Jacob Yates

Reveal Contact Info

Faculty

School of Optometry

Yates Lab

Current Research

Research in my lab is focused on understanding how the brain builds internal models of the world from sensory input—how it learns to infer structure and meaning from ever-changing visual signals. A key challenge is that the sensory input during natural vision is dynamic and shaped by eye movements. Yet the brain manages to extract stable, actionable representations.
We approach this problem from two directions. On the empirical side, we collaborate with experimental groups to analyze neural responses during natural vision, using modern spike train models to quantify how real neurons represent natural visual input. On the theoretical side, we develop brain-like generative models, incorporating spiking and eye movements into the latent structure, allowing us to ask how internal models can give rise to stable perception despite dynamic inputs. A central goal of the lab is to bridge normative theory and empirical neuroscience—to develop models of visual inference that are both computationally grounded and testable in real neural data.

Background

I began my career as an experimental neuroscientist studying the neural basis of perceptual decisions—how, on a single trial, we succeed or fail to see what’s there. I earned my PhD in Neuroscience in 2016 from the University of Texas at Austin, working with Alex Huk and Jonathan Pillow studying neurons in the parietal lobe in a well-established psychophysical paradigm based on motion discrimination. At the end of my PhD, I became convinced that the artificial paradigms used in much of sensory neuroscience—subjects fixating on a screen and making repetitive, constrained decisions—would not reveal the deeper principles of perception I was interested in. This was reinforced by a collaboration with Jon Matthis and Mary Hayhoe studying eye movements during natural locomotion. As a postdoc, I worked with Jude Mitchell, Greg DeAngelis, Michele Rucci and Dan Butts to build methods for enabling the study of vision during natural free viewing. I am now an Assistant Professor in the Herbert Wertheim School of Optometry and Vision Science and the Helen Wills Neuroscience Institute at UC Berkeley, where I continue to pursue the mechanisms of natural vision and the computational principles underlying perception.