main effect of verbal raw (v-RS) test scores on callosal thickness (residualized for all other effects) in segment 57 (red line). Plot b illustrates the interaction of v-RS with Sex with filled circles representing female (red line) and open circles representing male (blue line) participants. Plot c illustrates the main effect of performance raw (p-RS) test scores on residualized callosal thickness in segment 58 (red line). Plot d shows the interaction of p-RS with Sex with filled circles representing female (red line) and open circles representing male (blue line) participants

Fig. 5

Spaghetti plots at location of maximum effect of analysis step 2. Plot a illustrates the main effect of verbal raw (v-RS) test scores on residualized callosal thickness in segment 57 (red line depicting linear fit) after considering TBV differences. Likewise, plot b shows the main effect of performance raw (p-RS) test scores on residualized callosal thickness in segment 58 (red line depicting linear fit)

Fig. 6

Interaction of raw test scores (performance, p-RS; verbal, v-RS) and Age in predicting regional callosal thickness in analysis step 3. At each callosal segment, the direction and magnitude of the association is visualized by a circle. The size of the circle is proportional to empirical t value of the predictor coding the interaction; positive and negative associations are coded red and blue, respectively. Significance level is adjusted to a false-discovery-rate (FDR) of 0.05. The anterior corpus callosum is on the left side of each panel