FTA: "When Picard and el Kaliouby were calibrating their prototype, they were surprised to find that the average person only managed to interpret, correctly, 54 per cent of Baron-Cohen's expressions on real, non-acted faces. This suggested to them that most people - not just those with autism - could use some help sensing the mood of people they are talking to. "People are just not that good at it," says Picard. The software, by contrast, correctly identifies 64 per cent of the expressions."
I kind of wonder how they did this testing. Much of the information we get about a person's mood is from what context there is (far more than comes from what's on their faces). There are people who cry out of joy, but if I saw a picture of one of these people, I'm sure it would look like sadness to me.
Further, I don't like the sound of a future where people stop talking to other people because a light starts blinking.
However, for the purposes of autism research and development, this is good. Better than good - this is excellent. I really hope there will be more research in this area, for the purpose of helping those who can't communicate well. (I suppose that contradicts what I said above. Perhaps there is a particular scale on which to rank necessity of aid in communication?)
It makes sense that people's ability to detect true emotion is capped, though. It's useful to be able to successfully fake an emotion. Social skills are an arms race. 50-50 is somewhat lower than I'd expected in terms of the skill of the average person, but I'd be astonished (and disconcerted) if emotional detection was at 100%. Being able to lie is just too damn useful.
17/20 although I thought I had them all. Eyebrows don't move if you're faking. Also the way they returned to stone cold normality after the smile gave it away.
> Further, I don't like the sound of a future where people stop talking to other people because a light starts blinking.
I guess most people would become much better `raw' observers if they underwent a little training and got some feedback.
(I remember the study that some scientist did about clothing and the point in the monthly fertility cycle of women. After a while as a side effect the researchers became experts at guessing the point in the menstrual cycle.)
I agree, the part on the little red light telling you not to talk was disconcerting at best. I can't help but envision a slippery-slope wherein soon computers are regulating all aspects of our conversations, at which point they cease to ever be meaningful.
Open question: why do you feel computers regulating conversation is worse than computers regulating writing (from spell checking via grammar advice up to style advice)? It's not like you can't ignore either system.
Spell checking would be the equivalent of correcting pronunciation. I would hardly mind that. It's more the idea of computers controlling the actual flow and similarly the content of the conversation.
I kind of wonder how they did this testing. Much of the information we get about a person's mood is from what context there is (far more than comes from what's on their faces). There are people who cry out of joy, but if I saw a picture of one of these people, I'm sure it would look like sadness to me.
Further, I don't like the sound of a future where people stop talking to other people because a light starts blinking.
However, for the purposes of autism research and development, this is good. Better than good - this is excellent. I really hope there will be more research in this area, for the purpose of helping those who can't communicate well. (I suppose that contradicts what I said above. Perhaps there is a particular scale on which to rank necessity of aid in communication?)