Yes, learning about Neurodiversity will solve your Problems lmao

[Carl-Richard]

1 hour ago, Joshe said:

Give me more than just the eyes and I can read emotions better than NTs.

Like for example?

[Joshe]

1 hour ago, Carl-Richard said:

Like for example?

I can't demonstrate anything or isolate variables - it's the whole gestalt. Eyes, body, words, tone, what they're not saying, history, how the specific situation fits their patterns, what the situation typically motivates, whether the pieces are congruent or incongruent.

Also, the gestalt is just "known". I don't use linear thought to see it - I just see it. This is what makes me very good at poker. I (an autistic) can read people like books, which contradicts the premise of the test. 

The eyes are just one input. The test strips away everything except one narrow input and calls it "reading minds". But real mind-reading is integrating all of it. Someone can score high on that test and still oblivious to what's actually going on with people because they never learned to read the whole picture. 

The test implies that reading eyes = TOM = something autistics lack. I can say from my own experience this is bs. I'm not saying there isn't anything valuable about the test, just that it is wrong on its TOM premise. I'm not an academic so I can't argue much but beyond that.

Edited 18 hours ago by Joshe

[Carl-Richard]

2 minutes ago, Joshe said:

I can't demonstrate anything or isolate variables - it's the whole gestalt. Eyes, body, words, tone, what they're not saying, history, how the specific situation fits their patterns, what the situation typically motivates, whether the pieces are congruent or incongruent.

The eyes are just one input. The test strips away everything except one narrow input and calls it "reading minds". But real mind-reading is integrating all of it. Someone can score high on that test and still oblivious to what's actually going on with people because they never learned to read the whole picture. 

The test implies that reading eyes = TOM = something autistics lack. I can say from my own experience this is bs. I'm not saying there isn't anything valuable about the test, just that it is wrong on its TOM premise. I'm not an academic so I can't argue much but beyond that.

I have experience with autistics giving me a thorough in-depth intellectual account of how people work and how their own mind works, which is theoretically insightful and empirically plausible, but which in practice doesn't apply to themselves at all. The theory doesn't meet the reality as claimed. It might or might not be the case for you, and we can't really test that except through interactions. But that's a concept I believe also applies to autistics; not just a lack of TOM with respect to others but with respect to themselves. Because just like concrete rules and inferences based on those might be inefficient for understand other people's minds, it might be inefficient for understanding your own. But that's essentially gaslighting so take it with a grain of salt.

[Joshe]

1 hour ago, Carl-Richard said:

I have experience with autistics giving me a thorough in-depth intellectual account of how people work and how their own mind works, which is theoretically insightful and empirically plausible, but which in practice doesn't apply to themselves at all. The theory doesn't meet the reality as claimed. It might or might not be the case for you, and we can't really test that except through interactions. But that's a concept I believe also applies to autistics; not just a lack of TOM with respect to others but with respect to themselves. Because just like concrete rules and inferences based on those might be inefficient for understand other people's minds, it might be inefficient for understanding your own. But that's essentially gaslighting so take it with a grain of salt.

Just so we're on the same page, ToM is about modeling minds, not automatic emotional resonance. The more accurate your models, the higher the ToM.

It's not uncommon for me to annoy people by addressing their thoughts and emotions before they express them. Those closest to me know they can't hide anything, which makes them uneasy, but they also feel like I'm the one who knows them best, which makes me everyone's closest confidant. This is high ToM.

It works like this: right now, subconsciously, I'm storing this interaction into a map of "Carl" which will be available in the gestalt later. It grows from every interaction. For example, before I posted my AI screenshot, I knew you wouldn't like it. My mind seems to make a point of collecting data on people and that data is somehow used to form very accurate predictions about their state and behavior. 

To some reading this (I know who you are): don't project ego inflation or defensiveness onto me. I'm just explaining how ToM works and letting Carl know I'm not self-deceived regarding my own ToM, and that an autistic person can have very high ToM, which contradicts the idea that the eye test is a good proxy for ToM.

I'm sure autistics lack affective empathy (I know I do), but this doesn't necessarily hamper ToM. 

Edited 16 hours ago by Joshe

[Carl-Richard]

@Joshe The test seems like a good way to indicate whether somebody has autism. Whether or not it captures all aspects of theory of mind is a different story.

Different aspects of ToM is interesting. Autistics may be good at stacking up past experiences (concrete things) and using them to predict responses (a spectrumy friend of mine has essentially photographic memory and probably does the same). But the high-paced, highly contextual and subtle momentary changes in states, that may require indeed affective attunement and inference based on emotional expressions, that's what autistic people struggle with. And if you're meeting for example a new person, that might be all you have. Like if you had met a friend of mine from high school, what would you have to go by to understand their mind? What if they don't talk?

Edited 11 hours ago by Carl-Richard

[Basman]

On 1/24/2026 at 2:55 AM, Joshe said:

I scored below average: 22

It seems the premise is autistic people have low Theory of Mind, but this isn't true in my experience. I have very high TOM. 

The test seems to measure shallow, fluent social processing, which I think is very different from TOM.

Interesting idea though. 

Another critique is that NT people probably designed the whole test. It's basically asking "do you label this the same way NT people would label it". 

Walks like cope, swims like cope, and quacks like cope.

Probably cope.

[Joshe]

5 hours ago, Carl-Richard said:

@Joshe The test seems like a good way to indicate whether somebody likely has autism. Whether or not it captures all aspects of theory of mind is a different story.

I mean, it might be, but something just doesn't seem right about it.

My issue with it is it was developed to measure ToM - and it is a very creative idea - but my own experience with ToM shows how absurd that is. 

5 hours ago, Carl-Richard said:

And if you're meeting for example a new person, that might be all you have. Like if you had met a friend of mine from high school at a random place, what would you have to go by to understand their mind? What if they don't talk?

The gestalt works in real-time too.

When I sit down at a poker table full of strangers (some of them don't talk much), I can size them up faster than anyone else I know, and it all arrives via recognition, not inference. Like, I could know if someone is conflict-avoidant, externally validated, risk-averse, and projects confidence to mask insecurity all within 1-5 mins, and all without words, labels, or memories, and I can have an unconscious confidence score on that read which goes up or down as data comes in. I think this is much closer to ToM than affective attunement is.

That said, I wouldn't be surprised if affective attunement is a strong marker for autism, but the test is using ToM as a proxy, which isn't right. 

Edited 10 hours ago by Joshe

[Carl-Richard]

4 hours ago, Joshe said:

The gestalt works in real-time too.

When I sit down at a poker table full of strangers (some of them don't talk much), I can size them up faster than anyone else I know, and it all arrives via recognition, not inference. Like, I could know if someone is conflict-avoidant, externally validated, risk-averse, and projects confidence to mask insecurity all within 1-5 mins, and all without words, labels, or memories, and I can have an unconscious confidence score on that read which goes up or down as data comes in. I think this is much closer to ToM than affective attunement is.

The thing about Poker is it's one setting which you can practice and specialize in. But social games include many different settings where specialization might not be possible and all you can rely on is fast-and-frugal heuristics.

 

4 hours ago, Joshe said:

That said, I wouldn't be surprised if affective attunement is a strong marker for autism, but the test is using ToM as a proxy, which isn't right. 

It is right, because affective attuning is a central means by which you come to access soneone's mind. The thing you're doing is you're expecting a fully comprehensive measure of ToM. You won't ever get that in a test. Science is almost always "by proxy", especially in psychology. You create a measure and then it is used to point to an idea or variable. Very rarely do you get an essentially 100% 1-to-1 relationship between the measure and the variable, and it's basically not a thing when dealing with complex concepts like ToM.