Structural asymmetry of the insula is linked to the lateralization of gesture and language

Abstract

The control of gesture is one of the most left-lateralized functions, and the insular cortex is one of the most left-biased structures in the human brain. Therefore, we investigated whether structural asymmetries of the insula are linked to the organization of functional activity during gesture planning. We reconstructed and parcellated the insular cortex of 27 participants. First, we tested 15 strongly left-handed individuals because of a higher incidence of atypical organization of functions such as gesture and language in such a population. The inter-hemispheric structural asymmetries were compared with the lateralization of activity for gesture in the supramarginal gyrus (the hotspot of signal increase regardless of the gesturing hand) and Broca's area (the hotspot of signal increase for language production). The more pronounced leftward structural asymmetries were accompanied by greater left-hemisphere dominance for both of the studied functions. Conversely, an atypical, bilateral or rightward functional shift of gesture and language was accompanied by an attenuated leftward asymmetry of the insula. These significant relationships were driven primarily by differences in surface area. Subsequently, by adding 12 right-handed individuals to these analyses we demonstrated that the observed significant associations are generalizable to the population. These results provide the first demonstration of the relationships between structural inter-hemispheric differences of the insula and the cerebral specialization for gesture. They also corroborate the link between insular asymmetries and language lateralization. As such, these outcomes are relevant to the common cerebral specialization for gesture and language.

Keywords: Broca's area; human; praxis; right-hand; supramarginal gyrus; verbal fluency.

Figure 1

Regions of interest (ROIs) shown on the average cortical surface of all left-handed participants involved in this study. (A) ROIs used for structural analyses. The maps were derived from the Destrieux et al. (2010) atlas implemented in FreeSurfer. Colour codes: light blue denotes pars triangularis; pale green, pars opercularis; light green and pale yellow, insula; dark yellow, planum temporale; orange, supramarginal gyrus; red, superior parietal lobe. (B) ROIs used for analyses of the lateralization of functional activity. The binarized and bilateral maps of Broca’s area defined as BA44/45 (depicted in yellow), supramarginal gyrus (green) and insular cortex (blue). These stereotaxic maps were thresholded (to include only voxels with at least a 50% chance of belonging to a respective ROI) and binarized. These probabilistic masks were obtained from the Jülich histological atlas (Broca’s area, supramarginal gyrus; Amunts et al., ; Caspers et al., ; Eickhoff et al., 2007) or from the Harvard–Oxford atlas (insular cortex; Desikan et al., 2006) implemented in FSL.

Figure 2

Vertex-wise analysis of surface asymmetries measured across the whole brain shown on an inflated average brain of all study participants. Significant leftward surface asymmetries (depicted in warm colours) were located in the insular cortex – a structure of primary interest for the present study – supramarginal gyrus and post-central gyrus. Significant rightward surface asymmetries were present inferiorly in the parieto-occipital sulcus, and the anterior part of the cingulate gyrus.

Figure 3

Gesture- and language-related activity shown on the mean cortical surface of all left-handed participants’ brains. (A) Functional activity for gesture planning averaged across hands and gestures. In the majority of participants this activity was largely symmetrical, with the exception of the anterior supramarginal gyrus (SMG), anterior intraparietal sulcus (aIPS) and the superior parts of the superior parietal gyrus (SPG), where it was exclusively on the left. The bilateral involvement was observed in the insular cortex, the ventral premotor and dorsal premotor cortices, the supplementary motor complex (SMA complex), the middle frontal gyrus, the inferior temporal gyrus and the cerebellum. The angular gyrus activity was modulated on the right. (B) Functional activity in a language task (verbal fluency). Broca’s area, and insular and orbitofrontal cortices were engaged primarily on the left, whereas the fundus of aIPS was engaged exclusively on the left. The SMA complex, the anterior cingulate cortex and the majority of subcortical structures were modulated bilaterally. Given that the contrast reported was against the resting baseline, there is also a clear bilateral involvement of striate and extrastriate visual areas. (C) A conjunction test showing areas involved independently of the task (across gesture and language). Except for the middle frontal involvement on the right, the insular cortex and the SMA complex were engaged bilaterally. The small cluster of ventral premotor activity was exclusively left lateralized and the dorsal premotor cluster was exclusively right lateralized.

Figure 4

Laterality indices (LIs) for each left-handed participant. (A) LIs for gesture measured in the supramarginal gyrus (BA40). For display purposes individual LIs were sorted in ascending order. (B) LIs for language assessed in Broca’s area (BA44/45), presented in ascending order.

Figure 5

Correlations between the insular surface asymmetries and the lateralization of functional activity for gesture and language. (A) Gesture lateralization and insular surface asymmetries. There was a significant correlation between their respective LIs. (B) Language lateralization and insular surface asymmetries. The lateralization of activity in Broca’s area was significantly correlated with the insular surface asymmetry, both again expressed in LIs. (C) Gesture lateralization, measured at the whole hemisphere level, and insular surface asymmetries. Likewise, there was a significant correlation between gesture LIs and insular surface LIs. Black dots represent left-handers, whereas triangles represent right-handers.