Meta-analytical definition and functional connectivity of the human vestibular cortex
Introduction
Human equilibrium and our physical ability to navigate strongly depend on the processing of vestibular information. Although Aristotle did not include balance in his list of essential senses, we probably would not have evolved into upright stance and bipedal locomotion without permanent central nervous computation and representation of signals delivered by our peripheral vestibular organs. All other sensory systems have long been localized to specific areas within the human brain. The exact whereabouts of a well-defined vestibular cortex in man though is still very much in contention. This is in contrast to the undisputed human vestibular anatomy of the brain stem and its multisensory interconnections. During the course of time, four out of five brain lobes have been considered to “harbour” the cortical vestibular system (the exception being the occipital lobe) (Duque-Parra, 2004), thus showing the level of uncertainty as well as the possibly distributed nature of the vestibular network.
Studies employing neuron recordings and tracer injections in rats, squirrels, java monkeys and macaques have focused on several separate structures in the frontal, temporal and parietal cortex that undoubtedly receive vestibular input: the area 2v at the tip of the intraparietal sulcus, the area 3aV (a vestibular region within area 3a representing neck and trunk) in the central sulcus and an area called the parieto-insular vestibular cortex (PIVC) located posterior to the dorsal end of the insula. Area 2v in the parietal cortex was the first definite cortical vestibular area found in primates. It was initially thought to represent the vestibular area anterior to the suprasylvian sulcus (ASSS) found in the cat (Fredrickson et al., 1966). The ASSS had been the first cortical vestibular projection demonstrated in mammals. As early as 1973, Pandya and Sanides had already stressed in their findings that there is a distinctive cytoarchitectonic analogy between the retroinsular parietal cortex in primates and ASSS (Pandya and Sanides, 1973). The PIVC as the correlate for this retroinsular region was then discovered by Grüsser and his group in the Java monkey and later also confirmed in the squirrel and marmoset monkey (Grusser et al., 1990). It seemed more likely to represent ASSS in non-human primates than area 2v. Among the regions that have also been shown to receive vestibular information in the monkey are the ventral intraparietal area (VIP), area 7 in caudal inferior parietal lobe, the primary motor cortex (area 4) and the premotor cortex (area 6). A distinct vestibular cingulate region has been termed though it seems to only receive preprocessed vestibular information from the aforementioned areas 3aV and PIVC (for a more detailed comparative anatomy of the cortical vestibular representations in animals and humans see Lopez and Blanke, 2011) (Lopez and Blanke, 2011).
The aim of the present study was to pinpoint and segregate vestibular representations in the human cortex through means of a quantitative meta-analytical approach. We wanted to follow up on the quest for a unique vestibular cortex (Guldin and Grüsser, 1998). The use of a wide variety of stimuli (warm and cold caloric irrigation, galvanic vestibular stimulation, short tone bursts as an otolith impulse) as well as multiple imaging modalities (H2O2- and FDG-PET, electrical source and dipole reconstruction in EEG, and fMRI) in the last 30 years of human vestibular research has led to a variegated and still inhomogeneous level of knowledge about the cortical vestibular network. With our analysis, we hoped to condense and streamline the heterogeneous literature in this aspect of vestibular research. Based on previous hypotheses on laterality and interhemispheric dominance within the vestibular system, we anticipated to find lateralized cortical areas responding to a strictly unilateral side of stimulation as well as central nodes activated by all vestibular stimuli alike. Furthermore, it was our aim to look for unique patterns in vestibular activations with respect to the different subtypes of vestibular excitation (galvanic, caloric or otolith). Finally, we intended to gain knowledge about a cortical vestibular network by means of integrating our meta-analysis findings with a separate subsequent functional connectivity approach.
Section snippets
Study selection criteria
Functional neuroimaging studies were retrieved via searches in Pubmed, ISI Web of Knowledge and Scopus databases as well as identified by reference tracing and through reviews. Experiments reported in these papers that corresponded to the contrast of a vestibular stimulation against a resting baseline or a somatosensory control condition were included in the meta-analysis if they fulfilled the following criteria (cf. Table 1):
Analyses must be computed across the whole brain and not restricted
Meta-analysis of all vestibular activations
Activation likelihood (ALE) meta-analysis over the foci reported in all 28 suitable neuroimaging experiments (cf. Table 1) yielded significant convergence bilaterally in the peri-sylvian cortex (Fig. 1a, all overlays done with MRIcroGL by Chris Rorden). The extended clusters in both hemispheres included the posterior parietal operculum, the secondary somatosensory cortex, the inferior parietal cortex as well as the middle and posterior insula (Fig. 1b). Further significant convergence across
Discussion
Our analysis showed extended clusters in the temporo-parietal cortex, lateral and medial premotor cortex as well as parts of the insula when looking at the vestibular stimulation studies as a whole. They seem to resemble a network similar to one described by EEG recordings after intraoperative stimulation of the vestibular nerve in patients (de Waele et al., 2001). To find the common denominator for the vastly different kinds of vestibular incitements, we then differentiated the studies into
Conclusion
In conclusion, we want to follow-up on the question most prominently posed by Guldin and Grüsser at the end of the 20th century: “Is there a vestibular cortex?” Our results strongly support the existence of such a distinct and unique vestibular cortex in humans with its possible core region in the area OP 2 of the parietal operculum as the homologue to monkey PIVC (Guldin and Grüsser, 1998). Independent separate testing of functional connectivity revealed all other retrieved vestibular regions
Acknowledgements
We are grateful to Sara Duke for critically reading and editing the manuscript.
References (52)
- et al.
Unified segmentation
NeuroImage
(2005) - et al.
fMRI of thermal pain: effects of stimulus laterality and attention
NeuroImage
(2002) - et al.
ALE meta-analysis of action observation and imitation in the human brain
NeuroImage
(2010) - et al.
Assignment of functional activations to probabilistic cytoarchitectonic areas revisited
NeuroImage
(2007) - et al.
Co-activation patterns distinguish cortical modules, their connectivity and functional differentiation
NeuroImage
(2011) - et al.
Human somatosensory area 2: observer-independent cytoarchitectonic mapping, interindividual variability, and population map
NeuroImage
(2001) - et al.
Is there a vestibular cortex?
Trends Neurosci.
(1998) - et al.
Multisensory mechanisms in temporo-parietal cortex support self-location and first-person perspective
Neuron
(2011) - et al.
On the hemispheric specialization for categorical and coordinate spatial relations: a review of the current evidence
Neuropsychologia
(2003) - et al.
The thalamocortical vestibular system in animals and humans
Brain Res. Rev.
(2011)