Last chance to join @sinzlab.bsky.social and my lab as a postdoc for a (ok - I am biased...) extremely cool active vision project at the Neuro-AI interface:
www.fens.org/careers/job-...
It doesn’t? I should definitely tell that to a few people. ;)
Very cool @braininspired.bsky.social episode!
Important for anyone considering "naturalistic" behavior and what “ecological” and "affordance" actually mean in neuroscience...
(or why it may not be enough to just switch the PowerPoint Marr intro with a Gibson intro)
braininspired.co/podcast/232/
From our new paper out now in @currentbiology.bsky.social: www.cell.com/current-biol... w/ @neurofishh.bsky.social @gkafetzis.bsky.social @denilsson.bsky.social
Looking across animals, the vertebrate eye is an obvious outlier. Why is it so different that other highly visual animals?
Pace of ecology drives the tempo of visual perception across the animal kingdom www.nature.com/articles/s41... - new paper with Clinton Haarlem, Cliodhna Hynes and colleagues
Different species see the world as fast as they need to...
Build hands-on skills in optical & electrophysiological methods at the TENSS 2026 course & advance your skills in modern systems #neuroscience
🔗 Apply by 1 Mar: https://ibro.org/training-opportunity/perc4_romania/
#IBROinAsiaPacific #IBROinUSCanada #IBROinAfrica #IBROinLatAm #IBROinEurope #training
You have until March 3rd to apply for the Cajal summer school on Quantitative Approaches to Behavior and VR at Champalimaud! Come surf and track animals with us 🏄🪰🐟🚶
cajal-training.org/on-site/quan...
🧪🧠 New preprint: helping resolve a decades-long debate in synaptic plasticity
NMDA receptors are central to Hebbian learning. Yet for >30 years, the existence and function of presynaptic NMDA receptors have remained controversial.
📄 doi.org/10.64898/202...
1/6
Using our bee-tracking drone, we discovered that honey bees 🐝 have highly precise and individual routes. Now published at @currentbiology.bsky.social : doi.org/10.1016/j.cu...
Timeline of application deadlines for hands on couses in behavioral neuroscience.
We’re excited to team up once again with advanced training courses to help students with hands-on training about open source tools and methods for behavioral neuroscience.
Don’t forget to apply! Links below.
#OpenEphys #OEdu
bsky.app/profile/thea...
As conceptually fundamental as this paper is, assessing stability in motor control (and, frankly, anywhere else) is tricky. e.g. pmc.ncbi.nlm.nih.gov/articles/PMC... - or check out @olveczky.bsky.social and @asheshdhawale.bsky.social 's work.
Are you coming to cosyne? Would love to discuss that...
... absolutely. And the software/hardware engineering we can get done in a day now is just absurd. Feels like a superpower. But that's the thing - a very strangely ambivalent professional situation that I feel is not acknowledged enough in the Doomers vs. AI-Bros vs. Luddites spectrum.
(and if I may add a science thing: the fact that an allocentric variable can be so stable given the weird things animals do when running around nilly willy is cool - and did surprise us a bit in our own data as well).
You can't... repeal... a scientific finding. At that point it's just called lying about it.
After all, we had this settled once and for all years ago
...I do find it interesting (no judgement) how this illustrious journal is publishing drift "is a thing" "isn't a thing" stories at a high frequency these days, though.
Also: I am not sure if humans are _on average_ much better at reviewing literature. We just accept mediocrity far more easily if it comes from conspecifics...
I agree - and I don't use it like this. I am just not sure it will continue to be. My point was not to hype science AI, but - see OP - it would be foolish to trash current models based on outdated experience and ignore obvious effects on professional identity. This may be more obv. in data science.
I certainly had these tiring "ever-drifting" discussions to the same extent as the equally tedious "ever-stable". Maybe more - seeing change is far easier than seeing stability after all. It is sadly just so tempting to attach a narrative to categorical strawmen...
I disagree. Both sides of these (nonsensical) categorical claims have been made, when interpreting claims uncharitably. But yes: We absolutely have to move past that. Not everything flows - not everything is stable. We need to measure, define, and model better. And maybe get rid of this drift term.
Speaking of writing & review in AI times: That seems to be a concept that will hit a brick wall pretty soon. Until then, I'm faced with piles and piles of review requests for 50% LLM texts - my professional pride of spending days on somebody's work to give personal feedback is diminishing rapidly...
I guess what I want to say is: Yes. It is helping a lot with boilerplate, bureaucracy, grant-fluff parts, etc. But it is also 'helping' (reasonably proficiently - this is not a slop critique, quality is the problem) with the stuff that needed effort and yielded accomplishment payoffs. That's scary.
Writing. I love/hate writing. I teach writing. Nothing is more painful and satisfying than chiseling at this one sentence until it is just right. Now we have an instant fix ("improve using Strunk&White") that is usually not bad at all - and completely dissatisfying, even when good.
Digging deep, condensing knowledge, and writing review/opinion articles - is there a point in doing this anymore?
Sure.
The software / hardware engineering we do in the lab - complex integrated systems. Wonderful flow-sessions of figuring things out and getting them to run. Claude is now better at this.
This is what I find scary, though. The nitty-gritty technical details, the "I can do this" part of the job is something I really love. And in the last months, I came to realize that this may go away....
Same. Going from slob-prompting (*meh*) to full codebase integration (claude code / GPT codex) was pretty wild (and also more than a little scary...).
can we?
a neuroscientist: "I see drift in area X of species Y under conditions Z therefore it exists everywhere."
also a neuroscientist: "I think we can explain drift, therefore it happens."
