Cordelia Erickson-Davis Dissertation Defense

What it is to see: Artificial Vision as Constitutive Intra-Action

Visual prosthesis (VP) devices—devices that electrically stimulate the visual system with the goal of restoring vision to individuals who have lost it—have been heralded as being at the forefront of a scientific frontier that will allow for seamless integration between the biological and the mechanical. However, while hundreds of individuals have been implanted with VPs, few data are available on the perceptual experience associated with the devices. What does an individual see when the device is turned on for the first time, how does that change over time, and what does it mean for one of these devices to “work”? This dissertation is based on four years of interdisciplinary ethnographic research in which I tracked retinal implant devices from bench to bedside. What I found was that the reason we don’t know what artificial vision “is like” is because we don’t ask, and why we don’t ask is the story of clinical research and practice in the age of translational science and medicine, as well as the story of perception in the age of information and financialization. What has resulted is a biomedical landscape in which basic questions about the function of the devices have become practically unaskable, and what it means for the devices to “work” virtually irrelevant. When we do ask, however, we discover that there are large discrepancies between what was expected of these devices and what has been produced in terms of the perceptual experience of recipients, and that the “artificial vision” enabled by these devices is reported to be fundamentally different than what the individual remembered natural vision to be like. I argue that the failure of the devices stems in part from the problematic theory of vision that informed their design and implementation. I introduce an alternative theory of perception based on the work of empirical psychologist James J. Gibson that sees perception as the “direct” product of a constitutive intra-action of a perceiver-environment system (Erickson-Davis & Corwin 2020). I show how a theory of perception as constitutive interaction (PCI) is more congruous with recipient reports and enables us to see that “artificial vision,” or the perceptual experience associated with the VP, is a unique perceptual phenomenon. It is a finding that enables us to critically reflect on not only this class of devices, but also the dominant theories of perception that informed their design and implementation.

Public Zoom Session: https://stanford.zoom.us/j/97185753037?pwd=ZCtDUVpVbWUxQ2tkbVEwVXU4SU5vQT09