I think I speak for a lot of men (who don’t have daughters) when I say TIL
Miriam Posner@miriamposner.com · 4d agoNever in my life have my explanatory powers been as taxed as trying to explain to the 6yo that if she no longer wants bangs, she needs to grow them OUT, not cut them OFF. I tried drawing a diagram. I’ve had an easier time explaining Foucault to 19yo’s.
like us! No one is doing computer science and math with their naked brains. Pen & paper, math notation, computer language—these are our neurosymbolic add ons! @lampinen.bsky.social , @blaiseaguera.bsky.social , @kmahowald.bsky.social all get it. Not sure why this isn’t more obvious to folks
Andrea Lathrop@cabernet.bsky.social · 4d agoThis is a very good point. It's an LLM-system, at this point, not a pure LLM, so theoretical critiques about what an LLM can or cannot do should not be taken to be predictions of what an LLM-plus-a-bunch-of-stuff can or cannot do.
Exactly. And this is the sense of neurosymbolic intelligence that we humans inhabit as well!
Andrew Lampinen@lampinen.bsky.social · last wk.In short, I argue that these results align with a long-standing connectionist perspective that classical symbols systems are external to intelligence rather than a fundamental part of it — and that this more accurately characterizes how human mathematicians reason as well.
Applications are open for SFI's 2027 Complexity Postdoctoral Fellowships. SFI is looking for recent Ph.D. graduates with strong quantitative and computational skills who want to pursue independent, transdisciplinary research. Deadline: September 30, 2026 Apply: santafe.edu/sfifellowship
Alpha rhythm determines speech perception rate? And speeding it up makes unintelligibly fast speech comprehensible? This is nuts! www.pnas.org/doi/abs/10.1...
PNAS
Proceedings of the National Academy of Sciences (PNAS), a peer reviewed journal of the National Academy of Sciences (NAS) - an authoritative source of high-impact, original research that broadly spans...
pnas.org
@cogscisociety.bsky.social any word on when the website is coming back online?
Are the mental contents underlying implicit evaluations accessible to conscious introspection? Prior studies show high predictive accuracy, leading to a growing consensus that people can introspectively access their implicit biases. New paper out in JPSP! 🧵🧵🧵
Just up: my entry on #Categorization for @oecs-bot.bsky.social - Western vs Buddhist ”classical” views, William James, behaviorism, prototypes, theories, chinchillas & mantis shrimp, neural nets — it‘s got it all. oecs.mit.edu/pub/mrj0pnj5...
Categorization
oecs.mit.edu
#cogsci2026 peeps - hit me up if you wanna chat about language/thought/LLMs reasoning as pattern matching - all that good stuff arxiv.org/abs/2606.13607 arxiv.org/abs/2602.23928
Reasoning as Pattern Matching: Shared Mechanisms in Human and LLM Everyday Reasoning
When large language models (LLMs) fail to generalize or make haphazard errors in reasoning, it is often taken as evidence that LLMs are not truly reasoning, but rather performing a kind of pattern mat...
arxiv.org
Most of the federal agency science budgets goes into the salaries of students and trainees, and the decision to take that money and instead give it to industry and already rich venture capitalists is fundamentally a decision to have many fewer scientists and do far less science in the future.
Psychologists should found “participant island,” where we strand a large representative sample of people who must complete our tasks to earn airdrops of food and water. If you’re really good you get to temporarily board the Neuroimaging yacht
The lying, cheating Anthropic has hired a little man to live in my computer and talk to me
Quantian@quantian.bsky.social · 4w agoHilarious new conspiracy theory just dropped: OpenAI is secretly hiring human mathematicians to solve major open problems in mathematics and lying to pretend the AI did it.
As someone who increasingly uses AI (coding, exploring connections b/w ideas, summarizing, outlining, editing..) I find LLM-generated text to be clean and efficient. LLMs know how to communicate. But I am literally nauseated by the idea of putting any LLM-written text into the public sphere.
Reddit is one of the few repositories of human knowledge and experience over decades, which makes it ten times more valuable than any meta product. Protect it and use it for the public good. #ScienceSky 🧪
Alejandra Caraballo@esqueer.net · 4w agoThis is a pretty cool use of AI. They had it scan reddit comments to analyze adverse effects of GLP1s that weren't analyzed during trials. They noticed menstrual issues were prominent. Then they dug through the FDA adverse event reporting database and were able to verify and publish a paper on it.
This is the sort of thing that makes me think that in 50 years we’ll look at today as the dark age of neuroscience. Back when people believed in the doctrine of spikes.
Earl K. Miller@earlkmiller.bsky.social · 4w agoThe paper argues that two major puzzles, how the brain creates a unified experience and maintains an “edge-of-chaos” state, may be explained by continuous electromagnetic fields generated by neural activity. doi.org/10.3389/fnco... #neuroscience
Research measuring the effects of AI in education has a major quality control problem that should disqualify most of it from being considered in policy discussions or applied in teaching practice. Quick new post from me codeactsineducation.wordpress.com/2026/06/26/t...
The problem with evidence production on AI in education
Photo by Camille Villanueva on Unsplash The research field of artificial intelligence in education (AIED) has a long history of investigating the effects of AI on learning and other educational out…
codeactsineducation.wordpress.com
It’s almost as though the admin is planning to funnel this X-labs money to certain someones. Almost… 😐
Neil Lewis, Jr.@neillewisjr.bsky.social · last mo."Program managers would normally rush to inform potential and current grantees about such dramatic changes. But the memo tells program managers to keep their mouths shut. 'This information is highly confidential...Please do not communicate anything to PIs.'” www.science.org/content/arti...
—— Open Mind is MIT Press’s Diamond Open Access Cognitive Science journal. We now have our first Impact Factor: 2.9 as of June 2026 Submit your papers to Open Mind direct.mit.edu/opmi
Great piece. It may just be the algorithm but it does seem like there’s a recent vibe shift where a new group of people are having a “ohhh this is real” moment that others had a long time ago. Agentic LLMs workflows seem to have a lot to do with it. Others are reacting by digging into their denial.
Aeon Magazine@aeon.co · 2mo agoAdopting a cautiously optimistic perspective on AI, the literary critic and editor Martin Puchner proposes that we share something fundamental with the machines: the use of language. To explore this, he draws on a rich tapestry of theory, literature and life experience in this Essay
At the #Neurips2025 mechanistic interpretability workshop I gave a brief talk about Venetian glassmaking, since I think we face a similar moment in AI research today. Here is a blog post summarizing the talk: davidbau.com/archives/202...
What's so poignant about this heart-wrenching thread is that the students know AI can't be brushed off. This is a Copernican moment for human cognition and so many academics seem to be stuck on it-can't-even-spell-spell-strawberry level denial.
Jeff Sharlet@jeffsharlet.bsky.social · 2mo agoAt the end of the term I asked my college creative wriing students to submit anonymous thoughts on AI. No real surprises: Mood ranges from resignation to despair, capitulation from embittered erosion of standards to total, feelings of betrayal from deep to furious. 1/
I used to explicitly teach exactly this at university of bath computer science in a course called “research project preparation” How to read science articles was in the same lecture as how to not plagiarise / how to cite; iirc Good to have a reading for @hertieschool.bsky.social students, thx Pat
Pat Savage@patrickesavage.bsky.social · 2mo agoMy hot take: we faculty bear some blame for failing to teach students how to read efficiently. Reading every word was never practical, and LLMs tempt students we haven’t taught otherwise. Here’s what I wrote to teach this - please use with your own students if helpful! doi.org/10.31234/osf...
