Talking to some undergrads tomorrow about the why no one actually pays us to type model.fit() and asked them ahead of time to submit answers to a "trivially easy" data problem Fun to see the results come it, wild to know how many business run with a not dissimilar fog of war

G'day connection makers. On **April 16th**, we're hosting Georg Northoff telling us about intrinsic timescales. If it sounds interesting, here's the link to register: cam-ac-uk.zoom.us/meeting/regi...

Another rule is that if you actually challenge any of this stuff, some tech bro must say "Yeah but what do you know? They actually built these systems!" www.newscientist.com/article/mg26...

We have an exciting series of talks coming up this semester. Starting on April 22nd, Wednesday's 6pm Berlin time! Join us in going back to mental representations!

Recognizing LLM-generated text isn’t just a skill—it’s a life sentence.

This #infographic by Stephanie Phung is always a favorite example during my workshops. ✅ What WORKS in this infographic? • Consistency: colors, fonts, font sizes and illustration style are the same throughout the entire visual. Everything in the visual belongs together. 1/7

Anyone else spoiled by R's pkgdown and find python's mkdocs or sphinx a bit tedious and config-heavy. Check out @richmeister.bsky.social 's awesome new {great-docs} as a batteries-included alternative to spin up an effective docs site in <15 min! 1/n 🧵 github.com/posit-dev/gr...

From an old slide deck on replication research. Rewatching this episode of the Big Bang Theory hit differently after my own experiences with conducting replication research ;)

It's hard to pick my favorite widget of our course — but our new "Patchwork Playground" is a top candidate! 🏆 Learn the {patchwork} syntax 📦 as you interactively stitch up to 8 plots in any layout you can think of 🤯 Available now in "ggplot2 uncharted" 🤘 www.ggplot2-uncharted.com/module4/plot...

Paul Rosenbaum's Causal Inference book is strongly recommended even if you think you don't need another intro causal inference book. It's short, it's precise, and it's thoughtful about sensitivity analysis and using all the evidence we have. www.goodreads.com/review/show/...

Latest post, on why a lot of LLM-based results may just be a coin toss: kucharski.substack.com/p/ai-has-an-...

artefacts in MEG: here, the participant forgot to remove a belt with a metal buckle. the buckle moves when breathing, introducing low frequency artefacts. taking off the belt solves resolves this. 🙂

Biology is full of coconuts. 🥥

If you analyse time-resolved data (M/EEG, iEEG, pupillometry, force recordings…) and feel limited by cluster-based permutation tests (CBPTs); especially when trying to determine when an effect starts or ends; you may want to try our new R package: lnalborczyk.github.io/neurogam/ #rstats #brms #EEG

Aperiodic parameters are systematically dependent on the frequency range used to estimate them. We evaluated intracortical recordings of 62 patients using both Specparam and IRASA. The full article is finally out. @martinirani.bsky.social @medelero.bsky.social ieeexplore.ieee.org/document/113...

Wasn’t joking

I build a feed for european grant funding that's updated daily and (IMO) far more transparent then the official EU fundign website: michelnivard.github.io/eu-grants-fe... (inspired by the NIH equivalent @sashagusevposts.bsky.social build yesterday)

Astrocytes spread electrical influences but don't spike. Food for thought. Cell-type specific astrocyte activation is driven by cortical top-down modulation doi.org/10.64898/202...

Everyone should wait a few weeks for him to debug, but what @lakens.bsky.social did to my brittle idea is great, it'll become a huge success (and I'll happily contribute if he'll let me (i added it into a full app here for example)) his iteration seems the better idea here so happy someone did this!

#statstab #503 Reaction time distributions: an interactive overview Thoughts: Might be useful when you start off with this type of research. But I am curious what people think for no-nos with this approach. #reactiontime #rt #distributions #guide #r lindeloev.github.io/shiny-rt/

The next scikit-learn release will allow inspecting the type and values of attributes of fitted estimators in Jupyter notebooks & example code rendered as HTML in sphinx-gallery powered project websites. scikit-learn.org/dev/auto_exa...

Last week, I mentioned this in passing in a workshop: In 1938 Enrico Fermi won a Nobel Prize for discovering two new elements of the periodic table. Lise Meitner shortly showed that Fermi was mistaken and instead had produced known lighter elements by fission. She did not win a Nobel prize.

New post, on whether I could get Claude Code to complete a data task that had taken me AGES a decade ago… kucharski.substack.com/p/how-much-t...

This looks like really useful software: a GUI-driven EEG analysis shell on MNE, specifically targeted towards resting state EEG dx.plos.org/10.1371/jour...

Move over LLMs, it's time for wetware DOOM www.newscientist.com/article/2517...

AI is rapidly changing how we write code — but how should we use it in research? To kick off our 2026 “How-To” we welcome Russ Poldrack (@russpoldrack.org; Stanford University): 🧠 Better Code, Better Science 📅 22 April 2026 🔗 Register: forms.gle/WRUdGQEpd5is... #OHBM #OpenScience

This Thursday (March 5th), we're hosting Changsong Zhou, but mind the time shift (from our usual 15:00 UTC to 14:00). Here's the registration link + more information about the talk: cam-ac-uk.zoom.us/meeting/regi...

perennial reminder that this typographic thing: * * * is called a "dinkus" en.wikipedia.org/wiki/Dinkus

looks like an awesome setup for brms

Here's what they don't tell you about topic modeling: preprocessing determines everything. Corpus homogeneity, document length, and context windows matter more than algorithm choice. BERTopic won't fix bad inputs. No model will.

I am doing a short workshop in this symposium (quoted below) for folks in and around Leipzig. Here's the abstract. I really indulged myself with this one. A Guerilla Approach to Scientific Workflow:

The preprint is now out : arxiv.org/abs/2601.21016

I especially like to use a data dictionary to add labels! :) cghlewis.com/blog/dict_cl...

cool R trick for those unaware: the {labelled} package can be used to set variable labels which are automatically used by some other packages, including {ggplot2} and {gtsummary}

Love this. “Spontaneous” has always really meant “we don’t know why X is happening”, a perspective tied to the dominance of structured task paradigms for studying cognition, behavior, and brain activity.

The mechanics of conspiratorial reasoning and why it works. Next time you see a neat story, ask what has to be assumed for it to hold.

🧵“How do seemingly ordinary people become agents of state murder?” This is one of the guiding questions I ask students in my graduate class on genocide/state violence. With recent events, it is a question many Americans are asking. I do not have a definitive answer, but here is a reading list: 1/

This book is a wonderful, synthetic and richly illustrated journey through the natural history of the vertebrate brain 🤩 A big thank you to the authors 🙏 "A major theme in the evolution of the telencephalon has been the emergence of novel pathways... 1/2

These women helped to shape quantum mechanics — it’s time to recognize them www.nature.com/articles/d41...

A hitchhiker’s guide to information theoretical measures in psychology by @nielsvs.bsky.social with me and Yves Rosseel authors.elsevier.com/c/1mOwr53na-... osf.io/preprints/ps...

But wow - here's the beginning of the preface for Ruckmick 1938. I'm always taken aback when I read something from so long ago (here ~90 years) lamenting about the same things we're still lamenting about today. That's one reason I love these old books - they are so great for perspective.

Some really good tips here. Wish I had learned number 3 earlier, or ever for that matter 🥲 www.reddit.com/r/GradSchool...

Episode #34 in #TheoreticalNeurosciencePodcast: On balanced neural networks – with Nicolas Brunel theoreticalneuroscience.no/thn34 Cortical neurons seem to receive about the same amount of excitation and inhibition. One of the founders of this key idea in 1990s explains.

We propose applications to physiological networks, climate, and finance

Combo of two papers on partial information rate decomposition now out! journals.aps.org/prl/abstract... journals.aps.org/pre/abstract... Mini thread below 👇

We built the openESM database: ▶️60 openly available experience sampling datasets (16K+ participants, 740K+ obs.) in one place ▶️Harmonized (meta-)data, fully open-source software ▶️Filter & search all data, simply download via R/Python Find out more: 🌐 openesmdata.org 📝 doi.org/10.31234/osf...

#EconSky AI-assisted Programming May Decrease the Productivity of Experienced Developers by Increasing Maintenance Burden https://arxiv.org/pdf/2510.10165#EconSky "… find that productivity indeed increases. … the increase in productivity is driven by less-experienced (peripheral) developers. … 1/3

"Artificial intelligence (AI) models are 50% more sycophantic than humans, an analysis published this month has found." www.nature.com/articles/d41...