Neural populations can multitask to support decision-making
November 10, 2014
A new paper is out from my lab today in Nature Neuroscience. In this paper, we set out to understand how a single part of the brain can be used to support many behaviors. The posterior parietal cortex, for instance, has been implicated in decision-making, value judgments, attention and action selection. We wondered how that could be: are there categories of specialized neurons that are specialized for each behavior? Or, alternatively, does a single population of neurons multitask to support lots of behaviors? We found the latter possibility to be true. We recorded neurons in rat posterior parietal cortex while rats were making decisions about lights and sounds. We found that neurons could be strongly modulated by the animal’s choice, the modality of the stimulus, or, very often, both of those things. This multitasking did not provide a problem for decoding: a linear combination of responses could easily estimate choice and modality well.
We hope that our observations will change the way people think about how neurons support diverse behaviors because they challenge the prevailing view that neurons are specialized. Horace Barlow (the grandfather of computational neuroscience), argued that neurons in the frog’s retina were specialized for detecting particular kinds of motion. This is likely true in early visual areas, but in higher cortical areas, things are very different. Our observations about multitasking neurons point to a new way of encoding information that, we argue, confers flexibly in how the neurons are used, and could allow their responses to be flexibly combined to support many behaviors. The picture below shows me with co-first authors David Raposo and Matt Kaufman. 
New ideas for making sense of big data
October 18, 2013
Okay, so suppose you’ve just measured responses in hundreds of neurons, over time, during a complex behavioral task. Now what?? My lab members and I attended a conference at Columbia this week focussed on this issue. The conference, organized by Mark Churchland, Larry Abbott, John Cunningham and Liam Paninski was sponsored by Sandy Grossman and is a timely topic: advances in recording and imaging technology have made large neural datasets the norm and understanding how to analyze such datasets is nontrivial.
The talks included one from our lab, in which I described our recent ideas about the posterior parietal cortex and its response during a high dimensional decision task. Our work dovetailed with several others at the meeting: For example, Chris Machens spoke about his demixing principal components analysis, an analysis we have been using in our data. Chris, along with his student Wieland Brendel, 
developed this analysis to ask whether parameters that are mixed at the level of single neurons might be orthogonal at the level of the population. Observing an orthogonal representation in the population is important because it suggests that task parameters are represented in a way that could be trivially decoded by a downstream area.
In another talk, Jonathan Pillow described recent work from his lab on Bayesian nonparametric models for spike patterns in large datasets. The basic idea in “Bayesian nonparametrics” is to define models whose complexity grows gracefully with the amount of data available. Jonathan described an approach for modeling binary spike patterns using a Dirichlet process, which marries the parsimony of a simple parametric model (e.g., each neuron fires independently with probability “p”) and a “histogram” model that describes arbitrarily complex distributions over binary spike patterns. These models, which Jonathan’s group calls “universal binary models”, strike a happy medium between overly complex models and those that are so simple they fail to capture key features of spike data.
Churchland lab 