Transformers and cortical waves: encoders for pulling in context across time

L Muller, PS Churchland, TJ Sejnowski - Trends in neurosciences, 2024 - cell.com
The capabilities of transformer networks such as ChatGPT and other large language models
(LLMs) have captured the world's attention. The crucial computational mechanism …

Stability from subspace rotations and traveling waves

T Batabyal, SL Brincat, JA Donoghue, M Lundqvist… - bioRxiv, 2024 - biorxiv.org
Cortical activity shows stability, including the ability to recover from disruptions. We analyzed
spiking from the prefrontal cortex (PFC) of monkeys performing working memory tasks with …

Planar, Spiral, and Concentric Traveling Waves Distinguish Cognitive States in Human Memory

A Das, E Zabeh, B Ermentrout, J Jacobs - bioRxiv, 2024 - biorxiv.org
A fundamental challenge in neuroscience is explaining how widespread brain regions
flexibly interact to support behaviors. We hypothesize that a mechanism of neural …

Recurrent cortical networks encode natural sensory statistics via sequence filtering

CE Deveau, Z Zhou, PK LaFosse, Y Deng… - …, 2024 - pmc.ncbi.nlm.nih.gov
Recurrent neural networks can generate dynamics, but in sensory cortex it has been unclear
if any dynamic processing is supported by the dense recurrent excitatory-excitatory network …

Active filtering of sequences of neural activity by recurrent circuits of sensory cortex

CE Deveau, Z Zhou, PK LaFosse, Y Deng… - bioRxiv, 2024 - biorxiv.org
In daily life, organisms interact with a sensory world that dynamically changes from moment
to moment. Recurrent neural networks can generate dynamics, but in sensory cortex any …

Stability from subspace rotations and traveling waves

E Miller, T Batabyal, S Brincat, J Donoghue… - 2024 - researchsquare.com
Cortical activity shows stability, including the ability to recover from disruptions. We analyzed
spiking from the prefrontal cortex (PFC) of monkeys performing working memory tasks with …