Research Interests

As we move through the world around us our senses collect huge amounts of information about the state of our environment. It is the brain's job to process this information quickly and accurately in order to make decisions, implement coordinated motor responses, and predict future events. How does the brain do this? I am interested in studying sensory processing systems in order to better understand the fundamental computations that neurons perform during the processing of external stimuli. I am also interested in how internal states such as motivation and attention influence these computations.

More broadly, I am interested in how the fundamental principles underlying these computational strategies employed by the brain can be translated into artificial information processing systems. Over the last decade it has become clear that powerful new machine learning approaches like deep neural networks have a superficial but strong connection to biological information processing systems, like the visual system. Further developing these parallels will inevitably result in significant scientific and technological advances in the years to come.


Decoding in the presence of shared neural variability

It has long been known that variability in the spike counts of individual neurons, in response to identical stimuli, is correlated across populations of neurons. Recent work has shown that the impact of these correlations on the total amount of information contained in a population is highly dependent on the structure of these correlations.

This project aims to develop an analytical framework to understand the effects of correlated variability that is introduced by a small number of latent variables. Specifically, we are interested in how this variability affects decoding information that is contained in the population response. The conclusions drawn from this framework can then be tested in large neural populations to reach a better understanding of how the brain might ignore or utilize these correlations in processing sensory stimuli.

Latent variable models for analyzing large population recordings

The study of sensory cortical neurons has traditionally been carried out by recording from one or a few neurons. Recent advances in 2-photon microscopy and multi-electrode arrays now allow us to simultaneously record hundreds or even thousands of neurons in cortex. Since stimulus representation and processing is manifested in spatiotemporal patterns of activity in-vivo, these new experimental approaches have the power to revolutionize our understanding of cortical function.

Currently, however, there are no standard techniques for analyzing the vast amounts of data these experiments generate. The aim of this project is to develop a unified statistical framework that combines stimulus processing models and latent variable models of neural activity. I am interested in using this model to address questions about the nature of the interactions between stimulus processing and ongoing cortical network dynamics in large populations of sensory neurons.

paper | poster | code