Matthew Kaufman

Matthew Kaufman

Matthew Kaufman

Matthew Kaufman is a postdoctoral fellow in Anne Churchland’s lab at Cold Spring Harbor Laboratory. He received a B.S. with honors and distinction in symbolic systems from Stanford University in 2005. He performed his graduate work in neuroscience with Krishna Shenoy and received his Ph.D. from Stanford University in 2011.

In his graduate work, Kaufman trained monkeys to perform a challenging reaching task while he recorded motor areas of their brains. By applying novel analyses to these data, he and collaborators developed a theory of motor control: that these areas use “extra” patterns to make their job of controlling muscles easier. This theory provides the brain a natural way to set up, initiate and control movements.

In his postdoc, he has continued to develop these approaches in the rodent. Using neural recordings in the rat, he found that critical information for decision making is distributed throughout a neural population, which can be read out in a straightforward way even though individual neurons contain a confusing mix of “extra” signals. Moving forward, he is exploiting sophisticated microscopy techniques to record from many neurons while identifying where they are and how they connect. His goal is to unravel the neural circuits underlying a cognitive behavior: to determine how sensory information is recombined in the brain in order to arrive at a decision and produce an action command. By focusing on how information flows through neural circuits, he hopes to understand how brain areas communicate with one another during decision making.



“Networks for decision making: signal routing in and through parietal cortex”

Decision-making is a key building block of cognition. Whether it’s as simple as an insect deciding to fly or walk, or as complex as a basketball player deciding whether to pass or shoot, the brain is constantly engaged in evaluating sensory information, making decisions, and executing actions. One brain region has repeatedly surfaced as critical for making decisions based on multiple kinds of sensory information: the posterior parietal cortex, or PPC. While much has been discovered about decision-making by studying the PPC, we do not yet understand how populations of many neurons work together to produce a decision. Working in mice, we plan to use sophisticated optical techniques to monitor the activity of many PPC neurons at once while the mouse engages in a simple decision-making task. The mouse will be trained to evaluate the rate of a series of clicks or flashes. If the rate of clicks or flashes is above a certain threshold, the mouse will tilt a tiny lever to, say, the right. If it’s below the threshold, the mouse will tilt the lever in the opposite direction. With such as setup, we can ask: How do PPC neurons integrate sensory information over time? How are the neurons distributed spatially throughout the PPC? Are they grouped together by the way that they respond? Or is it more complicated? Finally, we can investigate how PPC neurons influence the motor cortex, a brain area that helps control movement. Since decisions result in actions, investigating this pathway will help reveal how decision-related information in PPC is used by the rest of the brain. Collectively, the work described here will lead to fundamental insights into how the brain represents the sensory evidence gathered as decisions are made, and how it organizes its neural circuits into functional units in order to make effective decisions.