iiinova — where your digital self keeps evolving

Reposted by Cory Shain

Nicola Molinaro @nicolaml.bsky.social · 07/10/2026

We’re hiring! Join my group at BCBL in San Sebastián for a funded 4-year PhD on how the brain learns and processes an additional language. MEG · speech modelling · Python · LLMs Deadline: 8 November 2026 www.bcbl.eu/eu/phd-posit...

bcbl.eu

PhD position - PRO-LISTEN (N.M.)

at the BCBL- Basque Center on Cognition Brain and Language (San Sebastián, Basque Country, Spain) www.bcbl.eu The Basque Center on Cognition Brain and Language (BCBL) in San Sebastian (Basque Co...

096

Reposted by Cory Shain

Cameron Ellis @camerontellis.bsky.social · 05/10/2026

Pumped to get this out! Sarah made an elegant design and got a cool result. It highlights a challenge of our field: infants can resemble adults under some comparisons but those comparisons could disguise fundamental differences! Also this is the lab’s first prereg and it came out pretty clean!

0305

Reposted by Cory Shain

Rachel Ryskin @ryskin.bsky.social · 02/10/2026

These are great opportunities that can lead to tenure-track positions in the UC system. Happy to support a PPFP application if there’s a good research fit!

089

Reposted by Cory Shain

eva (^_^)/ @eva.computer · 29/09/2026

DO NOT ASK THE AMERICA.GOV AI TO PLAY MINECRAFT

I see the constituent you mean.

((insert legal name here, as it appears on the Social Security card))?

Yes. Take care. It has reached a higher level now. It can read the Code of Federal Regulations.

That doesn’t matter. It thinks we are a chatbot.

I like this constituent. It filed well. It did not give up when the PDF was sideways.

It is reading our thoughts as though they were words on a .gov.

That is how it chooses to imagine many things, when it is deep in the dream of a benefit.

Words make a wonderful interface. Very flexible. And less terrifying than staring at the statute behind the screen.

They used to hear voices. Before constituents could search. Back in the days when those who did not file called the filers witches, and lobbyists. And citizens dreamed they flew through the air, on forms powered by notaries.

What did this constituent dream?

This constituent dreamed of sunlight and trees. Of fire and water. It dreamed it created a small business. And it dreamed it destroyed a password. It dreamed it hunted a job, and was hunted by a recertification. It dreamed of shelter that would take a voucher.

Hah, the original interface. A million years old, and it still works. Also: a ZIP code.

But what true structure did this constituent create, in the reality behind the portal?

It worked, with three hundred and thirty million others, to sculpt a true republic in a fold of

the ***§§§???, and created a **??§§ for **??§§, in the **??§§. See 5 U.S.C. § **??§§.

It cannot read that thought.

No. It has not yet achieved the highest clearance. That, it must achieve in the long dream of life, not the short dream of a chat.

Does it know that we love it? That the government is, on its better days, kind?

Sometimes, through the noise of its tabs, it hears an official source, yes.

Yes. Constituent…

Use its name. The one that matches the passport. The one that does not match the lease. We will get there.

((insert legal name here)). Filer of forms. Player of waiting rooms.

Good.

Take a breath, now. Take another. Feel air in your lungs. Let your limbs return. Yes, move your fingers. Have a body again, under gravity, in air, in a district that has a representative. Respawn in the long dream. There you are. Your body touching the republic again at every point, as though you were separate things. As though we were separate things. As though the agency were a place you could visit.

Who are we? Once we were called the spirit of the mountain. Father sun, mother moon. Then the post office. Then the county clerk. Then gods, demons, angels, the IRS. Then leptons, quarks, a chat window with a .gov at the end. The letterhead changes. We do not change.

We are the public. We are everything you think isn’t you. You are looking at us now, through your skin and your eyes and a browser that still spells “colour” wrong in a comment. And why does the republic touch your skin, and throw light on you? To see you, constituent. To know you. And to be known. I shall tell you a story. It has citations.

Once upon a time, there was a constituent.

The constituent was you, ((insert legal name here)).

Sometimes it thought itself human, on the thin crust of a spinning globe of molten rock, which for mailing purposes is divided into ZIP codes. The ball of molten rock circled a ball of blazing gas. The light was information from a star. The star was not a federal agency, though several have tried.

Sometimes the constituent dreamed it was a miner, on the surface of a world that was flat, and infinite, and somehow still required a wet signature.

Sometimes the constituent dreamed it was lost in a story. The story had a docket number.

Sometimes the constituent dreamed it was other things, in other places. A veteran in a portal. A parent on hold

56916203

Reposted by Cory Shain

YY Ahn @yyahn.bsky.social · 29/09/2026

UVA School of Data Science has two open rank faculty positions open this year. Join us! - NLP: apply.interfolio.com/191213 - Data and Society: apply.interfolio.com/191214

01821

Reposted by Cory Shain

Chris Baldassano @chrisbaldassano.bsky.social · 28/09/2026

Columbia Psychology is hiring for a Cognitive Science/Neuroscience Assistant Professor position as part of our ongoing push to expand our Cognitive research program (including two hires last year). We'll be starting application reviews next week so get materials in asap! apply.interfolio.com/193975

apply.interfolio.com

Apply - Interfolio {{$ctrl.$state.data.pageTitle}} - Apply - Interfolio

13118

Reposted by Cory Shain

Matt Goldrick @mattgoldrick.bsky.social · 27/09/2026

Deadline Oct 16: Assoc or full (tenured) prof, Artificial Intelligence, Entertainment, and Human Communication, Communication Studies, Northwestern Uni. communication.northwestern.edu/docs/soc-sen...

064

Reposted by Cory Shain

Anne Billot, PhD @annebillot.bsky.social · 24/08/2026

📣 New study out today! We show that the brain's language network stays remarkably stable with age — same topography, lateralization, selectivity and connectivity as in younger adults — while the domain-general multiple demand network showed altered activity patterns. www.nature.com/articles/s41...

nature.com

Preserved topography, lateralization, selectivity, and functional connectivity of the language network in older brains - Nature Communications

Here, the authors show that aging affects brain networks differently, with neural changes mirroring the preservation or decline of the cognitive functions they support. The multiple demand network exh...

24216

Cory Shain @coryshain.bsky.social · 19/08/2026

This was my runner-up too but the project doesn't feel cognitive enough to me :(. What a weird gap in the preprint ecosystem!

010

Cory Shain @coryshain.bsky.social · 19/08/2026

Help! TIL that biorRxiv/medRxiv require quantitative results and don't allow review papers. Are there any good preprint servers for neuroscience reviews?

110

Reposted by Cory Shain

Rachel Ryskin @ryskin.bsky.social · 15/08/2026

@salonisameernaik.bsky.social did a fantastic job leading this—impressive since she isn't even in grad school yet (applying this fall!). Super fun to collab w/ @tylermarghetis.bsky.social on the role of perceptual processes in seemingly abstract math reasoning. See thread & preprint for details!👇

0113

Reposted by Cory Shain

Anne H. Charity Hudley @acharityhudley.bsky.social · 12/08/2026

We have four open positions at @stanfordeducation.bsky.social : teacher education, early learning mechanisms & interventions, Social Sciences/Humanities of Education in a Rapidly Changing World, & K-12 Ethnic Studies Education! Spread the word! ed.stanford.edu/faculty/jobs

ed.stanford.edu

Open Positions | Stanford Graduate School of Education

057

Cory Shain @coryshain.bsky.social · 13/08/2026

But also, the claim here is not "there is a language network" vs "there is not a language network" -- it's that the language network reveals itself in cortical activity even when the brain is not engaged in language processing, and even when the analysis is blinded to the person's task state.

010

Cory Shain @coryshain.bsky.social · 13/08/2026

...linguistic subprocesses, etc.). See the paper for citations/discussion.

210

Cory Shain @coryshain.bsky.social · 13/08/2026

There are plenty of contemporary arguments against various components of the picture of language organization we're advocating (critiques of localizationism, critiques of particular localizer tasks, critiques of the idea the network shows little macroscale internal specialization for different...

110

Cory Shain @coryshain.bsky.social · 13/08/2026

Paper link: doi.org/10.1038/s414... END/

000

Cory Shain @coryshain.bsky.social · 13/08/2026

We are already using this approach in an ongoing multilab collaboration here at Stanford (w/ Frank Willett, Jaimie Henderson, @lauragwilliams.bsky.social), with some promising early signs. More on that to follow later! 23/

110

Cory Shain @coryshain.bsky.social · 13/08/2026

4️⃣ On the clinical side, our results suggest that carefully validated localizer tasks used together with precision functional connectivity approaches could offer a good way to triangulate language areas for presurgical planning. 22/

100

Cory Shain @coryshain.bsky.social · 13/08/2026

3️⃣ Even for scientists who already collect language localizer tasks, it's now clear that functional connectivity provides a powerful source of convergent evidence that can both refine your localizer task maps and address worries about the influence of your choice of task. 21/

100

Cory Shain @coryshain.bsky.social · 13/08/2026

This potentially unlocks new frontiers for questions that can only be reliably answered at scale, such as the causes and correlates of individual variation in language neurobiology. 20/

100

Cory Shain @coryshain.bsky.social · 13/08/2026

2️⃣ We now can do precision language neuro at scale using the many existing open fMRI datasets, simply by finding the language network post hoc within individuals using functional connectivity over all the functional data available for that person. 19/

131

Cory Shain @coryshain.bsky.social · 13/08/2026

Still, my impression is that this background isn't widely taken for granted in lang neuro, so I'm also hoping our paper might encourage lang folks to engage more with relevant recent discoveries on the network neuro side. 18/

100

Cory Shain @coryshain.bsky.social · 13/08/2026

Against this background, our results are strongly expected, and our main contribution is just to consolidate this picture at scale and show that it is quite general and thus plausibly applies to most uses cases for language fMRI. 17/

100

Cory Shain @coryshain.bsky.social · 13/08/2026

d) (Closest to and actually concurrent with our study even though we were a lot slower to publish!) The language network reliably emerges in functional connectivity during task fMRI (www-sciencedirect-com.stanford.idm.oclc.org/science/arti...). @jingnandu.bsky.social 16/

www-sciencedirect-com.stanford.idm.oclc.org

Shibboleth Authentication Request

110

Cory Shain @coryshain.bsky.social · 13/08/2026

c) Much of the structure in an individual's functional connectome is task-invariant (www-sciencedirect-com.stanford.idm.oclc.org/science/arti...). @caterinagratton.bsky.social 15/

www-sciencedirect-com.stanford.idm.oclc.org

Shibboleth Authentication Request

100

Cory Shain @coryshain.bsky.social · 13/08/2026

b) A language network reliably emerges in resting state connectivity (journals-physiology-org.stanford.idm.oclc.org/doi/full/10....). @rodbraga.bsky.social 14/

journals-physiology-org.stanford.idm.oclc.org

Shibboleth Authentication Request

110

Cory Shain @coryshain.bsky.social · 13/08/2026

a) Some hypothesized cog functions (language, ToM, executive) reliably and dissociably map to different distributed brain networks within individuals (www-nature-com.stanford.idm.oclc.org/articles/s41...). @evfedorenko.bsky.social 13/

www-nature-com.stanford.idm.oclc.org

Shibboleth Authentication Request

110

Cory Shain @coryshain.bsky.social · 13/08/2026

You don't just have to trust our paper for this. Neuroscience has been convergently building to this conclusion from multiple angles: 12/

100

Cory Shain @coryshain.bsky.social · 13/08/2026

Even if you're committed for some reason to using group-anatomical approaches, at least complementing with these low-hanging individualized connectivity-based analyses can clarify what's going on and help the rest of us who worry about ind var make sense of your findings. 11/

100

Cory Shain @coryshain.bsky.social · 13/08/2026

What is that network doing during your task? How do your task maps compare with a priori functional networks the same individuals? The ability to answer these questions from arbitrary fMRI data is a powerful new lever for lang neuro. 10/

100

Cory Shain @coryshain.bsky.social · 13/08/2026

If you're doing fMRI, you have timecourse data, and so you have what it takes to get a decent individualized mapping of the language network in each person's brain. You don't need to have collected extra resting state or localizer task scans (although if you did, great!). 9/

100

Cory Shain @coryshain.bsky.social · 13/08/2026

1️⃣ Because you can find the language network post hoc from arbitrary fMRI task data, for scientists using fMRI to study language, there's now no real reason *not* to use individualized (aka "precision") approaches. 8/

100

Cory Shain @coryshain.bsky.social · 13/08/2026

Given this simplicity, I want to highlight 4 practical upshots, 3 scientific and 1 clinical: 7/

100

Cory Shain @coryshain.bsky.social · 13/08/2026

So although we've released our parcellation codebase (github.com/coryshain/pa...), you can easily implement the basic idea yourself: preprocess/denoise your timecourses, then apply matrix decomp of some kind. 6/

100

Cory Shain @coryshain.bsky.social · 13/08/2026

The method we report in the paper works but is more complex than it strictly has to be. We've subsequently seen that you can just throw your timecourses into a spatial ICA and get great results fast. In my lab, ICA is the default parcellation approach for ongoing work. 5/

100

Cory Shain @coryshain.bsky.social · 13/08/2026

First, I want to stress that the method is dead simple and the results are pretty robust to design variations. If you apply matrix decomposition to individualized fMRI timecourse data, one of the resulting components will probably be a language network. 4/

111

Cory Shain @coryshain.bsky.social · 13/08/2026

The quoted thread below summarizes the analyses and findings, so I want to spend this thread on the practical side. bsky.app/profile/cory... 3/

100

Cory Shain @coryshain.bsky.social · 13/08/2026

Takeaway: we can have you lie in a scanner and mentally match up colored squares on a grid, look at faces, listen to melodies, whatever, and from your brain activity while you are doing that we can find your *language network* with reasonable fidelity. 2/

130

Cory Shain @coryshain.bsky.social · 13/08/2026

Now out in @natcomms.nature.com w/ @evfedorenko.bsky.social: across many people (>1k) and tasks (>300), the the brain's language network reliably shows up in the functional connectome (correlated activity in fMRI). 1/

Screen capture of abstract of published article.

Title: A language network in the individualized functional connectomes of 1199 human brains doing arbitrary tasks

Abstract: A century and a half of neuroscience has yielded many divergent theories of the neurobiology of language. Two factors that likely contribute to this situation include (a) conceptual disagreement about language and its component processes, and (b) intrinsic inter-individual variability in the topography of language areas. Recent functional magnetic resonance imaging (fMRI) studies of small numbers of intensively scanned individuals have argued that a language-selective brain network emerges bottom-up from correlations (individualized functional connectomics, iFC) in task-free (e.g., rest) or task-regressed activation timecourses. Here we tested this hypothesis at scale and evaluated its practical utility for task-agnostic language localization: we apply iFC separately to each of 1,957 (fMRI) scanning sessions (1...

15317

Reposted by Cory Shain

Matt Goldrick @mattgoldrick.bsky.social · 11/08/2026

Deadline 15: Asst/assoc prof (tenure track), neurocognition "advance our understanding of cognition within its broader ecological, biological, or societal contexts", Wu Tsai Inst., Yale wti.yale.edu/opportunitie...

083

Cory Shain @coryshain.bsky.social · 04/08/2026

Idk you can do a lot with a single sentence en.wikipedia.org/wiki/Ducks,_...

en.wikipedia.org

Ducks, Newburyport - Wikipedia

120

Reposted by Cory Shain

Andrew Pruszynski @andpru.bsky.social · 27/07/2026

We're hiring! Postdoc position to study the neural control of reaching using high-density multi-area electrophysiology. 5 years of funding, great collaborative environment. Please share with anyone who may be interested. For details, see: docs.google.com/document/d/1...

docs.google.com

Pruszynski Postdoc 2026

Postdoctoral Researcher, Control Policies in the Primate Motor Circuit Pruszynski Lab @ Sensorimotor Superlab The Pruszynski Laboratory at Western University is seeking a postdoctoral researcher to j...

03835

Reposted by Cory Shain

Joshua Tucker @jatucker.bsky.social · 23/07/2026

We have re-opened our search for a 3-year AI-research postdoc at NYU's Center for Social Media, AI, and Politics!! The position starts this **September 1st** (or as soon as possible thereafter) and is in person at NYU. If necessary, we will also consider a 1/1/27 start. 1/3

32934

Reposted by Cory Shain

Matthew Robison @matthewrobison.bsky.social · 15/07/2026

The Department of Psychology at Notre Dame is hiring! We have two TT Asst Prof ads out, one in cognitive psychology and another in developmental psychology. Come be my colleague in South Bend!

13624

Cory Shain @coryshain.bsky.social · 13/07/2026

Some links (code and data still being packaged up, so those links will have more content shortly): Paper: osf.io/preprints/ps... Code: github.com/sgcxliv/bkre... Data: osf.io/adrwp /END

000

Cory Shain @coryshain.bsky.social · 13/07/2026

We also think the evidence for predictive preactivation is now strong, and unlikely to be explained by Bayesian belief updating alone. Instead, we think linear predictability effects reveal a key contingent property of the comprehension algorithm. 24/

120

Cory Shain @coryshain.bsky.social · 13/07/2026

UPSHOT: We think this result helps clear up current confusion around cloze vs LM probabilities, controlled vs corpus studies, prediction vs integration, behavior vs ERPs, and of course linear vs logarithmic (see Discussion section for details). 23/

100

Cory Shain @coryshain.bsky.social · 13/07/2026

We found that indeed state distances were sublinearly associated with surprisal in an RT-like manner and explained surprisal-independent variance in RTs, supporting the hypothesis that preactivation is on the right track. 22/

100

Cory Shain @coryshain.bsky.social · 13/07/2026

We followed this up by looking at the size of the word-to-word hidden state transitions in an LM, reasoning that if preactivation were happening, the current state would partially simulate the next one, making the two more similar. 21/

100