Multisensory self-motion processing in humans

Humans obtain and process sensory information from various modalities to ensure successful navigation through the environment. While visual, vestibular, and auditory self-motion perception have been extensively investigated, studies on tac-tile self-motion perception are comparably rare. In my thesi...

Full description

Saved in:
Bibliographic Details
Main Author: Rosenblum, Elisabeth-Maria
Contributors: Bremmer, Frank (Prof. Dr.) (Thesis advisor)
Format: Doctoral Thesis
Published: Philipps-Universität Marburg 2023
Online Access:PDF Full Text
Tags: Add Tag
No Tags, Be the first to tag this record!
Summary:Humans obtain and process sensory information from various modalities to ensure successful navigation through the environment. While visual, vestibular, and auditory self-motion perception have been extensively investigated, studies on tac-tile self-motion perception are comparably rare. In my thesis, I have investigated tactile self-motion perception and its interaction with the visual modality. In one of two behavioral studies, I analyzed the influence of a tactile heading stimulus intro-duced as a distractor on visual heading perception. In the second behavioral study, I analyzed visuo-tactile perception of self-motion direction (heading). In both studies, visual self-motion was simulated as forward motion over a 2D ground plane. Tactile self-motion was simulated by airflow towards the subjects’ forehead, mimicking the experience of travel wind, e.g., during a bike ride. In the analysis of the subjects’ perceptual reports, I focused on possible visuo-tactile interactions and applied dif-ferent models to describe the integration of visuo-tactile heading stimuli. Lastly, in a functional magnetic resonance imaging study (fMRI), I investigated neural correlates of visual and tactile perception of traveled distance (path integration) and its modu-lation by prediction and cognitive task demands. In my first behavioral study, subjects indicated perceived heading from uni-modal visual (optic flow), unimodal tactile (tactile flow) or from a combination of stimuli from both modalities, simulating either congruent or incongruent heading (bimodal condition). In the bimodal condition, the subjects’ task was to indicate visually perceived heading. Hence, here tactile stimuli were behaviorally irrelevant. In bimodal trials, I found a significant interaction of stimuli from both modalities. Visually perceived heading was biased towards tactile heading direction for an offset of up to 10° between both heading directions. The relative weighting of stimuli from both modalities in the visuo-tactile in-teraction were examined in my second behavioral study. Subjects indicated per-ceived heading from unimodal visual, unimodal tactile and bimodal trials. Here, in bimodal trials, stimuli form both modalities were presented as behaviorally rele-vant. By varying eye- relative to head position during stimulus presentation, possi-ble influences of different reference frames of the visual and tactile modality were investigated. In different sensory modalities, incoming information is encoded rela-tive to the reference system of the receiving sensory organ (e.g., relative to the reti-na in vision or relative to the skin in somatosensation). In unimodal tactile trials, heading perception was shifted towards eye-position. In bimodal trials, varying head- and eye-position had no significant effect on perceived heading: subjects indicated perceived heading based on both, the vis-ual and tactile stimulus, independently of the behavioral relevance of the tactile stimulus. In sum, results of both studies suggest that the tactile modality plays a greater role in self-motion perception than previously thought. Besides the perception of travel direction (heading), information about trav-eled speed and duration are integrated to achieve a measure of the distance trav-eled (path integration). One previous behavioral study has shown that tactile flow can be used for the reproduction of travel distance (Churan et al., 2017). However, studies on neural correlates of tactile distance encoding in humans are lacking en-tirely. In my third study, subjects solved two path integration tasks from unimodal visual and unimodal tactile self-motion stimuli. Brain activity was measured by means of functional magnetic resonance imaging (fMRI). Both tasks varied in the engagement of cognitive task demands. In the first task, subjects replicated (Active trial) a previously observed traveled distance (Passive trial) (= Reproduction task). In the second task, subjects traveled a self-chosen distance (Active trial) which was then recorded and played back to them (Passive trial) (= Self task). The predictive coding theory postulates an internal model which creates predictions about sensory outcomes-based mismatches between predictions and sensory input which enables the system to sharpen future predictions (Teufel et al., 2018). Recent studies sug-gested a synergistical interaction between prediction and cognitive demands, there-by reversing the attenuating effect of prediction. In my study, this hypothesis was tested by manipulating cognitive demands between both tasks. For both tasks, Ac-tive trials compared to Passive trials showed BOLD enhancement of early sensory cortices and suppression of higher order areas (e.g., the intraparietal lobule (IPL)). For both modalities, enhancement of early sensory areas might facilitate task solv-ing processes at hand, thereby reversing the hypothesized attenuating effect of pre-diction. Suppression of the IPL indicates this area as an amodal comparator of pre-dictions and incoming self-motion signals. In conclusion, I was able to show that tactile self-motion information, i.e., tactile flow, provides significant information for the processing of two key features of self-motion perception: Heading and path integration. Neural correlates of tactile path-integration were investigated by means of fMRI, showing similarities between visual and tactile path integration on early processing stages as well as shared neu-ral substrates in higher order areas located in the IPL. Future studies should further investigate the perception of different self-motion parameters in the tactile modali-ty to extend the understanding of this less researched – but important – modality.