Sharona Pearl
๐ค SpeakerAppearances Over Time
Podcast Appearances
An interesting piece there, though, is that Brad Pitt is actually face-blind.
profoundly face-blind.
As an interesting anecdote, people who are face-blind tend to have significantly more diverse groups of friends and also tend in their romantic relationships to be attracted to people who are not conventionally attractive.
partly because if you have a friend group, and like you're saying, it can be really confusing even for non-face-blind people to know who is who, you might want to have a diverse group to make it easier to distinguish who is who.
And one of the adaptations that face-blind folks have is that if there's something unusual about how someone looks, that can make it easier for them to remember.
So while this doesn't hold true
in giant swaths of the aggregate, I think there's something to be learned from face-blind people about the kinds of judgments that we make of others.
Psychologists would agree with you.
There's something called the other race effect or the cross race effect, the CRE, and it's not just racism with an acronym.
There is an idea that faces that we have less exposure to are going to be harder to distinguish.
So that actually works for folks who are not of the race that they have a lot of exposure to, as well as folks who are of that race.
And as it happens, because of the global dominance of whiteness on television and in other ways, we tend to find it a lot easier to recognize white folks.
But it is a demonstrated phenomenon.
Essentially, your phone has a form of face recognition technology embedded in it.
And what it's doing...
is turning your face into a series of data points and measurements that it's comparing to a stored image of your face that it already has, that it has already converted into those data points.
So if those data points match, i.e.
the various distances between points on your face match, then it's decided that it's you.
That is the...
Easiest case by far, and by far the most successful version of facial recognition technology in the sense of having very few false positives, right?