SF Presidential Lecture: How Occam’s Razor Guides Human and Machine Decision-Making

Wednesday Oct 19, 2022, 6:00 PM → 7:00 PM America/New_York

Gerald D. Fischbach Auditorium/2-GDFA (160 5th Avenue)

Registration link: https://www.eventbrite.com/e/how-occams-razor-guides-human-and-machine-decision-making-registration-405283091637

Deciding something as potentially complicated as what to do next or as deceptively simple as where to look next requires our brains to deliberate; that is, to move beyond the rigidness and immediacy of sensory-motor reflexes and instead take time to process and weigh evidence in a flexible manner until arriving at a categorical judgment that guides behavior. Our understanding of this deliberation process, which represents a major building block of cognition, has benefited greatly from mathematically rigorous theories from some unexpected places.

In this lecture, Joshua Gold will describe two theoretical frameworks that support ongoing studies of deliberative decision-making in the brain, focusing on their historical origins. The first describes quantitatively the process by which uncertain evidence can be accumulated over time to balance the competing needs of maximizing decision accuracy while minimizing decision time. This framework is built on mathematical advances that Alan Turing and colleagues developed to decode messages sent via the Enigma machine during World War II. The second describes how biases can emerge in this information-accumulation process that can be helpful when considering options that differ in form and scope. This framework is a formalization of Occam’s razor, which states that all else being equal, simple solutions are better — an idea directly relevant to how biological and artificial brains can make effective decisions.

Speaker Bio:

Gold is a neuroscientist whose training included studying mechanisms of learning as an undergraduate and graduate student and mechanisms of decision-making as a post-doctoral fellow. He currently is a professor of neuroscience at the University of Pennsylvania, where he combines his training experiences into studies of how learning and decision-making interact in the primate brain. He is also chair of the Neuroscience Graduate Group, co-director of the Computational Neuroscience Initiative and father of three teenage children, all of which reflect his deep commitment to mentoring, education and patience.

Inquiries: lectures@simonsfoundation.org