Handedness-dependent functional organizational patterns within the bilateral vestibular cortical network revealed by fMRI connectivity based parcellation

Highlights

• Non-binary masked ICA addresses the non-binary (multisensory) nature of the vestibular cortical network.

• Handedness-dependent multiple symmetric integrative hub areas surround an asymmetric (lateralized, dominant) multisensory core region.

• The inferior insula is a target station for incoming lateralized vestibular information.

• Lateralized hemispheric function such as higher cortical vestibular functions might reflect most recent evolutionary developments.

Abstract

Current evidence points towards a vestibular cortex that involves a multisensory bilateral temporo-parietal-insular network with a handedness-dependent hemispheric lateralization. This study aimed to identify handedness-dependent organizational patterns of (lateralized and non-lateralized) functional subunits within the human vestibular cortex areas. 60 healthy volunteers (30 left-handed and 30 right-handed) were examined on a 3T MR scanner using resting state functional MRI (fMRI). The data was analyzed in four major steps using a functional connectivity based parcellation (fCBP) approach: (1) independent component analysis (ICA) on a whole brain level to identify different resting state networks (RSN); (2) creation of a vestibular informed mask from four whole brain ICs that included reference coordinates of the vestibular network extracted from meta-analyses of vestibular neuroimaging experiments; (3) Re-ICA confined to the vestibular informed mask; (4) cross-correlation of the activated voxels within the vestibular subunits (parcels) to each other (P-to-P) and to the whole-brain RSN (P-to-RSN).

This approach disclosed handedness-dependency, inter-hemispheric symmetry, the scale of connectedness to major whole brain RSN and the grade of spatial overlap of voxels within parcels (common/unique) as meaningful discriminatory organizational categories within the vestibular cortex areas. This network consists of multiple inter-hemisphere symmetric (not lateralized), well-connected (many RSN-assignments) multisensory areas (or hubs; e.g., superior temporal gyrus, temporo-parietal intersection) organized around an asymmetric (lateralized, “dominant”) and functionally more specialized (few RSN-assignments) core region in the parieto-insular cortex. The latter is in the middle, posterior and inferior insula. In conclusion, the bilateral cortical vestibular network contains not only a handedness-dependent lateralized central region concentrated in the right hemisphere in right-handers and left hemisphere in left-handers, but also surrounding inter-hemisphere symmetric multisensory vestibular areas that seem to be functionally influenced by their neighboring sensory systems (e.g., temporo-parietal intersection by the visual system). One may speculate that the development of an asymmetrical organized vestibular subsystem reflects a more recent phylogenetic evolution of various multisensory vestibular functions. The right hemispheric dominance of spatial orientation and its disorders, spatial neglect and pusher syndrome, may serve as examples.

Introduction

The vestibular cortex differs from other sensory cortices. Lacking a primary cortex in the narrower sense, the vestibular system relies on a network involving multiple multisensory vestibular cortical areas in both hemispheres organized around a core region in the posterior insula and retroinsular region, called the parieto-insular vestibular cortex (PIVC) in monkeys (Guldin and Grüsser, 1998). This network seems to be very similar in humans with the posterior insula, retro-insular region, and operculum 2 (OP2) representing the main components of said core region (Guldin and Grüsser, 1998; Brandt and Dieterich, 1999; Zu Eulenburg et al., 2012; Lopez et al., 2012). Another important feature of the vestibular system is a hemispheric asymmetry, namely the dominance of the ipsilateral hemisphere in relation to handedness (Dieterich et al., 2003). Further, there is a preference of the ipsilateral ascending projections from the stimulated ear (Bense et al., 2003; Dieterich et al., 2017), and a reciprocal inhibitory visual-vestibular interaction disclosed by functional imaging (Brandt et al., 1998; Brandt et al., 2002, 2003; Wenzel et al., 1996).

The vestibular cortical network relies on vestibular, visual, and somatosensory inputs in monkeys (Guldin et al., 1992; Chen et al., 2011a) and humans (Baumgärtner et al., 2010; Bucher et al., 1998; Dieterich et al., 1998; Emri et al., 2003; Fasold et al., 2002; Konen and Kastner, 2008). These multisensory inputs converge at multiple levels from the brain stem to the cortex: reflexive sensorimotor control of eyes, head, and body at the brain stem/cerebellar level; perception of self-motion and control of voluntary movement and balance at the cortical/subcortical level; and higher vestibular cognitive functions (e.g., spatial memory and navigation) at the cortical level (Dieterich and Brandt, 2015). This varying multisensory information is centrally computed, both bilaterally (Dieterich and Brandt, 2015) and hierarchically (Chen et al., 2010). It converges (Chen et al., 2011a; Carriot et al., 2013) in multiple spatially distributed (Brandt and Dieterich, 1999; Lopez and Blanke, 2011) and spatially tuned (Chen et al., 2011b) cortical areas. Within these areas both multimodal neurons as well as a mixture of multiple unimodal neuronal populations have been found in monkeys (Guldin et al., 1992; Yang et al., 2011; Dokka et al., 2015). Consequently, a binary understanding that allocates one specific label to one specific cortical area will most likely not do justice to the vestibular cortical network, especially at the scale being assessed by functional MRI (fMRI) (Logothetis, 2012, 2002; Tolias et al., 2005).

Resting state fMRI provides proxies of dynamic neuronal interactions that are thought to reflect mixtures of various cognitive processes and physiological factors (Buckner et al., 2008; Fox and Raichle, 2007; Smith et al., 2009). Functional connectivity based parcellation (fCBP) approaches oftentimes use independent component analysis (ICA) on resting state fMRI to divide a region of interest (ROI) into distinct subregions (Eickhoff et al., 2015; Kim et al., 2010). An fCBP approach that is restricted to a specific mask is then able to identify even more discrete functional subunits within this region (Kim et al., 2013; Mars et al., 2012; Smith et al., 2015). Most fCBP approaches then use a binary allocation of one label/parcel per voxel (Craddock et al., 2012; Neubert et al., 2014; Thirion et al., 2014). However, ICA naturally allows a multivariate (non-binary) separation of linear mixtures of signal sources that temporally correlate and spatially overlap (Beckmann and Smith, 2004). This non-binary methodical approach might be able to reflect multiple signals at the same spatial location, e.g., in multiple populations of neurons or a single multisensory population and might therefore be insightful when dealing with the vestibular system.

This study aimed to identify handedness-dependent organizational patterns of functional subunits within the human vestibular cortex areas whilst addressing its multisensory (non-binary) nature. To that end, 60 healthy volunteers (30 left-handed and 30 right-handed) were analyzed using masked ICA approaches resulting in binary and non-binary results. This mask was data-driven (composed of whole brain independent components) and specific to the vestibular cortical system as the independent components (ICs) used had to include vestibular reference coordinates derived from two meta-analyses of vestibular neuroimaging experiments pinpointing the vestibular cortex (Lopez et al., 2012; Zu Eulenburg et al., 2012).