🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

New preprint from myself, Justin Gardner and @JonWinawer - 'What has vision science taught us about fMRI?' We review how (and why) our knowledge about the visual system has been a useful tool for characterizing and understand many facets of the fMRI signal psyarxiv.com/76mez

Have you met people with a bad taste in art? What about a bad taste in clouds? Check out our new paper at Current Biology about individual differences in *daily* #aesthetic taste! authors.elsevier.com/sd/article/S09… (1/7)

Here's a great summary of a fun collaboration we worked on over the last year! For vision nerds interesting in natural images, contrast sensitivity, Gabors, visual search, and modelling....

This technology is incredible.

Shout out to @chelle__roberts who has published her honours work on fractals and natural scene perception! 🥳 Her research shows that 1/f tuning of the visual system is mostly driven by spatiotemporal structure (geometry) over fine luminance variations (photometry). More info👇

Great fun to be interviewed live by Boiling Point! Check it out to hear about my research on visual processing, restoration, and nature scenes (ft. a story about my time in the Amazon jungle 🦎🐜🦋)

Delighted to have a preprint of our "Anchor" manuscript for the HumanBrainAtlas up! biorxiv.org/content/10.110… Steve Kassem george paxinos Zoey Isherwood Thomas Shaw @MarkusBarth2

Galactica is basically GPT-3 for science. It can write whitepapers, reviews, wikipedia pages and code. It knows how to cite and how to write equations. It's kind of big deal 1/ 🧵

Our study identifying factors in persistence of detrimental behaviour in humans out now pnas.org/doi/10.1073/pn… Lovely working again w/ Jessica Lee, Liew Shi Xian, (twitterless?) Peter Lovibond, Gabe Weidemann, @gavanm Brief sum below

Glad our Hackathon students are having a great time at our first Junior UTS-CBA Hackathon 😊 Our students will be learning how to clean/analyse replication data, and how to present their findings in an industry setting. #experimentation #replication CommBank UTS Business School

Lets talk about a single texture that's been a huge game changer in the world of gaming. And the crazy part? You probably haven't even noticed it. This texture is literally EVERYWHERE. And there's something pretty special about it that's made it so popular. Another big🧵

Widely touted reconstructions of images from fMRI data of people looking at images hinge critically on the data selected for training and testing the algorithms. We are still a long way off from mind reading, and it’s going to get more and more important to vet claims carefully

🚨There's different types of expertise in facial recognition 🚨 We tested #DeepNeuralNetworks, #SuperRecognizers & #ForensicExaminers from 14 countries & found: 👉#AI + human expert = highest accuracy 👈 Read more: rdcu.be/dg0B9 Dr James Dunn David White @MoretonReuben

"could a large language model be conscious?" is now published (with an afterword, updating 8 months after i gave this as a NeurIPS Conference talk) in the Boston Review. bostonreview.net/articles/could…

The map is not the territory! Our paper challenging the canonical model of early extrastriate cortex is now available in eLife - the journal doi.org/10.7554/eLife.… @NandaRibeiro93 Liz Zavitz [email protected] Ash York

Very nice, haven't seen this one before!

The spirit of this ingenuity is still there at SPMIC UoN Physics & Astronomy ! Miss being a part of the action there!

Predicting human psychophysical judgments based on large language models (LLMs). The authors use GPT to elicit pairwise similarity judgments across six psychophysical datasets, recovering well-known psychophysical representations such as the color wheel and the pitch spiral. 🌈🌀