This is somewhat interesting. AI tools can be helpful. However I don't know that I agree it's teaching empathy. It's teaching a verbal empathetic response, one within a specific cultural context...
This is somewhat interesting. AI tools can be helpful. However I don't know that I agree it's teaching empathy.
It's teaching a verbal empathetic response, one within a specific cultural context in which it's being created... So likely a dominant culture American neurotypical (NT) verbal empathetic response. Which is not a bad skill to have, but it isn't the feeling of empathy.
In my personal experience, and from the mouths of autistic folks, often the issue is an overwhelming amount of empathy, sometimes to the point of experiencing physical pain, and a difficulty conveying that feeling to another (usually NT) person in a "socially acceptable" way.
The closest comparison I can make, and I'm allistic so this isn't my experience please feel free to correct me autistic folks, is to take the "cringe" feeling of watching the UK Office (or Scott's Tots from the US version) or watching someone do a big flash mob proposal and get rejected, or other second hand embarrassment and think of how that feeling is strong enough a lot of people stop watching the inciting media. Maybe it even makes you a little nauseated as you imagine yourself in that position, or you recall a similar situation that happened to you. But copy/paste that onto other emotions and imagine the difficulty of responding empathetically when experiencing whole body emotions, ones that may overwhelm your sensory limits.
So that's one thing. Another is that we see differences in how NT and neurodivergent (ND) people express empathy verbally. ND people share that they often express empathy through relating the situation to a similar time they felt that way. NT people often find this to be unhelpful and unempathetic. But instead of acknowledging both ways are valid and may be effective in different situations or with different people, ND people tend to get pushed to respond in NT ways instead.
Another example might be someone who when they say "I'll pray for you" genuinely means that with all the love and care in their heart. They want the best for you, not to judge or harm, but genuinely believe that this is how they can express their care and concern and empathy. Obviously to some people that will be read as totally un-empathetic, but to others this will be the kindest thing anyone has ever said to them. But we're not targeting teaching empathy to the deeply religious or the anti-theistic.
So it's frustrating when people just want autistic people to be less autistic and be more "normal;" a standard set exclusively by allistic people.
Finally
“I'd like to take a lot of the work that I've done over the decades that's face-to-face and see how much we can translate to AI,” said Koegel. “Since kids really like the computer, we want to see if instead of just spending time on their computer or iPhone we can create a learning experience
No. Because if the goal is to "teach autistic kids to verbalize empathy by dominant societal standards" and they do need practice to do that, cutting in-person socialization will do nothing for their ability to understand the non-verbal communication that is the majority of communication (though rarely is it 93% as often touted, probably more like 75% if I'm up on my research. ) Could this be a supplemental tool, maybe for folks that want it, for kids that struggle with understanding what to verbally say, maybe. But I personally wish society would give more space to ND folks to communicate in ways that work for them. (The equivalent of not enforcing or expecting eye contact but with verbal communication too)
If the goal changed from the above, this probably wouldn't be needed, or at least would not be an autistic targeted product. (And maybe we wouldn't cut Sesame Street funding where they teach and demonstrate empathyregularly. )
I wonder how much this has to do with the differences between written and verbal communication? Learning to write in a foreign language and being able to hold a conversation in that language are...
I wonder how much this has to do with the differences between written and verbal communication? Learning to write in a foreign language and being able to hold a conversation in that language are not entirely unrelated, but they involve different skills that both require practice. Similarly, a voice call is entirely verbal and that's different from a video call, which is different from meeting in person.
I can most easily imagine a chatbot being useful for writing practice. They say it can do spoken interaction, which might be similar to a phone call. (I wouldn't know; I avoid talking to machines and have never talked to a chatbot.)
It won't really teach people how to respond to a real person on a phone call. Real people don't always give you the response you expect, output A in response to input A. Active listening skills...
It won't really teach people how to respond to a real person on a phone call. Real people don't always give you the response you expect, output A in response to input A. Active listening skills and correcting missteps in real time are part of it too.
It could be practice, even if it puts a face on the screen. But it can't substitute for real interactions with people, even digitally, or by phone, or whatever. It's a bit like doing math homework vs doing math at the grocery store. You could learn math via practicing real world skills or practicing on paper but you will need real world skills eventually.
It's just not, IMO, doing what they claim nor should it be the goal to replace real world interactions with a bot. And yes that kind of OT can be expensive but what's not expensive is family doing it with you. Teaching family might be a useful bot... But also (good) children's media does this sort of thing already.
This is very true -- writing is a technology external to the actual meat of human language -- but I don't think the difference is super relevant to many of the skills involved here. Learning the...
Learning to write in a foreign language and being able to hold a conversation in that language are not entirely unrelated, but they involve different skills that both require practice.
This is very true -- writing is a technology external to the actual meat of human language -- but I don't think the difference is super relevant to many of the skills involved here. Learning the right thing to say to seem empathetic to a neurotypical person is definitely something that can transfer between spoken and written contexts to at least some degree.
Anecdotally, as someone with a recent autism diagnosis, I do find written communication a lot easier than spoken communication, in large part because there's a bit more time to respond even in highly synchronous textual communication, which lessens my anxiety a lot. But I both have social anxiety and have never really been considered unempathetic even in spoken contexts, so I've probably not got the same flavor of symptoms as the autistic people who benefit most from this type of coaching. So, y'know, grain of salt.
Have you seen Lindsay Ellis's latest video about empathy in children's educational television? It is extraordinarily emotionally powerful, and I'd save it until a day you're feeling quite mentally...
Have you seen Lindsay Ellis's latest video about empathy in children's educational television?
It is extraordinarily emotionally powerful, and I'd save it until a day you're feeling quite mentally strong because of the very challenging content. It is the most incredible video essay I've ever seen.
I have not yet. Between vet emergencies last week and about five days of my immune system trying to drown me in my own mucus, I've not had a great two weeks of headspace to get there. But it's on...
I have not yet. Between vet emergencies last week and about five days of my immune system trying to drown me in my own mucus, I've not had a great two weeks of headspace to get there.
Really, you've nailed it here. That kind of second-hand cringe used to be unbearable for me. As an adult it's a whole lot easier but in the past, man. I feel very uncomfortable with at least the...
The closest comparison I can make, and I'm allistic so this isn't my experience please feel free to correct me autistic folks, is to take the "cringe" feeling of watching the UK Office (or Scott's Tots from the US version) or watching someone do a big flash mob proposal and get rejected, or other second hand embarrassment and think of how that feeling is strong enough a lot of people stop watching the inciting media. Maybe it even makes you a little nauseated as you imagine yourself in that position, or you recall a similar situation that happened to you. But copy/paste that onto other emotions and imagine the difficulty of responding empathetically when experiencing whole body emotions, ones that may overwhelm your sensory limits.
Really, you've nailed it here. That kind of second-hand cringe used to be unbearable for me. As an adult it's a whole lot easier but in the past, man.
I feel very uncomfortable with at least the marketing of this research, teaching ND kids on what responses are considered better to NT kids may help them somewhat in a social setting but well, as you said, that's not teaching empathy. And I'm not convinced it won't cause other issues in the long run. I was effectively taught to behave more NT and it has been extremely damaging for me, and that's unfortunately not common. If people will begin using this to make NT kids 'more empathetic' it may actually do the opposite as they learn to just reply in certain circumstances, even if the tool may be useful.
Reading over the paper itself, they fortunately decided to not make it openly accessible and explicitly mention hallucinations caused by other LLMs which I'm very relieved about. We don't need more 'development' without any breaks. They also asked about the comfort in entering/exiting conversations though, I don't think they asked for how much they'd recommend this tool themselves, which I think is more important. (also, the gender imbalance is really high... 13/2)
I don't want to exclude the possibility of teaching ND kids or adults social appropriate responses(because that can help a lot) through bots if it really helps. Though the research for this tool makes me feel like we're continuing the same trend of trying to teach autistic boys to 'behave'. I know that that sounds harsh, but it's what I keep seeing: attempts to change behaviour with a high gender imbalance.
This is somewhat interesting. AI tools can be helpful. However I don't know that I agree it's teaching empathy.
It's teaching a verbal empathetic response, one within a specific cultural context in which it's being created... So likely a dominant culture American neurotypical (NT) verbal empathetic response. Which is not a bad skill to have, but it isn't the feeling of empathy.
In my personal experience, and from the mouths of autistic folks, often the issue is an overwhelming amount of empathy, sometimes to the point of experiencing physical pain, and a difficulty conveying that feeling to another (usually NT) person in a "socially acceptable" way.
The closest comparison I can make, and I'm allistic so this isn't my experience please feel free to correct me autistic folks, is to take the "cringe" feeling of watching the UK Office (or Scott's Tots from the US version) or watching someone do a big flash mob proposal and get rejected, or other second hand embarrassment and think of how that feeling is strong enough a lot of people stop watching the inciting media. Maybe it even makes you a little nauseated as you imagine yourself in that position, or you recall a similar situation that happened to you. But copy/paste that onto other emotions and imagine the difficulty of responding empathetically when experiencing whole body emotions, ones that may overwhelm your sensory limits.
So that's one thing. Another is that we see differences in how NT and neurodivergent (ND) people express empathy verbally. ND people share that they often express empathy through relating the situation to a similar time they felt that way. NT people often find this to be unhelpful and unempathetic. But instead of acknowledging both ways are valid and may be effective in different situations or with different people, ND people tend to get pushed to respond in NT ways instead.
Another example might be someone who when they say "I'll pray for you" genuinely means that with all the love and care in their heart. They want the best for you, not to judge or harm, but genuinely believe that this is how they can express their care and concern and empathy. Obviously to some people that will be read as totally un-empathetic, but to others this will be the kindest thing anyone has ever said to them. But we're not targeting teaching empathy to the deeply religious or the anti-theistic.
So it's frustrating when people just want autistic people to be less autistic and be more "normal;" a standard set exclusively by allistic people.
Finally
No. Because if the goal is to "teach autistic kids to verbalize empathy by dominant societal standards" and they do need practice to do that, cutting in-person socialization will do nothing for their ability to understand the non-verbal communication that is the majority of communication (though rarely is it 93% as often touted, probably more like 75% if I'm up on my research. ) Could this be a supplemental tool, maybe for folks that want it, for kids that struggle with understanding what to verbally say, maybe. But I personally wish society would give more space to ND folks to communicate in ways that work for them. (The equivalent of not enforcing or expecting eye contact but with verbal communication too)
If the goal changed from the above, this probably wouldn't be needed, or at least would not be an autistic targeted product. (And maybe we wouldn't cut Sesame Street funding where they teach and demonstrate empathy regularly. )
I wonder how much this has to do with the differences between written and verbal communication? Learning to write in a foreign language and being able to hold a conversation in that language are not entirely unrelated, but they involve different skills that both require practice. Similarly, a voice call is entirely verbal and that's different from a video call, which is different from meeting in person.
I can most easily imagine a chatbot being useful for writing practice. They say it can do spoken interaction, which might be similar to a phone call. (I wouldn't know; I avoid talking to machines and have never talked to a chatbot.)
It won't really teach people how to respond to a real person on a phone call. Real people don't always give you the response you expect, output A in response to input A. Active listening skills and correcting missteps in real time are part of it too.
It could be practice, even if it puts a face on the screen. But it can't substitute for real interactions with people, even digitally, or by phone, or whatever. It's a bit like doing math homework vs doing math at the grocery store. You could learn math via practicing real world skills or practicing on paper but you will need real world skills eventually.
It's just not, IMO, doing what they claim nor should it be the goal to replace real world interactions with a bot. And yes that kind of OT can be expensive but what's not expensive is family doing it with you. Teaching family might be a useful bot... But also (good) children's media does this sort of thing already.
This is very true -- writing is a technology external to the actual meat of human language -- but I don't think the difference is super relevant to many of the skills involved here. Learning the right thing to say to seem empathetic to a neurotypical person is definitely something that can transfer between spoken and written contexts to at least some degree.
Anecdotally, as someone with a recent autism diagnosis, I do find written communication a lot easier than spoken communication, in large part because there's a bit more time to respond even in highly synchronous textual communication, which lessens my anxiety a lot. But I both have social anxiety and have never really been considered unempathetic even in spoken contexts, so I've probably not got the same flavor of symptoms as the autistic people who benefit most from this type of coaching. So, y'know, grain of salt.
Have you seen Lindsay Ellis's latest video about empathy in children's educational television?
It is extraordinarily emotionally powerful, and I'd save it until a day you're feeling quite mentally strong because of the very challenging content. It is the most incredible video essay I've ever seen.
I have not yet. Between vet emergencies last week and about five days of my immune system trying to drown me in my own mucus, I've not had a great two weeks of headspace to get there.
But it's on my list to watch, ty
Really, you've nailed it here. That kind of second-hand cringe used to be unbearable for me. As an adult it's a whole lot easier but in the past, man.
I feel very uncomfortable with at least the marketing of this research, teaching ND kids on what responses are considered better to NT kids may help them somewhat in a social setting but well, as you said, that's not teaching empathy. And I'm not convinced it won't cause other issues in the long run. I was effectively taught to behave more NT and it has been extremely damaging for me, and that's unfortunately not common. If people will begin using this to make NT kids 'more empathetic' it may actually do the opposite as they learn to just reply in certain circumstances, even if the tool may be useful.
Reading over the paper itself, they fortunately decided to not make it openly accessible and explicitly mention hallucinations caused by other LLMs which I'm very relieved about. We don't need more 'development' without any breaks. They also asked about the comfort in entering/exiting conversations though, I don't think they asked for how much they'd recommend this tool themselves, which I think is more important. (also, the gender imbalance is really high... 13/2)
I don't want to exclude the possibility of teaching ND kids or adults social appropriate responses(because that can help a lot) through bots if it really helps. Though the research for this tool makes me feel like we're continuing the same trend of trying to teach autistic boys to 'behave'. I know that that sounds harsh, but it's what I keep seeing: attempts to change behaviour with a high gender imbalance.