until we see faces: recognition
7 Dec 2010 01:59 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
kaigouI've been considering this for several weeks now, but coming at it rather obliquely in terms of posting about it, and I think I may have been unclear in a previous post about familiarity-of-faces. I had followed that up with a second post (now deleted, which in hindsight was rather stupid of me to let my temper with myself get the best of me, but anyway), and I think some of the frustration I've been feeling recently (with myself, that is) can be traced to the lack of context. Part of that context is in providing a better definition of what I mean when I talk about recognition.
First off, here's the definition of faceblindness, from Wiki:
...but I get the impression this is not the same as something else -- the term of which, I'm not sure -- in which faces are simply, well, not-there. It's a situation in which someone doesn't register facial expressions, or can't interpret the expressions seen; that seems to be related to empathy-issues. (This is not an issue for me, but I'm mentioning it because the two -- face-recognition versus facial-based-empathy -- seem to get conflated, sometimes.)
Here's a demonstration of how I understand this recognition to work. Let's say you know you're about to walk into a room where there's a group of people. Some of whom you might happen to know, others you expect you won't know, and at least one person you do know, who you're expecting to meet there.
Let's say that friend of yours is a medium-height woman who loves the color green (because that's easy enough to remember, as a stand-in for "know her face really well"). All you need to do is find her, and once you have, come back to this page (where "come back to this page" equates with "sit down in a one-on-one visit with your friend").
Go see.
The ability to skip past a clump of unfamiliar faces and pinpoint just the one you want works in tandem with something called heuristics. There are plenty of explanations online about it, but one of the most equitable I've found is over at Ask A Korean. If you're curious, I highly recommend you read his posts on the topic, but here's the crux of the matter:
As AskaKorean explains, it's an evolutionary tool -- if you know sabretooth tigers are bad, and you see something that's big, yellow, and has large teeth, you don't stop to query all the rest of its features. You throw the big rock and run like hell: you're using heuristics to make a snap decision based on a few prominent features.
This goes hand-in-hand with identifying "who we know" in a crowd. For some, it seems they consider the rest of the faces to blend into a sort of blur. Here's the question to ask yourself about the previous link you just followed: who else did you notice? Do you remember anything at all about them? Did you register enough to know they weren't the woman-in-green, or did you register gender, ethnicity, height, or anything past that point?
The face we recognize -- that we know beyond the short-cut level of heuristics -- stands out, while the rest of the faces are indiscriminate. That phrase about "all ___ look alike" is really hiding a longer concept: "all ___ who I do not know look alike".
Yes, there's a lot more to unpack in that concept -- and AskAKorean does a pretty good job of trying to approach it fairly (to see that it has value, but that there is also an inherent danger in such stock/snap decisions). I'll come back around to the question of stock/snap group-trait decision-making, but for now, I'm mentioning only because it has a role in the process of recognizing faces.
When you can't recognize faces, you end up stuck halfway through that heuristic shortcut. I've never been able to see all faces as "looking alike," when I think over how I process meeting people. Dismissing faces as "looking alike" (that is, "looking like they belong to a group of people that I don't need to study carefully because I can say off-the-bat that they're not also in the 'familiar' category") means walking right past people I just might know. As I entered adulthood, I got tired of offending friends that way, and somewhere along the way I realized it was better to study all faces -- just to be sure -- rather than write off a bunch of faces as "automatically unfamiliar" for any reason.
Heuristics is still operating at some level, but it's a considerably reduced level, from the way it seems to me. Or maybe I should say, I'm slower. It has to be a conscious choice to disregard, based on what I can recall (which sometimes isn't that much) of the face I'm trying to find. If I can't recall details with any precision -- and nothing is jumping out as familiar -- then I can't disregard because that risks missing yet another friend in the crowd.
The keypoint about recognition is that it's almost like a switch, or so it feels to me, and it does feel as though the faces change upon the point of recognition. I don't mean as in "you don't see X or Y quality;" I mean that you see the same, but it means something to you now. Several people on the previous post mentioned the value/import of having an emotional reaction to a face/person, and I think that's the key -- when a face 'resolves' into something/someone familiar, that emotional context comes rushing in, and suddenly it's as though the face has subtly garnered more, somehow. More meaning: to you.
To demonstrate that lack-of-recognition, here's a quick exercise. Let's say I hand you these two photographs, telling you that you know someone in each of these pictures. Perhaps we're going through a box of photos, and I'm clearly assuming you know who's in the photographs, and your connections to them. You know that means it's your cue to do the recognizing, so you take a look...
First photograph.
Second photograph.
Here's what I'd bet you're going to do (or have already done): you'll look at each face carefully, and you wait for some flash of recognition. If you're really good with faces (or you recognize the stills and thus know who to look for), then you might not even experience any lag at all -- hence including two, since I figure it's a rare person who'll immediately recognize both (or a major movie buff).
At least one of those, I hope, contains faces completely unfamiliar to you: the faces mean nothing.
But if you look again...
First photograph.
Second photograph.
Hopefully, at least one of those prompted a major double-take on your part. That's the lag of recognition, and post-double-take, it's like suddenly the face carries a whole wealth of meanings that you'd been completely missing before.
That kind of recognition, I think, has little to do with heuristics -- which has everything to do, I think, with whether you take the time to study the face (or write it off as "in group that does not apply"). The recognition is a step beyond, where you take a single face and make the connection. Add in the emotional backdrop, the place and time (ie meeting a coworker in the hallway), or some other context, and you get flash of familiarity.
In other words, for the usual pattern of recognition/familiarity, we can break down the pattern like so:
The biggest key is the lag between seeing a face (and getting every cue that it should be familiar) but still lacking that reaction of emotional content. That's when I find myself staring at a friend's face, hoping I don't look completely dumbfounded, as I try to make sense of the face. I can see all the individual parts, but it's almost as though it doesn't make a comprehensible 'wholeness' because I haven't yet made the jump over to familiarity, to understand how I feel about this face.
That's the first big chunk, but I'm going to break this into parts, because there's a lot to work through and a lot to unpack, and I want to take my time considering each part carefully.
First off, here's the definition of faceblindness, from Wiki:
Prosopagnosia (sometimes known as face blindness) is a disorder of face perception where the ability to recognize faces is impaired, while the ability to recognize other objects may be relatively intact....
Few successful therapies have so far been developed for affected people, although individuals often learn to use 'piecemeal' or 'feature by feature' recognition strategies. This may involve secondary clues such as clothing, hair color, body shape, and voice. Because the face seems to function as an important identifying feature in memory, it can also be difficult for people with this condition to keep track of information about people, and socialize normally with others.
...but I get the impression this is not the same as something else -- the term of which, I'm not sure -- in which faces are simply, well, not-there. It's a situation in which someone doesn't register facial expressions, or can't interpret the expressions seen; that seems to be related to empathy-issues. (This is not an issue for me, but I'm mentioning it because the two -- face-recognition versus facial-based-empathy -- seem to get conflated, sometimes.)
Here's a demonstration of how I understand this recognition to work. Let's say you know you're about to walk into a room where there's a group of people. Some of whom you might happen to know, others you expect you won't know, and at least one person you do know, who you're expecting to meet there.
Let's say that friend of yours is a medium-height woman who loves the color green (because that's easy enough to remember, as a stand-in for "know her face really well"). All you need to do is find her, and once you have, come back to this page (where "come back to this page" equates with "sit down in a one-on-one visit with your friend").
Go see.
The ability to skip past a clump of unfamiliar faces and pinpoint just the one you want works in tandem with something called heuristics. There are plenty of explanations online about it, but one of the most equitable I've found is over at Ask A Korean. If you're curious, I highly recommend you read his posts on the topic, but here's the crux of the matter:
...heuristics is a mental shortcut. People engage in heuristics by extracting the most prominent information out of a certain situation; if people encounter a new situation that shares the same prominent information, they conclude that the new situation is the same as the previous situation. Heuristics is useful because it enables quick decision-making with little information.
As AskaKorean explains, it's an evolutionary tool -- if you know sabretooth tigers are bad, and you see something that's big, yellow, and has large teeth, you don't stop to query all the rest of its features. You throw the big rock and run like hell: you're using heuristics to make a snap decision based on a few prominent features.
This goes hand-in-hand with identifying "who we know" in a crowd. For some, it seems they consider the rest of the faces to blend into a sort of blur. Here's the question to ask yourself about the previous link you just followed: who else did you notice? Do you remember anything at all about them? Did you register enough to know they weren't the woman-in-green, or did you register gender, ethnicity, height, or anything past that point?
The face we recognize -- that we know beyond the short-cut level of heuristics -- stands out, while the rest of the faces are indiscriminate. That phrase about "all ___ look alike" is really hiding a longer concept: "all ___ who I do not know look alike".
Yes, there's a lot more to unpack in that concept -- and AskAKorean does a pretty good job of trying to approach it fairly (to see that it has value, but that there is also an inherent danger in such stock/snap decisions). I'll come back around to the question of stock/snap group-trait decision-making, but for now, I'm mentioning only because it has a role in the process of recognizing faces.
When you can't recognize faces, you end up stuck halfway through that heuristic shortcut. I've never been able to see all faces as "looking alike," when I think over how I process meeting people. Dismissing faces as "looking alike" (that is, "looking like they belong to a group of people that I don't need to study carefully because I can say off-the-bat that they're not also in the 'familiar' category") means walking right past people I just might know. As I entered adulthood, I got tired of offending friends that way, and somewhere along the way I realized it was better to study all faces -- just to be sure -- rather than write off a bunch of faces as "automatically unfamiliar" for any reason.
Heuristics is still operating at some level, but it's a considerably reduced level, from the way it seems to me. Or maybe I should say, I'm slower. It has to be a conscious choice to disregard, based on what I can recall (which sometimes isn't that much) of the face I'm trying to find. If I can't recall details with any precision -- and nothing is jumping out as familiar -- then I can't disregard because that risks missing yet another friend in the crowd.
The keypoint about recognition is that it's almost like a switch, or so it feels to me, and it does feel as though the faces change upon the point of recognition. I don't mean as in "you don't see X or Y quality;" I mean that you see the same, but it means something to you now. Several people on the previous post mentioned the value/import of having an emotional reaction to a face/person, and I think that's the key -- when a face 'resolves' into something/someone familiar, that emotional context comes rushing in, and suddenly it's as though the face has subtly garnered more, somehow. More meaning: to you.
To demonstrate that lack-of-recognition, here's a quick exercise. Let's say I hand you these two photographs, telling you that you know someone in each of these pictures. Perhaps we're going through a box of photos, and I'm clearly assuming you know who's in the photographs, and your connections to them. You know that means it's your cue to do the recognizing, so you take a look...
First photograph.
Second photograph.
Here's what I'd bet you're going to do (or have already done): you'll look at each face carefully, and you wait for some flash of recognition. If you're really good with faces (or you recognize the stills and thus know who to look for), then you might not even experience any lag at all -- hence including two, since I figure it's a rare person who'll immediately recognize both (or a major movie buff).
At least one of those, I hope, contains faces completely unfamiliar to you: the faces mean nothing.
But if you look again...
First photograph.
Second photograph.
Hopefully, at least one of those prompted a major double-take on your part. That's the lag of recognition, and post-double-take, it's like suddenly the face carries a whole wealth of meanings that you'd been completely missing before.
That kind of recognition, I think, has little to do with heuristics -- which has everything to do, I think, with whether you take the time to study the face (or write it off as "in group that does not apply"). The recognition is a step beyond, where you take a single face and make the connection. Add in the emotional backdrop, the place and time (ie meeting a coworker in the hallway), or some other context, and you get flash of familiarity.
In other words, for the usual pattern of recognition/familiarity, we can break down the pattern like so:
- Register/become aware of a large group of faces.
- Use heuristics to cross off and/or 'set aside' the largest possible non-matching groups (all men, all short people, all white people, etc).
- Of the faces remaining, use context (place/time) or features (hair, height, gestures, skin-tone, etc) to cross off the rest (has dark hair, is wearing a suit, is standing in work-hallway, etc).
- Having narrowed it down to a only a few, the brain can then handle studying each in turn with a bit more attention, and in that space, look for the emotional content (the person smiles, the person steps forward, you recognize the face, etc).
The biggest key is the lag between seeing a face (and getting every cue that it should be familiar) but still lacking that reaction of emotional content. That's when I find myself staring at a friend's face, hoping I don't look completely dumbfounded, as I try to make sense of the face. I can see all the individual parts, but it's almost as though it doesn't make a comprehensible 'wholeness' because I haven't yet made the jump over to familiarity, to understand how I feel about this face.
That's the first big chunk, but I'm going to break this into parts, because there's a lot to work through and a lot to unpack, and I want to take my time considering each part carefully.
no subject
Date: 8 Dec 2010 02:55 am (UTC)I like the new layout, btw.
I'm not great at recognizing people, but I'm not face blind, either. I'm reading your essay with great interest.
no subject
Date: 8 Dec 2010 03:17 am (UTC)All I can recall is that there are several different recognition/cognition issues with faces, but I don't know the names for them -- I just recall reading various articles that talked about them. Inability to recognize faces (in part, after delay, at all) vs inability to read faces, of which the latter I find particularly interesting but I think it... uhm, not sure, but I seem to recall reading about it in connection with children who spend far too much time isolated in orphanages (when there's not enough staff to give very young children more than the barest amount of interaction). I think that's the situation where the children start to display, as they grow up, an inability to 'read' an expression and understand that it means 'anger' or 'happiness', and in turn the children themselves display unusual or inconsistent facial expressions as related to emotion. Hrm, my memory could be faulty, but I could've sworn that's one situation where there's a possible connection between early childhood/interaction and the visual 'cues' we learn.
As for my own facial recognition... I think I've realized (especially after the discussion on the last post) that maybe the best way to put it is that I rely very heavily on context, because that's where I root narratives. If I don't have context, then I can't dig out the narrative ("this person worked across the hall" or "this person was in my english class") and I can't place the person. Bleah. People should wear name-tags.
no subject
Date: 8 Dec 2010 03:53 am (UTC)no subject
Date: 8 Dec 2010 03:58 am (UTC)Or if nothing else, at least it's another opportunity for me to serve as a bad example! I never turn those down.
no subject
Date: 9 Dec 2010 06:05 am (UTC)The other is strictly a media thing - I am very bad at identifying identical copies of people when they show up in advertising, movies, whatever (the following is a spoiler for The Empire Strikes Back, but I figure it's been long enough - the first time I saw it I had no idea who the face was that Luke saw when he cut off Vader's head in the vision sequence. I needed the book adaptation to tell me it was his own). Fortunately I only have one set of identical friends (two of triplets; the third is nonidentical) and I tell them apart by speech mannerisms (and topics of conversation) anyway.
no subject
Date: 13 Dec 2010 05:52 am (UTC)My personal take is that it's the first option -- expectation. The whole "everyone looks alike" really only (truly) applies, IMO, when we're drunk or we're not wearing our glasses. Otherwise, even on the face of it, we know it's nonsense. And even if you go for a really bizarre example like, oh, the town I used to live in that was settled by two German families (and only those two, for about the first 100 yrs of the town's existence) such that, well, you didn't need a rock to hit at least three red-heads in the surrounding five feet... you still wouldn't mistake them. Clothes, speech, height, weight, age, blah blah blah.
But if you think about the facets of what we put together to make a snap recollection -- the tone of voice, height, build, age, face shape, hair style, clothing... and say, you see three of those that are an awful lot like someone you know, then maybe your brain is expecting that the rest of the information will arrive with the message, "yes, you know this person!" Maybe to the extent of not even noticing the rest, because you've already snapped-to. I mean, if we do double-takes when we don't recognize -- and then do recognize -- someone who's in a place we don't expect, it makes sense to me that we could do double-takes of recognizing -- and then un-recognizing -- someone who's in a place we don't expect, eh?
I might not be able to recall names and always delay on whether I know someone, but I've never had trouble telling twins or triplets apart. I think that's mostly because I'm so attuned to speech patterns and mannerisms, and no two people are identical on that. There's always going to be some sort of clue. I had identical twins for dorm-mates in college, and after stumping everyone (including the twins) on how I could tell them apart, I finally had to admit the secret: one twin always walked two feet in front of the other, with her chin just a little bit down, and the second twin always walked with her shoulders back, at a slightly slower pace. As long as I saw them walk towards me, or walking into the room, I knew which was which.
(After that, they started sitting down as soon as they saw me coming. *snort* )