Richard Naud

How are the fluctuations in electric field organized across the whole brain? We analyzed the 3.4 M samples from the international brain lab using and found a surprising structure made of communities and landmarks. doi.org/10.1101/2025.1… @JCBeique @XingyunWang4

@JulioMTNeuro Something like this 👉🏼 cell.com/neuron/abstrac…

@karimjerbineuro Link to the paper biorxiv.org/content/10.110…

Very cool talk by Richard Naud (University of Ottawa) @ #MAIN2025 “Modular Organization of Electrical Fluctuations in the Mouse Brain”

@karimjerbineuro Thank you for organizing such a vibrant event.

@s_y_chung Inspiring talk !

Thanks for the invitation! Having a blast visiting the Montreal AI & Neuro community this weekend. Learning a lot from excellent neuro-AI talks, and the interactions have been inspiring and productive. #MAIN2025

Via Rail does it again. The track rental is not an excuse for all the mediocre performance of this organization. If ever we get a passenger train in this country it needs both tracks and a new business model. cbc.ca/news/canada/ot…

1. New preprint resolving a conundrum in systems neuroscience with an AI scientist, and humans @ReillyTilbury, @dabin_kwon, @ali__haydaroglu, @JacobMRatliff, Valentin Schmutz, Matteo Carandini, @kevinjmiller10, @neuro_kim biorxiv.org/content/10.110…

For those interested in methods for analyzing a specific frequency of neural activity as a series of discrete events, the CBASS code is posted here with an excellent wiki by @QuentinPerreno1: github.com/cardin-higley-…

This op-ed published in the @nytimes by @stephenwitt thoughtfully captures the urgency and complexity of navigating AI's risks, but also my sincere conviction that technical solutions are possible — we still have agency and an opportunity to act. nytimes.com/2025/10/10/opi…

Flexible perceptual encoding by discrete gamma events nature.com/articles/s4158… #neuroscience

@rudzinskimaciej @JCBeique @XingyunWang4 Thanks!! 😁 And yes, totally with you re KNN, we'll look into this.

The longer I think about the paper the more I like it It's an "easy"(in comparison) way to find first structures, patterns to search for, sanity check and robustness (as you got the same result from 3 different arch.) And allows to check gigantic datasets The error in comparison to fancy small scale white box method is probably smaller (in Ideal scenario it shouldn't be like that but 🤷) So this study is more valuable than most I read long time ago 🥲 Still most basic knn on corr as a comparison would be nice (probably there is something more renowned but I like slightly modified knn due to simplicity and breadth of possible adjustments)

How are the fluctuations in electric field organized across the whole brain? We analyzed the 3.4 M samples from the international brain lab using and found a surprising structure made of communities and landmarks. doi.org/10.1101/2025.1… @JCBeique @XingyunWang4

@rudzinskimaciej @JCBeique @XingyunWang4 Same hope here... To answer your question, we didn't try every method, and I share your intuition that there should be another method, maybe even a simpler one able to reveal such an organization. KNN could like a good candidate. I would like to try..

I know why with DL it works better - it diminishes between datasets, electrod type placement quality local mater filtering etc effects But I hoped we would be able to overcome that with white box methods 🥲 Still did you think about comparison to knn on correlation distance or mutual information or any similar simple method that allows comparison of which datasets, locations how often and robustly show stable communities?

How prediction error drives memory updating: role of locus coeruleus–hippocampal interactions cell.com/trends/neurosc…

Thanks to @IntlBrainLab for assembling and sharing their data!!

I’ve often made the bet that we’ll never really understand what’s in the LFP (or the EEG). While I feel these findings leave me with many many further questions, I don’t think I’ll make that bet again.