Welcome back to Reading the Research, where I trawl the Internet to find noteworthy research on autism and related subjects, then discuss it in brief with bits from my own life, research, and observations.
Today's article checks in with emotion-reading technology. Essentially, businesses want to skip asking their customers how they feel about their products and instead want to take pictures of their faces and get the data that way. Why is this relevant to autism? These businesses are having the same problem reading faces that autistic people do, because a facial expression does not necessarily correspond to how a person is feeling.
It's humorous to me that they're having to find this information out by trying and failing, rather than simply asking themselves whether this would work... but I guess this is another situation where neurotypical people just do things automatically, without comprehending what they're doing, how it works, and why it works.
The examples they give in this research for why this technology isn't working are contextual, and that's the big hurdle for autistic people too. We're taught in elementary school that a happy face means the person is happy, a sad face means the person is sad, and an angry face means the person is angry. Simplistic, but you'd think it would be accurate. It's not. You can't, in adult humans, look at a smiling face and presume accurately that they're happy.
Are they:
That's not going to stop marketing firms from trying to make such an AI, of course. With enough refinement, they can probably match or even exceed what I've done with my brain. Having such an AI serve marketing and businesses makes me uneasy, but perhaps such AIs could also be used to teach autistic people how to read situations and faces more accurately.
(Pst! If you like seeing the latest autism-relevant research, visit my Twitter, which has links and brief comments on studies that were interesting, but didn't get a whole Reading the Research article about them.)
Today's article checks in with emotion-reading technology. Essentially, businesses want to skip asking their customers how they feel about their products and instead want to take pictures of their faces and get the data that way. Why is this relevant to autism? These businesses are having the same problem reading faces that autistic people do, because a facial expression does not necessarily correspond to how a person is feeling.
It's humorous to me that they're having to find this information out by trying and failing, rather than simply asking themselves whether this would work... but I guess this is another situation where neurotypical people just do things automatically, without comprehending what they're doing, how it works, and why it works.
The examples they give in this research for why this technology isn't working are contextual, and that's the big hurdle for autistic people too. We're taught in elementary school that a happy face means the person is happy, a sad face means the person is sad, and an angry face means the person is angry. Simplistic, but you'd think it would be accurate. It's not. You can't, in adult humans, look at a smiling face and presume accurately that they're happy.
Are they:
- on the job at a customer service counter or receptionist desk? They're probably smiling politely rather than actually happy, because customer service jobs tend to drain your will to live but you're required to appear friendly.
- looking at something on their phone? They've probably just read something funny. This may entertain them for a few seconds, but their overall mood may not be happy.
- in a situation where sadness would be a more appropriate response, like just after finding out their pet or family member died? They're probably putting on a brave face because they don't want to publicly acknowledge or handle the pain right now.
That's not going to stop marketing firms from trying to make such an AI, of course. With enough refinement, they can probably match or even exceed what I've done with my brain. Having such an AI serve marketing and businesses makes me uneasy, but perhaps such AIs could also be used to teach autistic people how to read situations and faces more accurately.
(Pst! If you like seeing the latest autism-relevant research, visit my Twitter, which has links and brief comments on studies that were interesting, but didn't get a whole Reading the Research article about them.)
No comments:
Post a Comment