Skip to main content
Upcoming Event

Neural Mechanisms of Strategy-dependent Decision-making in the Prefrontal Cortex

  • December 17, 2024
  • 4:00 PM - 5:00 PM
  • A Level Physics Seminar Room, Room A30, Smith Hall Annex (CRC)

Event Details

Type
Center for Studies in Physics and Biology Seminars
Speaker(s)
Giancarlo La Camera, Ph.D., associate professor, Stony Brook University
Speaker bio(s)

The ability to make decisions according to context is a hallmark of intelligent behavior. The prefrontal cortex (PFC) is known for processing contextual information, but many questions remain open. This is especially the case for ``strategic" behavior where the context follows from abstract rules rather than dedicated input cues. In this work, we investigate the neural basis of two strategies called `repeat-stay' and `change-shift' strategy, respectively. These strategies have been observed in monkeys performing certain types of context-dependent tasks; in the task studied here, one of three targets are chosen based on an instruction stimulus and the outcome of previous trials. The same stimulus may instruct different decisions and the same decision may result from different stimuli, requiring the ability to develop strategic rules that span multiple trials. We found that PFC activity makes sharp transitions across latent neural states encoding task variables such as strategy, decision, action, reward, and previous-trial decisions. We compared two models able to perform the same task: a recurrent neural network (RNN) trained via backpropagation through time, and a multi-modular spiking network (MMSN) containing realistic ingredients of real cortical networks. Both models successfully attain levels of performance comparable to the monkeys’; however, the RNN seems to learn specific combinations of task conditions while the MMSN adopts the abstract strategies. The MMSN also reproduces the sequence of sharp transients observed in the PFC data, and explains some behavioral errors as the consequence of temporally misplaced transitions. In summary, the spiking network’s modular architecture suggests possible mechanisms for storing information across trials and subserve strategic behavior in complex tasks.

Open to
Public
Contact
Melanie Lee
Phone
(212) 327-8636
Sponsor
Melanie Lee
(212) 327-8636
leem@rockefeller.edu



Calendar of Events & Lectures

Browse upcoming and past Events and Lectures by keyword, program, date and type


For virtual attendance
When available, remote attendance options can be viewed while connected to the campus network or VPN.