How does the brain perceive? Does it use the information coming in through the various senses, such as our eyes and ears, and build up a perception of the world outside from the bottom up? Or is it doing something quite different? New thinking in neuroscience suggests that the brain builds models of what’s out there and uses these models to interpret the incoming sensory data—an idea that goes back to Plato, Kant and most recently to German physicist and physician Hermann von Helmholtz. Perception—of any sort, including perception of our body—is inherently an inference on the causes of the information impinging on our senses. According to this predictive processing hypothesis, the brain is always making informed predictions, and it’s these predictions that we perceive at all times.

A panel of scientists and writers will discuss the predictive brain from multiple perspectives. Celeste Kidd, professor of psychology, will discuss the predictive brain in the context of a developing child’s brain. Computer scientist Christos Papadimitriou will share his thoughts on the intersection of these ideas with computer science. Neuroscientist Bruno Olshausen will shed light on visual perception and the predictive brain. Michael Pollan, author of How To Change Your Mind, will speak about his experiences with psychedelics and how psychedelic hallucinations tie in with current ideas of perception as prediction. Anil Ananthaswamy, author of The Man Who Wasn’t There, will moderate the discussion and share ideas about how neuropsychological conditions such as schizophrenia are being understood from the point of view of the predictive brain.

Formal registration for "The Predictive Brain" is now closed. Overflow seating of 70 seats will be available in an adjacent room, and seats may free up in the lecture hall. We encourage you to arrive for refreshments at 5:30 pm and check in at the registration table, where you'll receive a spot in the waiting list for a seat.

Questions? Contact Christine Wang at +1 510.664.5459 or chris24 [at] berkeley.edu.

Theoretically Speaking is a lecture series highlighting exciting advances in theoretical computer science for a broad general audience. Events are held at the David Brower Center in Downtown Berkeley, and are free and open to the public. No special background is assumed.

Light refreshments will be served before the lecture, at 5:30 p.m.