“#TheoryOfMind Might Have Spontaneously Emerged in Large Language Models”

https://osf.io/csdhb/

😅 😂 🤣

#ActuallyAutistic #TeamBullshitScience

📢 Exciting News! Our latest paper "White and gray matter correlates of theory of mind in autism: a voxel-based morphometry study" has been published at Brain Structure and Function! 📚🧠 #AutismResearch #Neuroscience #TheoryOfMind

🔗 Read the full paper here: https://rdcu.be/dgWvm

Kudos to my co-authors & collaborators, Kaitlyn May & Rajesh Kana! Many thanks to the anonymous reviewers & the editor for taking the time to review our paper! 🙌🏼

@yourautisticlife @dyani @actuallyautistic Well, of course, it WAS obvious — to her. #TheoryOfMind

🌟 The Many Faces of Fish: How Much Do We Really Know About Their Emotions and cognitive abilities? https://www.pawssnouts.site/2022/12/the-many-faces-of-fish-how-much-do-we.html #Fish #fishcognitivecapabilites #fishemotions #pets #theoryofmind

#Autism #Solipsism
I'm wondering what other #Autists think about the "idea that only one's mind is sure to exist".
"Solipsism" has never been a concern for me... Perhaps as an #ActuallyAutistic I compensate for my weaker #TheoryOfMind , but I have no difficulty believing that other people have their own mind!
This article is worth a read 🔗 https://johnhorgan.org/cross-check/the-solipsism-problem

💡 The Many Faces of Fish: How Much Do We Really Know About Their Emotions and cognitive abilities? https://www.pawssnouts.site/2022/12/the-many-faces-of-fish-how-much-do-we.html #Fish #fishcognitivecapabilites #fishemotions #pets #theoryofmind

The Many Faces of Fish: How Much Do We Really Know About Their Emotions and cognitive abilities? https://www.pawssnouts.site/2022/12/the-many-faces-of-fish-how-much-do-we.html #Fish #fishcognitivecapabilites #fishemotions #pets #theoryofmind

Good News, Everyone!

This weekend I cracked[0] Sir Simon Baron-Cohen's laptop, and uploaded one gibibyte[1] estrogen to his hard drive.

I took a screenshot of his lock screen as evidence that I didn't just make this up.

#Spectrum10K #TheoryOfMind #ToB #ExtremeMaleBrain #EMB

[0] Hacking ain't cracking, hmkay.
[1] This is why your own hard drive is always smaller than it says on the tin.

It's our mistaken #TheoryOfMind that will get us into trouble here, not "the threat of AGI" or whatever.

The fundamental opacity of the model means that we are unable to predict sharp-edged failure cases, even in areas where the model has demonstrated some basic competence, e.g. passing bar exams.

This opacity and sharp-edged-ness combine to produce a system that is *unsafe by design*.

https://youtu.be/rd8sITBnijg #functionalism #theoryofmind

@dgolumbia @garymarcus

David -

I don't think that Michael Kosinski is even close to suggesting that the ability to infer unobserved states is mind is in itself a form of consciousness.

#SentientSyllabus #ChatGPT #TheoryOfMind #ai

A paper on arXiv finds that an emergent ability to solve Theory-of-Mind (ToM) tasks, in ChatGPT (Thanks @kcarruthers). Such emergent behaviour is particularly interesting because it has not been built into the algorithm by design.

https://arxiv.org/abs/2302.02083

I find particularly intriguing (although the authors don't discuss that point) how beliefs change simply with the length of the conversation, even when no new facts are added. The philosopher Paul Grice stated four maxims of communication: quantity, quality, relation, and manner; aspects that allow speakers and listeners to establish contextual information _implicitly_. It is intriguing to think that this need to evaluate implicit context is a necessary condition for natural communication, and that this is the stimulus for ToM emergence.

I'm intrigued - but not totally surprised. The ability of LLMs to pass the "Winograd Schema Challenge" already showed that there is something going on. Example:

Human:
(1) The cat ate the mouse, it was tasty. Who was tasty: the cat or the mouse?
(2) The cat ate the mouse, it was hungry. Who was hungry: the cat or the mouse?

AI:
(1) The mouse was tasty.
(2) The cat was hungry.

... and you can easily try that for yourself.

That paper is here:
https://arxiv.org/abs/2201.02387

#SentientSyllabus #ChatGPT #HigherEd #AI #Education #TheoryOfMind #Mind #Intelligence

Hot off the press! 🔥

Title: Children's understanding of values as mental concepts: Longitudinal changes and association with theory of mind

Authors: Kinneret Misgav, Anat Chomsky, & Ella Daniel

The study examined how children develop in conceptualizing their values in internal-mental terms (IMS). Also, the authors found a cross-lagged association between IMS terminology and ToM comprehension.

https://onlinelibrary.wiley.com/doi/10.1111/sode.12666

#OpenAccess #TheoryOfMind #Children

And I fully expect that to happen again. Why?
Well, she's 19 and we're pretty sure that she doesn't have #TheoryOfMind. At least not particularly well-developed. And it's unlikely that she will develop it.
I'm still puzzled as to what she was thinking? I don't think it was resentment/frustration - she was having a pretty good day. My best guess is just she saw the opportunity and thought it was funny.
4/

I used to worry that I might be a psychopath because I have to really think about the fact that other people are people but recently all I can see is the benefits of not really caring what other people think (because I don’t think about the fact that they can do that.)

And when I am reminded that like… other people have MINDS (gross) and are agents of their own reality (WTF) I am too busy being grossed out by the idea that other people can CAN think about me to worry about what those thoughts actually ARE. 😅

I think the reason that remembering other people have agency is that it reminds me that apparently I do too and the idea that I can actually… make stuff happen or not based on decisions I make is honestly a little distressing to me (maybe a lot?)

#ActuallyAutistic #TheoryOfMind

"Anjan Chatterjee - Can Art Affect the Self?"

Interesting! Think how an #Autist, with a different #TheoryOfMind and #Social behaviour, is influenced (or not) by this perception of the #Self and others.
🔗 https://youtu.be/P_zGeGXQOJ4

At the meeting of the French #robotics research group #GdRRobotique, I liked @Amp77’s presentation about #socialrobotics:
- #theoryofmind is a requirement
- social interaction model is based on Emergent dialogue […] (Antle et al., 2014) : https://doi.org/10.1145/2593968.2593971
- The FAtiMA toolkit is used for social decision-making : https://fatima-toolkit.eu/
- #AI Planning is used for social decision-making
- We wish mixing #machinelearning and expert models #ai, but for now we often use #woz

@nerthos Thanks. Yep - prejudiced people are lacking something.

But that's the point - they are not aware of what that "thing" is. #TheoryOfMind

#ChatGPT can "dream" about an interactable virtual machine that runs Linux. Further, it can connect to an imagined web and dream about the OpenAI chat assistant website. There, the imagined chat assistant can generate text answering the user inputs.
This sounds like the beginnings of #theoryofmind.
https://www.engraved.blog/building-a-virtual-machine-inside/

@nerthos Thanks. Yep - prejudiced people are lacking something.

But thats the point - they're are not aware of what that "thing" is. #TheoryOfMind