Braden Purcell is a postdoctoral fellow at the Center for Neural Science at New York University. He received his B.A. in psychology from Miami University in 2007 and a Ph.D. in psychological science from Vanderbilt University in 2013. For his Ph.D. work, he studied neural mechanisms of visual attention and perceptual decision making with Jeffrey Schall and Thomas Palmeri. At New York University, he is currently studying neural mechanisms of adaptive behavior in the lab of Dr. Roozbeh Kiani. This work uses human psychophysics, computational modeling and neurophysiological recordings in behaving monkeys to understand how the brain can update decision strategies in a dynamic environment.
“The Role of Perceptual Certainty in Adaptive Decision Making”
Our confidence in previous experiences can strongly drive our future decisions. For example, imagine that you are driving a car and running late, but then the car begins to make a strange noise. Should you pull over and risk being late, or should you just keep driving? In making this decision, you would want to consider how confident you are that the noise indicates a problem. If the engine is clearly close to failure, then you should pull over immediately. On the other hand, if the noise is more ambiguous, then you might want to continue driving and avoid any delay. When making this choice, your brain must first determine your confidence about the perceived noise and then use that confidence to decide whether to stop or keep driving. In our laboratory, we study how the brain represents confidence and uses it to support decision-making and learning. Using simple visual stimuli, we have trained monkeys to perform a new task in which they make decisions based on their confidence about past choices. While the monkeys perform this task, we use newly developed techniques to record the neural activity of many brain areas at once. We build computer models of these brain areas to understand how their activity relates to the monkey’s confidence, decision-making, and learning. This work will provide insight not only into how humans make decisions based on confidence in general, but also will inform our approach to instances in which decisions fail to take confidence into account, which may occur in disorder such as autism, schizophrenia, or drug abuse.