Profile avatar
markkula.bsky.social
Prof at Leeds trying to model human behaviour, the weirder the better Into road safety, neuroscience, machine learning, and connections between fundamental and applied science https://environment.leeds.ac.uk/transport/staff/957/professor-gustav-markkula
136 posts 2,990 followers 821 following
Prolific Poster
Conversation Starter

When I worked in UK academia I recall hearing of a Japanese man who was baffled at how Britain had decided to run its universities like firms. “Why? Your universities are excellent and your firms are terrible.”

When I have parallelised my code

The Trump admin is causing chaos in transport research. Studies have been canceled, datasets have vanished, and experts have been fired. The Nat'l Academies’ Transportation Research Board has caved to Trump’s demands for censorship, triggering upheaval. Its future is in doubt. From me (w/scoops) 🧵

Is 2025 the year we’ll finally see some long overdue changes in the academic publishing system? So far the big publishers have been able to mostly stick to business as usual thanks to these “transitional” deals, but here’s hoping universities can keep a united front in this year’s renegotiations

Incredible diversity of eyes and vision across animals! These reviews really make you appreciate the evolution of the eyes and their perfect adaptation to the animal's lifestyle. Mice don't have a fovea, Hawks have two! Review 1: doi.org/10.1146/annu... Review 2: www.nature.com/articles/s41...

Read this if you feel doomery

I've written a free and accessible guide to Cephalopod Sentience with Alex Schnell, Piero Amodio and Peter Morse, stunningly illustrated by Roksolana Tkach. Please download and share! It's worth it for the illustrations alone! 🐙 thebrooksinstitute.org/sites/defaul...

Lots of own-body science 🧪 this week: 1) Used a microscope to classify a foreign object extracted from my own finger 2) Jitsu’d my back, sprained some weird deep core muscle - an ablation study to determine exactly which 200 everyday movements that little underappreciated guy was needed for

As US and the world becomes more authoritarian and anti-science, hugely important for the international community to get involved and show solidarity and support Great thread here 👇👇

𝗕𝗿𝗮𝗶𝗻 𝗱𝘆𝗻𝗮𝗺𝗶𝗰𝘀? If interested in brain dynamics this one is a feast! www.nature.com/articles/s41... #neuroscience

Got started a bit too late on your road user modelling? Deadlines for this workshop have been extended to Mar 3 (abstracts) and Apr 14 (full papers). We've already received some exciting submissions, so it's looking like it'll be an awesome conference and workshop. Would love to see you there! 🧪

A new permanent Global Citizens’ Assembly for People and Planet is being established for the 2025 UN Climate Change Conference (COP30). #CitizensAssembly #COP30

I increasingly think that spiralling inequality is the key reason behind so much of our current geopolitical and sociopolitical mess. Gary Stevenson is the best voice on this topic that I’ve come across, so if you havent seen him he’s def worth a watch youtu.be/FDpVj64DrcQ?...

Check out @markkula.bsky.social’s fantastic tips for “pain-free(?) academic writing”! Perfect for some weekend reading. Don’t miss his insightful list.

The agents are coming! ‘Digital humans’ in a virtual world www.thetransmitter.org/neuroai/digi... cc @thetransmitter.bsky.social

This cool paper on brain signatures of pedestrian decision making by Siwei Ma is now out in TRF. Open access for 50 days here: authors.elsevier.com/c/1kagW4tTwD... (and full postprint in the quoted post below) 🧪🧠📈

Happy to have this published …https://www.science.org/doi/10.1126/sciadv.adr6698

The new OpenAI model announced today is quite wild. It is essentially Google's Deep Research idea with multistep reasoning, web search, *and* the o3 model underneath (as far as I know). It sometimes takes a half hour to answer. Let me show you an example. 1/x

Our latest work on understanding the behavior of bounded agents in more naturalistic tasks accepted at #ICLR2025: Inverse decision-making using neural amortized Bayesian actors, with @dominikstrb.bsky.social @tobnie.bsky.social and @jan-peters.bsky.social based on @tobnie.bsky.social MSc thesis

A recipe for pain-free(?) academic writing gmarkkula.github.io/writing-reci... I’m a big fan of “top-down iterative writing”, for all kinds of academic texts. This is my take on why it works and one way to do it, in a series of (individually pain-minimised) steps from blank screen to submission 🧪

A recipe for pain-free(?) academic writing gmarkkula.github.io/writing-reci... I’m a big fan of “top-down iterative writing”, for all kinds of academic texts. This is my take on why it helps, and one way to do it, in a series of (individually pain-minimised) steps from blank screen to submission 🧪

Haha this is the best thing in a while, don’t even need to show it to my kids 😂

www.nature.com/articles/s41...

Maybe it's the wrong time for me to ask, but maybe not? What are some tangible actions a person can do who is based in the UK right now to prevent the rise of the far right and facism, especially with Farage being best pals with Musk, right now? How do we act before it's too late?

This contrast - whether statistics is about using data to define rules for decisions (like a state) or make inferences (like a person) - is at the heart of one of the biggest controversies in 20th century statistics you've never heard of. A quick primer here: conjugateprior.org/2014/06/spea...

New publication 📢 "Pedestrian crossing decisions can be explained by bounded optimal decision-making under noisy visual perception" by Yueyang Wang, Aravinda Ramakrishnan Srinivasan, Jussi P.P. Jokinen, Antti Oulasvirta, & @markkula.bsky.social. Full text: doi.org/10.1016/j.tr...

It makes no sense to me that it’s 2025. It’s weird, right? Quarter of the way into the 21st century is weird

I once naively thought that since child mortality was so much higher back in the day people would be inured to it. But reading the Stoic philosophers (Epictetus and Marcus Aurelius in particular) made it apparent: no, it was experienced as fully devastating, it took immense fortitude just to go on.

📣 Workshop on computational models of human behaviour for testing automated vehicles 📣 With Waymo and leading academics at IAVVC Sep/Oct next year in Baden-Baden, Germany Hope to see you there! Submission deadlines for abstracts / full papers are Feb 16 / Mar 30 2025.iavvc.org/program/work...

We have to define what we mean by “AI.” Because the model that advanced our understanding of protein folding and the LLMs that churn out the nonsense slop that fills the internet both get called “AI.” And the latter is co-opting the usefulness of the former to create this exactly perception.

‘Hey I didn’t bring a present but I was thinking I could play my snare drum for the newborn baby’ uhh no dude. You cannot

New paper in Imaging Neuroscience by Gabriel Weindel, Leendert van Maanen, and Jelmer P. Borst: Trial-by-trial detection of cognitive events in neural time-series doi.org/10.1162/imag...

Scholars, want to write a fundable grant in 2024? Terrific. Grants help you gain control over your career. Here are 9 tips for writing fundable grants: #AcademicSky #PhDSky

Going to adopt this as my new strategy.

I remember being up watching weekend morning kids TV with my little brothers when the news broke, and my parents still in bed refusing to believe what I was telling them en.m.wikipedia.org/wiki/Assassi...

Did you know that Claude Shannon also developed maze solving algorithms? Which he implemented physically using a wooden mouse named Theseus. To solve a given maze, on a grid, Theseus used two modes: 🧵

Walking and cycling can be a visceral aesthetic experience. While time in a car often feels like time wasted, walking and cycling can feel like time gained. (This was just before sunrise, which is at 9.15 this time of year)