@kenjilee
Ph.D. student studying the in vivo identification of cell types and the neural dynamics of decision making in prefrontal cortex. Chand Lab @ BU; NINDS F31 Fellow; prev. UW, Allen Inst., and U. Puget Sound. From Hawaii 🌴
Psyched to announce our COSYNE workshop on social behaviors (Mar 17th, Cascais). We have a stellar lineup of speakers on topics from animal cooperation and aggression to the social dynamics of LLM agents.
Co-organized with Libby Zhang (Allen Institute + UW).
cosyne-social-behavior.github.io
When I was choosing between PhD programs, I had a hard time gauging what mattered most. Everyone I asked valued something different, which made it even harder to figure out what I should be weighing.
Wrote up the framework I wish I'd had: www.yaelcourtney.com/resources-an...
If you’re a rock climber, Vetbond is clutch for closing skin splits. Fast, doesn’t get hot, and non-toxic! Other hacks include the dremel sanding bit for finger calluses and presentation laser pointer for pointing out holds on a route. I of course would never use lab supplies for such things!
📣 Excited to announce the 2nd edition of our workshop
“Agent-Based Models in Neuroscience: Theory, Autonomy, Embodiment & Environment”
at @cosynemeeting.bsky.social #CoSyNe2026!!
🧠🤖🌍🪰🐟🐭💪🧘🏃
🗓️ March 17, 2026
📍 Cascais, Portugal
🔗 Speaker lineup and schedule: neuro-agent-models.github.io
Your figure quality is 🤯. Can't wait to read!
I am totally pumped about this new work . "Task-trained RNNs" are a powerful and influential framework in neuroscience, but have lacked a firm theoretical footing. This work provides one, and makes direct contact with the classical theory of random RNNs:
www.biorxiv.org/content/10.6...
Going to #COSYNE2026? Don't miss our tutorial on our open neural dynamics data resources.
🗓️ March 12, 9:15-10:15am
🧑🏫 Presented by @sejdevries.bsky.social
🔗 More info: https://www.cosyne.org/tutorials
@cosynemeeting.bsky.social
“I want to study how the brain takes knowledge about some topic then applies it to solve a totally new problem...”
tinyurl.com/ynvyv685
Monitoring gene expression in the brain with a simple blood test?
In collaboration with @jerzyszablowski.bsky.social & his lab we figured out how to do it nonhuman primates using noninvasive neuroengineering of synthetic serum markers.
nam11.safelinks.protection.outlook.com?url=https%3A...
I whipped up another learning related Skill! Smaller than Learning Opportunities, but very complementary: interactive guidance through a quick research-backed psychological intervention that helps improve learning plans, motivation and commitment
github.com/DrCatHicks/l...
NYU's Center for Neural Science is seeking a faculty candidate that would be jointly appointed with our Tandon School of Engineering. We are looking for post-doc applicants with neuroengineering or computational backgrounds.
apply.interfolio.com/182074
Another work “of potential interest” from the Chand lab! Nicole’s work on identifying cell types is now in press! With careful dissection, high-density probes can tell you quite a lot about underlying physiology!
Most of the quotes are similarly strongly worded. They don’t imply much or any openness to how AI can be useful at all much less how it can be used judiciously. And if they never use it, how will they ever be convinced otherwise?
I took at some of the “emblematic quotes”, Here’s one of them on what excites them about AI, “Literally nothing: it's a dismal
prospect of short cuts, lazy
scholarship, and people without
training in a discipline creating
"researched" publications that are
inaccurate and misleading.”
I guess I don’t mean to say the existence of the relationship is surprising but just that this category is so strongly obstinate. I worry these friends are being left behind.
This to say I don’t disagree with anything. I think users of AI have a duty to corral the enthusiasms of less experienced trainees and also to help inform academics who see AI as completely without use.
I would characterize as a "careful optimist" but most of the "sceptics" and "challengers" in my personal life I have not found to have good knowledge of the current state of the tools. Most of the takes cite the old study on dev productivity which is now far outdated. metr.org/blog/2025-07...
What's interesting is in the data, it seems that the majority of those "completely against AI" also have never used it. Seems there's a strong contingent of "never have, never will" even though there are surely use cases for them. I would've liked more investigation of discipline-specific breakdowns
Thrilled to finally share this work! 🧠🔊
Using a new reinforcement-free task we show mice (like humans) extract abstract structure from sound (unsupervised) & dCA1 is causally required by building factorised, orthogonal subspaces of abstract rules.
Led by Dammy Onih!
www.biorxiv.org/content/10.6...
Key to efficient learning is realizing how we ACTUALLY learn, not just what FEELS like learning. I wrote a Claude Skill for some friends to help them think about this and they've liked it -- see Principles for some directions you could explore
github.com/DrCatHicks/l...
My take on the Defining Cell Types perspective at @thetransmitter.bsky.social
Defining brain cell types is no longer a matter of classification alone, but of embedding their genetic identities within the dynamical organization of population activity, writes @lmprida.bsky.social.
#neuroskyence
www.thetransmitter.org/defining-cel...
Cosyne Viewing Parties
Visas, costs, care responsibilities, and environmental concerns all limit Cosyne attendance. Luckily, the talks are livestreamed; but watching alone is the high road to an aneurism. Hence: viewing parties! Gather regionally to watch Cosyne talks! More info: shorturl.at/3DHZX.
As per constant toolkit learning, I don’t think that’s a bad thing. The reason devs are learning new tooling is because it’s improving so rapidly. Every few months has felt like a paradigm shift. I’ve found helpful to ask Claude to ask me questions to check my understanding of what it has done.
I think certainly true about the skill degradation: I find myself resisting the urge just to accept everything Claude spits out. But I struggle to differentiate this from say people telling me to learn C instead of Python because I would learn memory management.
Finally! Cerebellum gets put at the forefront!
If I recall the research study correctly, it’s the MIT one from a year ago which studied even older models. That result was believable at the time. Today, not so much.
I think this may have been true of LLMs last year but afaik, the newest stuff is a completely different product. Claude Code and Codex has turned single experienced devs into entire teams of devs. I don’t know a single dev not using it and loving them.
Reminds me of the time Jeff Hawkins had this “brain” on one of his slides unironically. It was like the first slide and I just dissociated for the rest of the presentation.
NIH, CDC, HHS, and all of US cancer research and public health needed and still needs protection from Russell Vought and the P2025 White House.
The defining feature of 2025 at NIH was not merely grant changes.
It was unprecedented levels of presidential interference with NIH decision-making.
1/