Abstract
Several works have revealed a robust visual–vestibular interaction at the cortical level in humans. The same has been shown at an infratentorial level in other primates. Most stimuli applied in the context of ocular motor neuroimaging research are fixed and therefore highly artificial in nature for methodological reasons. Advances in imaging techniques now allow for robust brainstem fMRI, thus opening up a whole new terrain for investigation. The aim of this study was to investigate the neural correlate for ocular motor adaption control (AC) with changing stimulus parameters (speed/direction) and the functional connectivity (FC) of the localised governing hubs at the level of the brainstem and the cerebellum. The effects of constant and linear accelerating small-field optokinetic (OKN) stimuli in the horizontal and vertical direction were studied in 20 volunteers lying supine with their ears plugged in a clinical 3T scanner (Siemens Magnetom Trio, Erlangen, Germany). The protocol included two sessions, each consisting of a T2*-weighted EPI sequence (2mm isotropic resolution, TA 2.93s) in alternating blocks of visual fixation and during OKN stimulation (left/right, up/down). Horizontal (vertical) OKN speeds were either fixed at 30°/s (20°/s) or accelerated during stimulus blocks from 15° to 45°/s (10°/s–30°/s). Resting-state data was acquired prior to task fMRI for 5min (3mm isotropic resolution, 37 slices, TA 2.67s, whole-brain coverage). Data was analysed with SPM8 and DPARSFA after standard preprocessing. T-contrasts were calculated with respect to the rest condition and were considered significant at p . The task revealed a cerebellar network including the dorsal ocular motor vermis (OMV), the fastigial ocular motor region, the posterior inferior vermis, the flocculus and paraflocculus, the culmen (lobule VI) and the declive. In the brainstem, we were able to significantly activate the ocular motor and abducens nuclei, the paramedian pontine reticular formation (PPRF), the dorsolateral pontine nucleus (DLPN), the nucleus reticularis tegmenti pontis (NRTP), the prepositus nucleus (NPH), riMLF and the interstitial nucleus of Cajal (INC). AC was localised to the OMV and pontine structures (NTRB and PPRF). Activation clusters throughout all OKN tasks were significantly larger with higher peak maxima in the left cerebellar hemisphere. We found a significant deactivation of the ventral uvula and the cerebellar tonsils (lobule VIII) during constant more so than during accelerating OKN stimulation. FC maps with the dorsal ocular motor vermis as a seed region gave a significant connectivity with all other regions found in the task-based part of our experiment as well as the dorsal visual stream and all cortical eyefields. Our study gives in vivo evidence for the basic visual–vestibular interactions at the level of the infratentorial networks. Ocular motor areas of the human brainstem could be mapped at an unprecedented detail. AC for the OKN task itself independent of direction was found to be localised mainly in the OMV. Velocity-fixed more so than accelerating OKN stimuli lead to a significant deactivation of a cerebellar core vestibular area known to govern the velocity storage mechanism, the ventral uvula. Extending the cortical concept of a right-hemispheric predominance for visual-spatial processing we found a complementary left-sided dominance for OKN responses in the cerebellum (Dieterich et al., 2003). Taken together, these findings should open up another chapter in our understanding of visual–vestibular interactions in the human brain.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call