Investigation of Embodiment for Upper Limb Amputees

Overview

Today, prosthetic hands are numb. They provide no tactile or proprioceptive sensory information back to the user. The lack of sensory feedback has been shown to reduce the utility of a prosthesis by half. The prosthesis is seen as a tool, not as an incorporated part of the body schema. Only now are there chronically-implantable technologies which can provide physiologically appropriate sensory feedback to upper limb amputees to recreate tactile and proprioceptive percepts. These sensations are the building blocks to enable the embodiment of the device. Furthermore, newly developed outcome measures are now available which can detail the improved embodiment such neural interfaces can create. The investigator's mission is to enable the embodiment of artificial devices using peripheral nerve stimulation and thereby close the gap between the experience of our intact physiological systems and those using prosthetic remedies. This investigation of embodiment for upper limb amputees is organized into three main areas of work including 1) normative data collection, 2) device development, and 3) characterization of embodiment using peripheral nerve stimulation. The normative data collection will quantify the embodiment of conventional cosmetic, body-powered, and myoelectric prosthetic hand options using a modified Rubber Hand Illusion protocol (Specific Aim 1). This thrust will ask how does the amount of embodiment vary among conventional prosthetic hands as well as probe the relationship between agency and embodiment. The device development project entails the design of multi-modal sensors in order to study full-hand embodiment (Specific Aim 2). The ability to measure and then elicit sensation on the passive surfaces of the hand (palm, ulnar border, and dorsal surface) has never been explored. Here, a multi-modal sensor which can detect proximity, contact, and force will be integrated into a commercially available prosthetic hand in order to provide detailed measurements across the palm, ulnar border, and dorsal surfaces in order to study embodiment in more depth. Finally, the characterization of embodiment using peripheral nerve stimulation will take place over a multiple subject factorial experiment which quantifies the effects of quantity and spatial parameters of the peripheral nerve stimulation on the embodiment of prosthetic hands (Specific Aim 3). This study asks what somatosensory percepts from the hand are most critical for embodiment by varying the parameters of the peripheral nerve stimulation (quantity and spatiality) and measuring the level of embodiment in each case.

Leveraging advancements in neural interfaces, biomechatronic devices, and myoelectric control algorithms, my research mission is to enable the embodiment of artificial devices by providing physiologically appropriate somatosensory feedback. Specific Aim 1: How does the amount of embodiment vary among conventional prosthetic hands?. Data from Specific Aim 1 will be used to test the following hypotheses: H1a. The body-powered prosthetic devices are embodied more than passive and myoelectric prosthetic devices. H1b. Passive cosmetic devices are embodied less than actuated cosmetic devices (agency). H1c. Body-powered terminal devices are embodied less than myoelectric terminal devices (agency). Specific Aim 2: Design of Multi-Modal Sensors for Full Hand Sensation (No human subject experiments.) Specific Aim 3: What somatosensory percepts from the hand are most critical for embodiment? Data from Specific Aim 3 will be used to test the following hypotheses: H3a. The maximum number of channels elicits more embodiment than the minimum number. H3b. The sensory feedback from passive spatial locations of the hand increases the embodiment compared to sensory feedback just from the grasping spatial locations.