|Ph.D Student||Bennet Rotem|
|Subject||Virtual Re-Embodied Cognition: Bodily-Cognitive|
Dynamics under Illusory Virtual Embodiments
|Department||Department of Education in Science and Technology||Supervisor||Professor Miriam Reiner|
Human cognition has developed “for action”, has co-evolved with the body and the world, and even abstract thought is in fact deeply grounded and situated in our bodily form, sensory experiences and physical interactions. Despite decades of embodied-cognition research in support of this view, not much is known on the generalization of it to alternative cognitive dynamics expected to emerge from the interaction with alternative body-world couplings, as may become highly relevant with the increased prevalence of virtual-reality (VR) embodied experiences. The goal of the current dissertation is therefore two-fold: 1. Study the instantaneous cognitive changes when re-embodied in an illusory bodily form, and 2. Study the continuous adaptation of motor-cognitive mechanisms to altered virtual sensorimotor dynamics. The research used three quantitative empirical experiments with different virtual re-embodiment environments, to tackle these goals. The first experiment used visuo-tactile synchronization to induce and maintain an illusory sense of ownership over a virtual hand of different lengths, and by this to actively explore the limits of the recently found near-body (Peripersonal) cognitive effects. The second and third experiments used an illusory virtual-hand that mirrored in VR the physical-hand’s motion, while operating a virtual device whose visuo-motor dynamics were altered to induce various sensorimotor adaptations. These two experiments used measurements of individual mental-imagery and motor-planning capacities, to study presumed changes in the extent of their correlations, representing the extent of the neural mental-motor mechanisms overlap.
The results of the three experiments show that virtual re-embodiment seems to demonstrate a generalization of the known body-related cognitive patterns, and most notably increases the apparent overlap of mental and motor shared components, as reflected in comparably-high correlations of individual mental-imagery and motor-planning capacities (up to r = 0.89, compared to r = 0.47 for the control). The results also reveal a surprising gender-specific difference in the connection of these mental and motor capacities, which may suggest additional explanations for the known gender-related gap in mental-rotation performance. In addition, the motor-cognitive re-coupling is shown to increase the cognitive improvement of low-performance subjects following motor training in the virtual re-embodied settings.
We interpret this seeming convergence of mutual mental-motor mechanisms to support the conjectured role of the cerebellar sensory-prediction forward-model as a shared mental-imagery and motor-control brain system. While most motor-cognition studies so far have consistently shown that mental-imagery and motor-control are functionally, temporally and physiologically “equivalent”, our results suggest a unique support for the stronger claim of a “shared” mental-motor component jointly serving both.
We conclude by analyzing some methodological limitations of our study, most notably a relatively-low number of subjects, and suggest follow-up research to validate and refine our results while expanding it to higher-cognition capacities. We expect the found increase in mental-motor correlations to possibly enable development of novel methodologies for motion-based cognitive training and assessment, which may also enhance future personalized active-learning environments. We suggest follow-up research to use the same paradigm to study additional motor-cognitive phenomena, such as the conjectured coupling of various motor and cognitive deficiencies, both in children and the elderly.