James Strachan 🏳️‍🌈

Experienced Humboldt Fellow and Visiting Scientist at UKE in Hamburg, working on coordination and communication in teaching interactions. He/him

[James Strachan 🏳️‍🌈 @jamesstrachan]

We have a new paper out today in Perspectives on Psychological Science!
"Flexible Cultural Learning Through Action Coordination"
https://t.co/7R6Ae3pZJJ
Link to preprint and 🧵 in the QT from first author @matcharbonneau1

[Mathieu Charbonneau @matcharbonneau1]

Look at this new preprint!
https://t.co/92t4ZBEjM2
With Arianna Curioni @luke_mcellin @jamesstrachan
(Thread; 1/10)

[James Strachan 🏳️‍🌈 @jamesstrachan]

Our results indicate that LLMs, particularly GPT-4, show impressive performance on standardised tests of Theory of Mind, but that they also show key differences in how they appear to behave 7/7

[James Strachan 🏳️‍🌈 @jamesstrachan]

LLaMA, on the other hand, appeared to show a bias in responding that may explain its superior performance on this one test (6/7)

[James Strachan 🏳️‍🌈 @jamesstrachan]

Our follow-up experiments indicated that GPT's poor performance was driven by an overly conservative stance as it claimed uncertainty despite being able to generate correct answers when probed (5/7)

[James Strachan 🏳️‍🌈 @jamesstrachan]

Very pleased to see this out now Open Access in Nature Human Behaviour
We investigated how well different LLMs performed on tests of Theory of Mind and compared their results against human performance (1/7)

[Nature Human Behaviour @NatureHumBehav]

Systematically testing language models on a broad battery of Theory of Mind tasks with comparison to human data, a study by @jamesstrachan et al. demonstrates human-like performance by AI chatbots such as GPT-4. @ASTOUND_project
https://t.co/mHrcFYZ5ew

[James Strachan 🏳️‍🌈 @jamesstrachan]

GPT-3.5 also performed at or only slightly below human levels on most tests, while LLaMA2-70B performed below human levels

The only task where this pattern didn't hold was the faux pas, where GPT models performed significantly worse than humans while LLaMA did better (4/7)

[James Strachan 🏳️‍🌈 @jamesstrachan]

Humans (unsurprisingly) found more complicated tasks more challenging than simpler tasks

What was more surprising was how well the LLMs did. Particularly GPT-4, which performed significantly better than humans on 3/5 tests (irony, hinting, Strange Stories, 3/7)

[James Strachan 🏳️‍🌈 @jamesstrachan]

We tested a broad range of Theory of Mind tests, including false belief, irony comprehension, faux pas, hinting, and the Strange Stories

We also made sure to test humans and LLMs in as similar a way as possible to ensure a species-fair comparison of their performance (2/7)

[Max Marschner @marschner_max]

Excited to share the first paper out of my PhD, now published @CognitionJourn! The joint outcome of work with David Dignath and Günther Knoblich.
https://t.co/tQS0BdhbxF

We find new evidence that co-actors represent joint actions on a group level. WE > ME.

Thread: 1/

[Emma McEwen @EmmaSMcEwen]

Very pleased to announce that the first study from my PhD thesis is out now 🥳 with thanks to @ConstableMerryn and all my other wonderful co-authors!

[nature @Nature]

Nature research paper: Synaptic wiring motifs in posterior parietal cortex support decision-making https://t.co/RpuueOS55J

[Mathieu Charbonneau @matcharbonneau1]

I am very excited to share with you the news that our book ‘The Evolution of Techniques: Rigidity and Flexibility in Use, Transmission, and Innovation’ is coming out in exactly 2 months.
https://t.co/iyehWl6bvX

1/5

[James Strachan 🏳️‍🌈 @jamesstrachan]

Check out our paper, "Flexible Cultural Learning Through Action Coordination" in Perspectives on Psychological Science

[Mathieu Charbonneau @matcharbonneau1]

As we move into 2024, time for some shameless self-promotion. Check out our paper in Perspectives on Psychological Science:

"Flexible Cultural Learning Through Action Coordination"

Paper: https://t.co/1koWyEpc88
Preprint: https://t.co/ZaKlYnXXp3
🧵👇 1/10

[James Strachan 🏳️‍🌈 @jamesstrachan]

And then it hits you...

[Massimo @Rainmaker1973]

This photo is an example of how optical illusions mess with your mind.

First you see a rock floating in the air and then...

[James Strachan 🏳️‍🌈 @jamesstrachan]

Fantastic evening tonight seeing Rodrigo y Gabriela (@rodgab) live in Hamburg

[James Strachan 🏳️‍🌈 @jamesstrachan]

@dora_kampis

[James Strachan 🏳️‍🌈 @jamesstrachan]

@GloucesterQuays @Lindt Thanks, Alia! She had a great day and is looking forward to coming back when she's run out of Lindor ;)

[James Strachan 🏳️‍🌈 @jamesstrachan]

Huge thanks to @Lindt and the team at @GloucesterQuays who made my Granny's 100th birthday special with a surprise gift of 100 of her favourite white chocolates 💜

[James Strachan 🏳️‍🌈 @jamesstrachan]

Here she is picking out some additional decoy chocolates to protect her new stash

Quote of the Day: "If I'd known being 100 was this fun I'd have done it ten years ago."

[James Strachan 🏳️‍🌈 @jamesstrachan]

Thank you to Carol (in the picture above) and her team at Lindt, and to the ladies on the Customer Service desk who all made my Granny's day