Yet again, machine learning — even gussied up via the transformer architecture — encodes and reinforces societal biases. This study reveals that LLM-based peer review relies heavily on author institution in its decisions. arxiv.org/abs/2509.15122

"In a landmark study, OpenAI researchers reveal that large language models will always produce plausible but false outputs, even with perfect data, due to fundamental statistical and computational limits." www.computerworld.com/article/4059...

Can large language models stand in for human participants? Many social scientists seem to think so, and are already using "silicon samples" in research. One problem: depending on the analytic decisions made, you can basically get these samples to show any effect you want. THREAD 🧵

I'm sorry, worldwide, irrevocable, non-exclusive, transferable permission to my voice and likeness? For what now? In any manner for any purpose??? This is in academia/.edu's new ToS, which you're prompted to agree to on login. Anyway I'll be jumping ship. You can find my stuff at hcommons.org.

We need new rules for publishing AI-generated research. The teams developing automated AI scientists have customarily submitted their papers to standard refereed venues (journals and conferences) and to arXiv. Often, acceptance has been treated as the dependent variable. 1/

Finally! 🤩 Our position piece: Against the Uncritical Adoption of 'AI' Technologies in Academia: doi.org/10.5281/zeno... We unpick the tech industry’s marketing, hype, & harm; and we argue for safeguarding higher education, critical thinking, expertise, academic freedom, & scientific integrity. 1/n

Do NOT perform "image enhancement" with LLMs or AI for any forensic purposes whatsoever. The people doing this have no clue of how images are formed, nor of how AI/LLM image manipulations function. AI/LLMs introduce *NEW DATA* that leads to mis/dis information.

🧠 New preprint: Why do deep neural networks predict brain responses so well? We find a striking dissociation: it’s not shared object recognition. Alignment is driven by sensitivity to texture-like local statistics. 📊 Study: n=57, 624k trials, 5 models doi.org/10.1101/2025...

📣 NEW -- In The Economist, discussing the privacy perils of AI agents and what AI companies and operating systems need to do--NOW--to protect Signal and much else! www.economist.com/by-invitatio...

What is it about human brains that allows us to reason symbolically whereas most other animals cannot? And why does AI sometimes struggle to generate images of objects in statistically unusual spatial relationships to each other, when people wouldn’t struggle to imagine such relationships?

The Bundeswehr (German Army) officer graduation ceremony is held on the site where the men who tried to assassinate Hitler were executed. They are admonished to take those people’s examples, so that when you are given an order that is unethical, it is your duty to refuse or disobey it.

The psych job market may not be dead... but it is gravely injured 😬 So far it's looking like the Trump administration's attacks on higher ed/research are going to have more than 2x the impact on the job market as the covid-19 pandemic. #psychjobs #neurojobs #academicjobs

Around 80 M1 across the 6 tracks and 65 M2 students, we are proud to welcome the new 2025-2026 #cog-SUP cohort! Master in Cognitive Science from @upcite.bsky.social and @sorbonne-universite.fr

As #ECVP2025 wraps up, I’m delighted that #ECVP2028 in Paris was accepted at the Business Meeting! ECVP will meet again before but the Parisian organizing committee is already planning an unforgettable 50th anniversary!

Results so stunningly clear they inspired this classic xkcd (xkcd.com/2400/):

does someone good at coding & analysis want to work remotely w/ us in the coming few months (before end of 2025), as a paid consultant? project will be on neurofeedback (fMRI, ECoG, calcium imaging). we'll work towards developing the experiments & analysis pipelines together. if so pls DM me ur CV🧠📈

Our target discussion article out in Cognitive Neuroscience! It will be followed by peer commentary and our responses. If you would like to write a commentary, please reach out to the journal! 1/18 www.tandfonline.com/doi/full/10.... @cibaker.bsky.social @susanwardle.bsky.social

Hey #ECVP, we got our colleague @joshuamartin2902.bsky.social on BSky! Give him a follow! :)

Pretty nice city for a pretty nice conference (jubilee).

G'day #ECVP! Today, three more colleagues present their posters. 38: @celinehonekamp.bsky.social will introduce you to dual adaptation in the temporal domain. 48: Tarek will present a relative depth experiment conducted on the multifocal display he built.

There's a lawsuit about AI stealing your work. It's the same lawyers taking on Elsevier et al in a separate case. Academics: 1. Check if your work is in LibGen at www.theatlantic.com/technology/a... 2. If so, let the lawyers know at www.lieffcabraser.com/anthropic-au...

Good morning #ECVP! Today, two of my colleagues are presenting their posters in the morning and afternoon session 🎉 Lina will introduce you to (mismatching) causal relationships and @lreining.bsky.social will introduce you to MooneyMaker, a Python package to create ambiguous two-tone images.

Valuable methods work at #ecvp2025! @olaf.dimigen.de draws a crowd talking about the extensive benchmarks he did with our 240 and 480 Hz oled monitors. Preprint 👉 doi.org/10.1101/2024...

#ECVP2025 be ready: my poster on repeated exposure to delay in two different tasks is up on Thursday 15:30-17:00 (No 38) Come and say hi

I won't get to join you all at #ecvp2025 @ecvp.bsky.social this year because I'm on holiday in Australia... but my lab is there in force! Please do go say hi, and check out my students' and postdocs' amazing work! @ben.graphics @lschmittwilken.bsky.social @lreining.bsky.social

Looking forward to meeting you #ECVP2025 Mainz this week, including collaborative work with @tobnie.bsky.social @dominikstrb.bsky.social @ookenfooken.bsky.social @fatatai.bsky.social @tsawallis.bsky.social @mamassian.bsky.social @guidomaiello.bsky.social @mariaeckstein.bsky.social and many others

Excited to share our new paper w/ @cibaker.bsky.social in @natcomms.nature.com linking active vision & memory! We provide evidence that gaze reinstatement & neural reactivation are deeply related phenomena that jointly reflect the experiences constructed during recall. doi.org/10.1038/s414... 🧵1/9

I won't get to join you all at #ecvp2025 @ecvp.bsky.social this year because I'm on holiday in Australia... but my lab is there in force! Please do go say hi, and check out my students' and postdocs' amazing work! @ben.graphics @lschmittwilken.bsky.social @lreining.bsky.social

End of August means ECVP! This year, I am here (in beautiful Mainz) with a special mission - to find postdocs who will join us in (equally beautiful) Krakow, PL! Drop me a DM if you are interested! @ecvp.bsky.social

And now without bluesky making the background black...

ATTN🚨: I will be looking for PhD students through NYU's Center for Data Science PhD program this year. Applicants should have an interest in either NeuroAI (specifically biological attention or AI interpretability) or ML for Remote Sensing. Visit my lab website for more info: lindsay-lab.github.io

Okay everyone, things are getting serious. I’m going to teach research methods again, 1st year psychology undergraduates. What would you cover with respects to philosophy of science, the research process etc.? I’m not very happy with the textbook stuff so I’m open to all ideas!

trying this with GPT-5 and charting new frontiers in gaslighting

The shitty first draft is and always will be the number one strategy for writing well. Editing is a craft you can hone. Writing from scratch is a different dumpster fire every time you put a pen to a blank page.

fascinating write-up by @edzitron.com suggests the #GPT5 router can cause massive infrastructural load. Almost as if cognitive science work on decision problems and computational complexity is relevant if you try to mimic metacognition 🫠 www.wheresyoured.at/how-does-gpt... 1/2

Position open for a neuroimaging focused postdoc to come work with us in Sydney Feel free to get in touch for a chat if you have any questions about it :) usyd.wd105.myworkdayjobs.com/en-GB/USYD_E...

10 years after I came as a participant in 2015, tonight I had the pleasure of giving an evening lecture at the Studienstiftung summer academy in the Swiss mountains ⛰️ My talk on brains-as-computers fell on open ears and there were 45 min of super exciting questions. What a joy!

Slightly diminish a band: Rail against the machine

“The AI in the study probably prompted doctors to become over-reliant on its recommendations, ‘leading to clinicians becoming less motivated, less focused, and less responsible when making cognitive decisions without AI assistance,’ the scientists said in the paper.”

New from Arnold @visnerd.bsky.social lab: Mental rotation is often regarded as paradigmatic for #mentalimagery. But it turns out people often don't use imagery for mental rotation - & when they do it is often not useful (same viewpoint trials). #visionscience #psychscisky doi.org/10.1016/j.co...

I still get chills Meet Mike *30+ years severe depression *first hospitalized @ 13y *20 meds *3 rounds of ECT *2 near-fatal suicide attempts Mike felt joy for the first time in decades after we turned on his new brain pacemaker or PACE see videos, read paper, follow thread doi.org/10.31234/osf...

music lessons don't make kids smarter, but many people have assumed that music lessons would have near-transfer effects, e.g. improving aspects of auditory perception big new study led by Andrew Oxenham says "nope". music lessons make kids better at music & that's good enough reason to do 'em !

Hiring/Seeking a *permanent* programmer in ecological modelling at my lab @unibonn.bsky.social. We mechanistically model plant communities, island biogeography, species range dynamics, eco-evolutionary feedbacks, diversity gradients, tropical forests, vascular epiphytes. Pls rt! shorturl.at/Bzzdt

Another email to Bhattacharya about his interview and his continued statements that NIH's gain of function research caused the COVID pandemic and killed millions of people. 1/8

Stuff like this makes me laugh as someone who has worked in education all around the world. Nowhere is perfect. But to posit that U.K. schools and teachers are super far ahead in thinking on learning etc is ….interesting. I find U.K. teachers often “know” cogsci stuff as pop science only,

At PhD/post-PhD level, academia is fundamentally dialogical and unless you intend to hide behind a prompt your whole life, you are going to look like a fool when you open your mouth and don't have a working understanding of the things you profess to write about.

Motherfucking wind farms…

The "on the thought train" chapter starting at 24:20 is the crux of the whole thing and it's wonderful www.youtube.com/watch?v=TMoz...

1 To predict the behaviour of a primate, would you rather base your guess on a closely related species or one with a similar brain shape? We looked at brains & behaviours of 70 species, you’ll be surprised! 🧵Thread on our new preprint with @r3rt0.bsky.social , doi.org/10.1101/2025...