Neural mechanisms of active vision in the fovea - Neural mechanisms of active vision in the fovea
In many ways human vision is like a camera, with a lens that forms an image on a spatially
arranged sensor (the retina). However, it is unlike a camera because the sensor has uneven
sampling and is constantly moving with the eyes. Recent behavioral and theoretical work suggest
these eye movements serve a faciliatory role in high acuity vision – where the eye movements
are part of the computations and enhance spatial resolution. However, the neurophysiological
mechanisms to support this facilitation remain unknown. More broadly, little is known about the
neural mechanisms that integrate across the retinal motion generated by eye movements,
especially in the central visual field (the fovea). This is particularly important because over 8
million Americans suffer from central vision loss due to retinal disorders. Even if the retinal signals
could be repaired, it is imperative to understand how the brain reads out foveal signals to ensure
recovery of high-acuity visual processing, and fixational eye movements are a part of that process.
The proposed career development plan aims to address these questions by measuring visual
processing in the foveal representation of primary visual cortex (V1) during natural visual
behavior. This proposal uses custom high-resolution eye-tracking, a novel visual foraging
paradigm, largescale neurophysiology, and state-of-the-art machine learning to make these
measurements possible. The proposed research will not only generate fundamental
understanding of how eye-movements facilitate visual processing, but also will integrate the
experimental and theoretical tools required to support neurophysiological studies of active visual
processing without a loss of rigor or detail. The candidate has extensive expertise in awake-
behaving neurophysiology and computational modeling and the training plan is designed to
support his further training in statistical modeling, high-resolution eye-tracking, and modern
machine-learning techniques for analyzing neural population data. The primary mentor, Dr. Daniel
Butts, is a world expert in statistical models of neural activity during active vision; Co-mentor, Dr.
Michele Rucci, is a world leader in high-resolution eye tracking and theoretical approaches to
active vision; and Co-mentor, Dr. Jude Mitchell, is a pioneer in establishing the marmoset model
of visual neuroscience and an expert in neurophysiology of visual attention. Together, they will
provide the guidance to establish the candidate’s transition to a successful independent research
career.