Dániel Barabási
@bdanubius.bsky.social
300 followers 23 following 49 posts
Eric and Wendy Schmidt Fellow @BroadInstitute || Harvard Biophysics PhD '23 || ND Physics ‘17 || 🇺🇸🇭🇺🇸🇪🇷🇴
Posts Media Videos Starter Packs
bdanubius.bsky.social
OSNAP!

I received an NIH Outstanding Scholars in Neuroscience Award

This award makes my heart sing -- between college and PhD I spent a year researching at the NIMH, and loved every minute of the campus and community.

Feels full circle to be honored by NIMH post-PhD!
bdanubius.bsky.social
Recent PhD Grad, or planning on defending soon?

With current funding uncertainties, postdoc fellowships, often funded by private donations, can provide job stability in the coming years.

I put together a short guide and list of fellowships I applied to 👇
bdanubius.bsky.social
Tragically beautiful article from @harvardmagazine.bsky.social about the absurdity of letting death machines into our cities, and how simple fixes can return "complete" streets to the people.

Amsterdam and Copenhagen transformed, what's stopping Boston?

www.harvardmagazine.com/2025/03/harv...
Making Roads Safer for Pedestrians and Cyclists | Harvard Magazine
Working to curb road deaths
www.harvardmagazine.com
bdanubius.bsky.social
In the vein of Women's History Month and International Women's Day, Nature published a number of fascinating women's health studies recently, including:

Hormone cycle's effect on breast cancer treatment.

Brain restructuring under pregnancy.

X chromosome + brain aging.

Links👇
Reposted by Dániel Barabási
sarthakc.bsky.social
1/ Our paper appeared in @Nature today! www.nature.com/articles/s41... w/ Fiete Lab and @khonamikail.bsky.social .
Explains emergence of multiple grid cell modules, w/ excellent match to data! Novel mechanism for applying across vast systems from development to ecosystems. 🧵👇
Global modules robustly emerge from local interactions and smooth gradients - Nature
The principle of peak selection is described, by which local interactions and smooth gradients drive self-organization of discrete global modules.
www.nature.com
bdanubius.bsky.social
Are you a PhD student thinking about PostDoc options?

Fellowships provide early independence, and some early deadlines are in spring and summer!

Check out my guide below for navigating the process, plus a curated fellowship list with deadlines, salaries & more.
bdanubius.bsky.social
We call those the "Four Ss," but unfortunately the title got cut.

Also in the paper, the "Four Fs": Feeding, Fleeing, Fighting, and Mating.
manojdoss.bsky.social
Sometimes fancy science-y words are unnecessary (from Nature Reviews Neuroscience www.nature.com/articles/s41...).
bdanubius.bsky.social
So while certain elements can be reused, and indeed on different timescales, in many cases the System Two learning is a quick realization, which can be seen in the neural dynamics, where System Three places the knowledge into the animal's repertoire through more extensive synaptic rewiring.
bdanubius.bsky.social
Thanks for sharing!

I distinguish System Two and System Three as follows:

System Two, the Eureka Moment, can be implemented dynamically, e.g. a shift in attractor state, with a sprinkle of STDP.

But then System Three exists to "deepen the well", like with hippocampal replay.
Reposted by Dániel Barabási
bdanubius.bsky.social
Takeaways:
• Most innate circuits (System One) are pre-coded genetically.
• “Learning” (System Two) can be shockingly rare but potent when it happens.
• Ongoing plasticity (System Three) is mostly to stabilize or fine-tune your existing wiring.
bdanubius.bsky.social
But why do we think we learn?
• Human babies are born relatively immature → System One finishes outside the womb.
• Language/semantic memory makes us feel like everything is learned.
• AI hype around “learning from scratch” feeds the misconception that all brains do the same.
bdanubius.bsky.social
In this vein, we separate circuit formation into three “systems,” each deployed at different times and contexts:
• System One: Developmental Maturation
• System Two: Eureka Moments
• System Three: Staying Tuned
bdanubius.bsky.social
We resolve this nature–nurture conflict by proposing that:

(1) Critical knowledge for engaging the world is realized by development,

(2) Novel information isn’t strictly required for daily competence, and

(3) Plasticity mainly provides homeostatic feedback stabilization.
bdanubius.bsky.social
Dogmatically, circuits assembly has been split into two phases:

(1) predetermined, genetically driven coarse wiring of the nervous system.

(2) pruning and refinement through interactions with the environment, which is thought to fine-tune mission-critical neural connectivity.
bdanubius.bsky.social
How much do we *really* learn?

In @natrevneurosci.bsky.social‬ with Florian Engert and
André Ferreira Castro, we address why most of you humans firmly believe that patterned activity plays a necessary and instructive role in shaping neural circuits.
bdanubius.bsky.social
Mating proximity blinds threat perception

Valentine's Day themed paper: how a dopamine-governed filter during courtship turns off the serotonergic wave that would otherwise force them to abort and flee.

https://www.nature.com/articles/s41586-024-07890-3
bdanubius.bsky.social
🔥🔥🔥 Efficiency, resiliency tradeoff + behavioral reconfiguration of circuits provide clean interpretations of the multi-scale structure of brain activity 🔥🔥🔥