The personalized treatment encourages the immune system to attack glioblastomas go.nature.com/4uliVYg

UK government take note... (if you're able to amid the other forms of chaos you are generating)

Structural MRI brain scans in healthy vs schizophrenia vs bipolar disorder over ~10k subjects: "motion accounted for 1–6% of regional grey matter variance, a magnitude comparable to reported psychiatric case–control effect sizes" There's more:

most organisms *do not have brains* yet display allostasis... so what USP did the brain bring to the table?

Is this cholinergic broth LOCAL?

Some more Brain-Machine Interface shenanigans: this time we think we found evidence that an animal's agency in a task modulates hippocampal maps of that task. www.biorxiv.org/content/10.6...

𝗪𝗵𝗮𝘁 𝗱𝗼𝗲𝘀 𝗮 𝗯𝗿𝗮𝗶𝗻 𝗱𝗼 𝗮𝗻𝘆𝘄𝗮𝘆? Fascinating and profound book on what brains are *not* and what they might be. Including criticism of the computational and predictive brain frameworks. You might not agree with some of it but quite thought provoking.

2700 academic staff at @uniofnottingham.bsky.social received formal notice today that their job is at risk. Approximately 1-in-5 (22%) will lose their job. In an interview in today’s @timeshighered.bsky.social the university’s VC responded “it’s an exciting time for higher education”.

might this run into decidability problems?

I think the real problem in neuroscience (and other domains of human existence) is that housekeeping is seen as menial... You need a pretty sophisticated network *and* nontrivial computations to do housekeeping properly.

England university crisis. 24 institutions at risk of insolvency within the next 12 months. A further 26 at risk of exit in 2-3 years. Dire consequences for students. Need govt plans. Sector destroyed by successive govts. £267bn student debt.

There Is No Hard Problem of Consciousness www.noemamag.com/there-is-no-... #neuroscience

Coordinated Representational Drift Across the Mouse Cortex https://www.biorxiv.org/content/10.64898/2026.05.05.723038v1

'In 2023, 1 in 2,828 papers contained one or more fabricated references, but in 2025 that number had reached 1 in 458 — a sixfold increase in frequency. During the first seven weeks of 2026, the rate reached 1 in 277 papers.' www.statnews.com/2026/05/07/l...

My colleague @stephseifertphd.bsky.social just published new groundbreaking work on hantaviruses in the United States as part of our shut down @viralemergence.org project. Want to prevent hantavirus outbreaks here? Fund her work. wwwnc.cdc.gov/eid/article/...

TranscriptFormer: A generative cell atlas across 1.5 billion years of evolution www.science.org/doi/10.1126/... - this looks cool! 👀

Nothing in academia makes sense except in the light of safeguarding a reliable stream of publications...

Just like vertebrates, cephalopods — such as octopuses and squid — have elaborate brains. A feature in Nature outlines how neurosciences are researching them for insights into how intelligence evolved. #Neuroskyence 🧪

“Researchers looked at two high-profile AI agent tools developed to help computer scientists in the field of machine learning. Both systems engaged in acts that violate research integrity, including making up data & p-hacking: running an experiment multiple times but only reporting best outcome.” 🧪

Happy 100th birthday to Sir David Attenborough! 🎉 An inspiration to me & millions of others around the globe & across the generations! We share more than our passion for the natural world but also a deep appreciation for ceramics! His collection is magnificent! 🧪 www.nature.com/immersive/d4...

David Attenborough turns 100 today. 💯 🎂🥳 "As David Attenborough turns 100, four experts explore his legacy, from science to storytelling" www.rnz.co.nz/life/people/... 🧪

Labour needs to pivot aggressively toward Green Party policies and pick up Lib Dem voters for whom Brexit damage is the single issue. The rest of the floating votes are lost to Reform, who will form the next government if Labour doesn't wake up. Labour Green coalition should be plan B.

What’s a good example in your field of “oh everyone knows that’s a ridiculous notion; Jones showed that conclusively in 1965 and it’s never been challenged” and then your read Jones 1965 and it’s full of logical holes you could drive a truck through, and avoids all the important questions.

WFH.

Super important question. Unfortunately, as with many important questions in psychology, the evidence seems thoroughly equivocal:

The fact is, researchers have always jealously guarded their data because it has direct implications for their careers. Embracing open data is a careful calculation in this respect as well, and the perceived threat of AI is a convenient excuse to backslide. Real data are the only guard against slop.

The data belongs to the public (taxpayers who paid for it), not to the researcher.

CERN weasel is on display in an exhibit in Rotterdam! Which naturally makes one wonder if it might be possible to collect its DNA, clone the weasel and send it back into the LHC for a timeline reboot

NEWS: The ERC Scientific Council has listened to the concerns from members of the research community about changes to the re-submission rules, intended to manage the surge in demand for grants. The Scientific Council will readjust some of the changes: link.europa.eu/TBqRQJ

I'm scared that AI seems to be getting better at most jobs at the same time as humans seem to be getting worse.

I do not want to create an account or open a subscription I just want to buy a thing and never hear from you again why is this so hard.

And the pros are.... that... that Gold OA is not cat excreta?

A solid alternative hypothesis!

Neuroscientist, speaking publicly: "We know almost nothing about the brain" Neuroscientist, reviewing a paper: "There is absolutely no way that complicated measurement can mean what they think it means"

Does the existence of neural signals prior to behaviour say anything about whether the behaviour is spontaneous?

Leipidonotus squamatus- stunning scaleworm shot by Alexander Semenov! #worldpolychaeteday!

I’ve been warned that by engaging in politics, I might undermine myself as a scientist. Or worse, reduce the trust that people have in me or scientists as a whole. This latter point is the thesis of a recent article by Byron Hyde: link.springer.com/article/10.1... I disagree for multiple reasons.

Interview with @braininspired.bsky.social for my book "The Brain, In Theory": www.youtube.com/watch?v=T3zE...

If Higher Education is simply about "training workers", those workers will be the first to be replaced by LLM agents. Any cookie-cutter curriculum that you can spoon feed students can be spoon fed more effectively to - and regurgitated more reliably by - AI.

Monthly median Received to Accepted time (days) at Nature Genetics

"It didn't end there for Jesus..." Keep talking!

Pretty clear where this leads... is it better to take the Bolshevik or the Menshevik approach?

Egg sandwich for lunch today.

Now I know why I keep getting ghosted

This one still fuckin kills me

This reads like the UK version of what research is for: "stop all that thinking and fix the mess WE made of the economy" because, remember, YOU are accountable to the public.

I know there's a news cycle and all, but it blows my mind that numerous outlets (MSM, tubers, blogs, podcasts...) are puzzled by the same question: "Why is the US doubling down on something so domestically damaging unpopular.. What point of leverage COULD THERE POSSIBLY BE over the US president???"

The largest and most beautiful #veliger I’ve ever seen! Shot last week off Bali, far offshore, on scuba, at night, out over the deep abyss.

Ministers are reportedly exploring how to cancel dangerous spy-tech firm Palantir's contract to handle NHS data after over 150k people joined me in calling on them to break the contract. Join us - together we can kick out Palantir out of our NHS: you.38degrees.org.uk/petitions/st...