The building blocks of experience
Brian Isett has long been interested in the big, difficult questions about perception. His drive to understand individual experience led him to study philosophy and write poetry, but that was not enough. He wanted answers. Rather than throwing up his hands, he rolled up his sleeves and started to do research.
How do percepts form, and how do we understand the building blocks of experience? You can assume these questions are unanswerable or you can make some incremental step towards understanding something that seems impossible.
As an undergraduate at Cornell University, Isett worked in Carl Hopkins’ Lab and graduated with a B.S. in Biological Sciences with Distinction in Research. As a UC Berkeley Neuroscience PhD Program student, he worked in Dan Feldman’s Lab and just completed his thesis on “Simultaneous coding of microscopic and macroscopic features of touch in mouse somatosensory cortex.” He will submit two manuscripts about his PhD research soon.
Isett’s PhD research aims to determine how sensory information, specifically from touch, is processed in the brain. He is particularly interested in answering what one might think is a very simple question – how does the brain use touch to perceive object texture and shape?
Isett is able to rigorously study this question in mice, which use their whiskers for touch sensation. The touch sensitive neurons in mouse whiskers project to a region of the brain called the primary somatosensory cortex. The somatosensory cortex is organized into columns, with each column receiving information primarily from a single whisker. This is an extensively studied system, such that the neural connections are largely mapped out. However, the big questions about how an animal can use this brain structure to perceive an object, and make decisions about how to act based on that information, remain unanswered.
To tackle the big questions, Isett needs to have high speed tracking of the whiskers as they are touching different surfaces, while recording neuronal activity in a single cortical column, while the mice are making perceptual decisions. You can get some insight into the technical difficulties he has faced by browsing through his blog, which I found to be a surprisingly fun and interesting read considering that it includes posts such as “DIY Lickometer: Measuring Fast Licks From a Small Tongue.” His blog represents his interest in contributing to open science, especially in sharing methodology with other scientists.
Reflecting on his years as a PhD student, Isett recognizes that he has had to scale back his expectations about getting answers to the big questions. And that is ok. By maintaining a balanced lifestyle and continuing to explore ideas through poetry, his curiosity has remained intact. Thanks to the persistence of Isett and neuroscientists like him, the methodical and tireless probing will gradually make the brain yield its secrets.
The following Q&A with Brian has been edited for brevity.
How did you become interested in neuroscience?
In college I found myself taking a combination of biology, philosophy, and psychology classes. I started to triangulate ways to answer questions that were really interesting to me, like how do percepts form, and how do we understand the building blocks of experience. I was influenced by the philosophy classes I had taken, where they were posing these big questions. A really famous one is “What is it like to be a bat?” This is a famous philosophical essay by Thomas Nagel. He uses that question as a thought experiment to say it is not totally possible to understand some of the biggest questions about experience.
At the same time I was in Carl Hopkins’ lab studying how nocturnal electric fish communicate in dark water. I think on some basic level I thought, this is a lot like the bats! You can assume these questions are unanswerable or you can make some incremental step towards understanding something that seems impossible. That was what set me on the path of neuroscience.
Why did you decide to come to UC Berkeley for grad school?
During my interview I found that the graduate students at Berkeley were really smart but also really down to earth. They seemed like they had a balanced lifestyle in terms of researching really hard questions, but then also making time to get outdoors and maintain other passions that they were interested in. That was very appealing. It seemed like Berkeley had the kind of people that I would enjoy doing research with and that I could also make connections with outside of the lab.
In terms of the content of the work, I was interested in perception and systems level neuroscience. Berkeley is well known in disciplines relevant to studying those questions. In particular, there is important dialogue between computational neuroscientists and the systems and molecular neuroscientists here.
Did you learn coding and engineering as a PhD student or before?
I greatly expanded my coding ability while here and completely started my electrical engineering skill set. I wanted to create systems to test out new perceptual questions, and you can’t just buy this stuff off the shelf, you have to tinker around.
How will your research advance our knowledge of how the brain works, and why is that important?
Oh god, is it Monday or what? [laughs]
If we go back to the questions that got me interested in neuroscience – what are the building blocks of experience? – that is a generalized way of asking, how do individual neurons in the brain generate experience, and are there general patterns that can describe experience across individuals? This starts to address deep and important questions about the relationship between the structured circuitry in a person’s brain and individual experience.
And the really big questions have really big payoffs. For example, understanding how the brain generates experience is crucial for helping people with brain damage or altered brain circuitry. From strokes to schizophrenia to autism, we are going to have to understand how brain circuits generate experience in fine detail to make precise or individualized cures. Right now we only have brute force ways of addressing these problems. We give drugs that affect the whole brain, or we have bulk ways of stimulating the brain. I think when we want to create nuanced fixes to the most difficult neurological problems, those are going to require a very deep understanding of brain circuitry and activity patterns.
What is one interesting thing you discovered through your research?
Our experiments showed that cortical neurons represent multiple aspects of touch simultaneously and we were able to decode which kind of surface mice touched from the activity of these neurons. We even saw evidence that elementary tactile inputs could build basic representations of tactile shape. This could greatly expand the types of questions people are able to address using the whisker system, which would be very exciting.
What is the most difficult aspect of your research?
One difficult aspect is looking at a project, and honestly evaluating what is working and what is not working about it, and being able to reframe around the aspects that are working, even if it is not the full ambitious scope you set out for. Our original project was going to be on shape discrimination, but that changed.
The main reason for the change was that we really didn’t know enough about some of the more basic aspects of touch encoding to jump to a higher order feature like shape. It was difficult for me to admit, but we needed to answer more basic questions first: how does spatially precise information get from whisker-surface contact to the brain? We reframed our questions about shape to start from what was known and unknown about texture.
Despite all of that, the project worked out well. I think the compromises ended up making it a stronger paper in the end.
On your website you describe yourself as a “neuroscientist and poet.” How do these aspects of your personality interact?
Both are a way for me to try to understand and communicate something about experience. What excites me about the neuroscience I do is trying to understand how brain activity generates truths about the world. For me, poetry is an opportunity to create experiences that pass between people. It’s fun to try to condense a complicated world to a nugget of text that can then evoke a similar experience in someone else. It has been good for me to have the balance of staying excited about these bigger questions while I’m answering these things that often require a lot of painstaking detail. I am excited to say that I’ve been working on a book of poems called “Grid Poems Vol. I” which will be available soon.
by Georgeann Sack
Isett, B.R., Feasel, S.H., Lane, M.A., and Feldman, D.E. (2018). Slip-Based Coding of Local Shape and Texture in Mouse S1. Neuron 97, 418–433.e5.
Brian is currently a postdoc in the Gittis Lab at Carnegie Mellon University (started January 2018).