What determines the pattern of working memory activity in the mouse cortex? We tackled this question in our new model of the mouse cortex and thalamus. This gives us a new cell-type-centric view on the specialisation of mouse cortical function @eLife 1/9.

RT @UlysseKlatzmann: What happens in the brain when a stimulus enters consciousness?. We built a biophysical model of the cortex to zoom in….

This is the fruit of an intense 3-year collaboration with.@ulisespereirao (who is the real driving force). It was guided thoughtfully by Xiao-Jing Wang. It has been a wonderful experience to think through problems with and learn from Ulises and XJ.

This work is inspired by work on cognitive networks by.@bttyeo , Bucker, Margulies, Spreng, work on communication subspaces by Semedo, Kohn, Machens, Yu, Kohn, and recent modelling work by Ostojic, @jmourabarbosa , @jorgefmejias and several others.

The model goes beyond our previous work in at least two other ways : 1) representations - now the same macroscale activity can represent many different stimuli depending on the pattern of neural activity within areas; 2) dynamics - chaos✅, sequences✅, fixed-point attractors ✅.

In the model, all cognitive networks interact with each other through communication subspaces (an idea from systems neuroscience). This subspaces allow selective routing of specific information across network (such as only salient stimuli being propagated to the salience network).

Why can't you typically have DMN, FPN and DAN activity all together? Because the combination of the DMN-FPN and FPN-DAN attractors is not itself an attractor of the system. The FPN can only maintain one of the two activity patterns at any time.

We propose that the DMN and DAN may each have within-network connectivity that is too weak to maintain a high-activity attractor state. However, if they recruit part of the FPN into their network they can maintain a distributed attractor state (DMN-FPN or FPN-DAN).

A simple idea emerged which could explain the flexible coupling between the brain's frontoparietal network and the Default Mode and Dorsal Attention networks, as well as the anticorrelation between the latter two.

I feel this is a big step forward in our work modeling cortex-wide activity during cognition. Inspired by systems neuroscience and nonlinear dynamical systems, we propose a mechanism for interactions between the brain's major cognitive networks.

RT @ulisespereirao: I am excited to share the last work of my postdoc as a Swartz Fellow at NYU on the dynamic routing of large-scale cogn….

RT @BristolUni: Congratulations to Dr Naomi Millner, Dr Thom Sharp, and Dr Konstantinos Tsetsos, who between them have been awarded a total….

RT @k_tsetsos: Deeply honoured to receive an #ERCCoG ERC Consolidator Grant to explore the Dynamics of Attribute Weighting in Multiattribut….

RT @russo_eleon: 📢PhD position @BristolUni (with Ross Purple and @seanfw, UK) and joint supervision @SantAnnaPisa (with @russo_eleon, IT) o….

RT @StanDehaene: Please take the time to read this Open letter of psychologists and neuroscientists for peace in the Middle East — I signed….

Absolutely cracking paper by super team @DrBMunn @eli_j_muller @jmacshine . Down Under, Top of the World.

RT @sjblakemore: Undergrads/graduates: if you want to do a PhD, can’t afford to do a masters and know the area you want to do a PhD in, a p….

RT @misicbata: Integrating brainstem and cortical functional architectures | led by @JustineYHansen in @NatureNeur….

RT @LLogiaco: At #SfN2024 on Wed. 1:30 PM? Checkout the talk by (X-free) Chris Cueva who - w/ @anzahorodnii, @MendozaHalliday, N. Qian & B.….

RT @fennamk: not a Nobel but still pretty exciting to receive a New Innovator Award from the @NIH_CommonFund! We'll be using it to study th….

Amazing to see Hopfield and Hinton honoured for their world-changing work. As with many Nobel prizes, there are many big contributions that also deserve recognition. Including Shun-Ichi Amari's:.