WEBVTT - Facial Recognition Machinery, Part 2 0:00:03.000 --> 0:00:04.840 Welcome to Stuff to Blow Your Mind, a production of 0:00:04.880 --> 0:00:13.680 I Heart Radios How Stuff Works. Hey you, welcome to 0:00:13.720 --> 0:00:16.040 Stuff to Blow your Mind. My name is Robert Lamb 0:00:16.160 --> 0:00:19.200 and I'm Joe McCormick, and we're back today with part 0:00:19.200 --> 0:00:23.480 two of our exploration of facial recognition machinery. Last time, 0:00:23.520 --> 0:00:26.920 of course, we talked about uh some tech biz world 0:00:27.120 --> 0:00:30.200 stuff that that may be highly relevant to your life, 0:00:30.360 --> 0:00:33.519 especially in the near future. We talked about an artificial 0:00:33.560 --> 0:00:36.200 intelligence company that was recently profiled in the New York 0:00:36.240 --> 0:00:40.240 Times as uh selling a service to law enforcement that 0:00:40.280 --> 0:00:43.319 would use while they stole your face right off your 0:00:43.320 --> 0:00:46.240 head and scraped it from the Internet, and now they're 0:00:46.280 --> 0:00:49.199 selling that to law enforcement as a tool supposedly for 0:00:49.360 --> 0:00:53.080 identifying people with a high rate of accuracy, uh, linking 0:00:53.120 --> 0:00:56.440 your anonymous face to all of the digital information that's 0:00:56.440 --> 0:00:59.840 out there about you. Long story short, we're all boned 0:01:00.840 --> 0:01:04.240 unless we, uh, you know, we actually you know, put 0:01:04.280 --> 0:01:08.160 into place various laws and protections that that either keep 0:01:08.280 --> 0:01:12.399 these technologies from fully coming online or make sure that 0:01:12.480 --> 0:01:17.880 they are restricted from destroying the privacy at least of 0:01:18.200 --> 0:01:20.560 you know, private individuals. And we'll talk more about that 0:01:20.600 --> 0:01:23.040 aspect of the subject. I think in the next episode, 0:01:23.040 --> 0:01:26.399 when we get more into the modern technology today, we 0:01:26.440 --> 0:01:31.160 wanted to focus more on the biological world of facial recognition. 0:01:31.800 --> 0:01:35.000 What's been learned in in recent decades in psychology and 0:01:35.080 --> 0:01:39.520 neuroscience about the recognition of faces by animals like us, Right, 0:01:39.560 --> 0:01:42.640 Because ultimately, I guess the counter argument is, Hey, we're 0:01:42.640 --> 0:01:45.880 just trying to teach computers and phones to do what 0:01:46.560 --> 0:01:49.200 humans do and what animals can do, and that is 0:01:49.240 --> 0:01:52.320 look at a face and respond to it, identify the 0:01:52.320 --> 0:01:55.200 individual behind that face. Right. And while that might be 0:01:55.240 --> 0:01:58.240 something that scary as a capability for the machine to have, 0:01:58.760 --> 0:02:01.960 it's something that's uh part of our survival history and 0:02:02.120 --> 0:02:04.520 an important part of our social lives. Oh yeah, because 0:02:04.560 --> 0:02:08.720 we go around every day, we're walking around, we're driving where, 0:02:08.800 --> 0:02:11.520 you know, in an exercise class, etcetera. And our brain 0:02:11.600 --> 0:02:15.000 is engaging in that exercise. Of which human is that? 0:02:15.080 --> 0:02:16.600 Do I know that human weight? I think I know 0:02:16.639 --> 0:02:19.919 that human weight? Further analysis reveals I do not. It's 0:02:19.919 --> 0:02:23.360 a really funny thing actually, when you notice how much 0:02:23.440 --> 0:02:26.600 your brain is just going who's that is that? Yeah, 0:02:28.080 --> 0:02:31.079 it's like a it's a ridiculous amount of your your 0:02:31.120 --> 0:02:35.280 processing power is eaten up with that narrative. Yeah. In fact, 0:02:35.360 --> 0:02:37.799 I mean it makes that That's why solitude is sometimes nice, 0:02:37.800 --> 0:02:41.720 because it just removes us from that exercise. Now, granted, 0:02:41.840 --> 0:02:43.680 you could have too much solitude, and I guess maybe 0:02:43.680 --> 0:02:45.600 the brain ends up using all that energy that it 0:02:45.720 --> 0:02:48.960 would use towards identifying or trying to identify strangers towards 0:02:49.760 --> 0:02:53.160 new and destructive things. But yeah, for the most part, 0:02:53.200 --> 0:02:55.799 it is an important part of making your way around 0:02:55.919 --> 0:02:58.200 human society. Now, at the risk of sounding like I'm 0:02:58.240 --> 0:03:02.040 making excuses, I gotta say, hand, this is a complicated subject. 0:03:02.240 --> 0:03:04.240 This is one of those where the deeper I dug 0:03:04.240 --> 0:03:06.360 into it, the more and more it just seemed like 0:03:06.400 --> 0:03:08.960 we were missing out on So I mean, I think 0:03:09.000 --> 0:03:11.399 we just have to preface this by saying it's impossible 0:03:11.440 --> 0:03:14.760 for us to do the whole subject of biological facial 0:03:14.760 --> 0:03:17.640 recognition justice in this episode. We'll do our best in 0:03:17.680 --> 0:03:20.720 a reasonable length of time. Yeah, I find like it's 0:03:20.720 --> 0:03:23.680 easy to sort of glimpse the complexity of it when 0:03:23.680 --> 0:03:26.800 you engage in exercises like say, attempting to draw a 0:03:26.840 --> 0:03:30.240 face that you know. And granted that involves artistic ability 0:03:30.240 --> 0:03:34.000 and talent. That is, sometimes the talented is underdeveloped, but still, 0:03:34.040 --> 0:03:37.320 like even I find myself without having that talent, just 0:03:37.360 --> 0:03:40.520 even the mental exercise of them trying to figure out, Okay, 0:03:40.760 --> 0:03:43.440 if I was to draw Joe's face, Wait, what does 0:03:43.560 --> 0:03:45.640 Joe look like? Again? Okay, I have to form the 0:03:45.680 --> 0:03:48.160 picture in my mind and then I have to then 0:03:48.160 --> 0:03:49.800 I second guess it. I'm like, is that really what 0:03:49.880 --> 0:03:52.320 Joe looks like? Or it's it's even it's even harder 0:03:52.360 --> 0:03:54.960 if I'm not physically in the room with that individual 0:03:55.000 --> 0:04:00.520 to really have horns and pointed teeth like that. Yeah, 0:04:00.720 --> 0:04:03.240 So that's that's That's one side, but also just the 0:04:03.240 --> 0:04:06.640 the idea of recalling faces like and granted we're dragging 0:04:06.640 --> 0:04:09.520 in the complexity of of memory when we're doing that, 0:04:09.880 --> 0:04:13.880 but I think it also hints that the it hints 0:04:13.920 --> 0:04:17.359 at how difficult this is to really unwrap what happens 0:04:17.360 --> 0:04:20.599 when we look at another face and identify it much 0:04:20.640 --> 0:04:23.320 less when we recall it from memory. Now, we've discussed 0:04:23.320 --> 0:04:25.880 face perception in the brain before, for example, in our 0:04:25.920 --> 0:04:28.919 episodes on face blindness and in an episode called the 0:04:28.960 --> 0:04:32.719 Doppelganger Network h And in these previous episodes, something that 0:04:32.760 --> 0:04:36.080 we definitely talked about was the history of how our 0:04:36.200 --> 0:04:40.480 understanding of facial recognition in the brain was illuminated by 0:04:40.520 --> 0:04:45.320 studying cases of people with with malfunctions of facial recognition 0:04:45.360 --> 0:04:48.640 in one way or another. Uh, primarily with the condition 0:04:48.680 --> 0:04:51.440 that we talked about in the face blindness episode. It 0:04:51.560 --> 0:04:55.120 is known as face blindness or prosopagnosia, which is a 0:04:55.160 --> 0:04:57.960 condition with a somewhat misleading name if you go with 0:04:58.040 --> 0:05:02.320 face blindness, because people with face blindness, I think it 0:05:02.320 --> 0:05:05.400 would be best explained by saying they actually see face 0:05:05.560 --> 0:05:08.880 is just fine. The real issue is that people with 0:05:08.960 --> 0:05:14.120 this condition have difficulty recognizing faces, not seeing them right. Like. 0:05:14.120 --> 0:05:15.880 One example I always come back to, and I think 0:05:15.880 --> 0:05:17.640 I've probably brought this up in the show before, is 0:05:17.680 --> 0:05:21.560 there's there's an excellent episode of that that television series 0:05:21.600 --> 0:05:25.400 Hannibal about Hannibal Lecter, in which there's a character that 0:05:25.520 --> 0:05:28.920 also has face blindness, and when they behold Hannibal Lecter 0:05:29.000 --> 0:05:30.960 and a key scene, all they see is like a 0:05:31.400 --> 0:05:35.520 featureless flesh mask because it's like they can't see the 0:05:35.560 --> 0:05:38.440 face at all. That is not based on any of 0:05:38.440 --> 0:05:40.560 the material we've looked at in an accounts that we've 0:05:40.560 --> 0:05:43.480 read that is not what face blindness is. It sounds 0:05:43.520 --> 0:05:46.320 like face blindness. The experience of face blindness is more 0:05:46.320 --> 0:05:49.920 akin to say, when I look at some vegetation and 0:05:50.080 --> 0:05:52.320 I asked myself, is that poison ivy? I know, I've 0:05:52.320 --> 0:05:54.080 looked at a picture of poison ivy. I'm not sure 0:05:54.080 --> 0:05:57.000 if that's poison ivy or not. It's not like I 0:05:57.040 --> 0:06:01.400 don't see a plant. I just cannot identify it compared 0:06:01.440 --> 0:06:04.960 to other plants of similar form and function, And therefore 0:06:05.000 --> 0:06:07.680 I have to fall back on on Okay, well, I'm 0:06:07.680 --> 0:06:09.640 gonna try and remember what are the what are the features? 0:06:09.680 --> 0:06:11.760 Three leaves? Let it be how many leaves does this have? 0:06:11.839 --> 0:06:13.320 And I started to have to gauge in a more 0:06:13.680 --> 0:06:16.720 and a different kind of cognitive exercise to try and 0:06:16.760 --> 0:06:19.159 make a positive identification. Yeah, I mean, I think it 0:06:19.279 --> 0:06:22.400 might be even more complicated a task than that. It's 0:06:22.440 --> 0:06:25.640 like the people who have typical powers of facial recognition 0:06:26.240 --> 0:06:30.200 don't even recognize what a superpower this is that comes effortlessly. 0:06:30.720 --> 0:06:33.280 The point of comparison I've used before, and I think 0:06:33.320 --> 0:06:35.560 I heard back from some people after this episode saying 0:06:35.560 --> 0:06:38.200 it was a good one. Was the idea of holly bushes. Like, 0:06:38.200 --> 0:06:40.800 if you look at one holly bush, you can see 0:06:40.839 --> 0:06:42.839 it just fine. You can note all the colors and 0:06:42.880 --> 0:06:45.480 the shapes and all that. But imagine you're walking down 0:06:45.560 --> 0:06:47.400 the street and you happen to pass by a place 0:06:47.440 --> 0:06:50.440 where that same holly bush you looked at earlier has 0:06:50.440 --> 0:06:53.080 been like dug up and replaced somewhere else. Would you 0:06:53.160 --> 0:06:56.280 notice it was the same bush? I mean, it looks 0:06:56.279 --> 0:07:00.080 like just another bush, right, unless you were engaging a 0:07:00.120 --> 0:07:04.120 far more tedious exercise of like counting the branches on 0:07:04.200 --> 0:07:06.600 the first bush, you know, really getting in there, or 0:07:06.800 --> 0:07:09.600 you know, marking it with a sharpie, that sort of thing, exactly. Yeah, 0:07:09.600 --> 0:07:13.800 because our brains are not specially wired to casually notice 0:07:13.840 --> 0:07:17.480 and remember minor visual differences in individual plants of the 0:07:17.520 --> 0:07:22.160 same species. But it appears that typical human brains are 0:07:22.320 --> 0:07:26.560 specially wired to notice and remember minor visual differences in 0:07:26.640 --> 0:07:30.800 the hundreds of honestly pretty similar oblong orbs of meat 0:07:30.840 --> 0:07:34.080 and teeth that we interact with every day. Yeah, I mean, 0:07:34.080 --> 0:07:36.800 because a lot of faces are similar, you know. And uh, 0:07:36.840 --> 0:07:39.120 and and that's often where we get that initial like 0:07:39.240 --> 0:07:41.920 miss characterization, where we glance and we think we see 0:07:41.960 --> 0:07:44.600 somebody we know, but then we realize we don't. And 0:07:44.600 --> 0:07:48.000 occasionally you'll get that kind of like triple tech moment 0:07:48.080 --> 0:07:50.640 where somebody, Oh, at first glance, it seems right, and 0:07:50.680 --> 0:07:52.960 then a second glance it seems almost right, And then 0:07:52.960 --> 0:07:56.080 you always know there's just a very similar looking um 0:07:56.120 --> 0:07:58.840 person to someone that you know, someone I've encountered before. 0:07:58.840 --> 0:08:00.640 But this is in fact a strange Do you have 0:08:00.760 --> 0:08:04.960 that one person who you see doppelgangers of all the time, 0:08:05.520 --> 0:08:09.320 like one specific friend or celebrity that you always think 0:08:09.400 --> 0:08:13.360 you see somewhere? I guess. I mean there are certain 0:08:13.640 --> 0:08:15.920 you know there, there's certain looks that are you know, 0:08:16.200 --> 0:08:20.720 that are common, certain styles addressing that are common. Um, 0:08:20.760 --> 0:08:22.480 I've got a very weird one. Do you want to 0:08:22.520 --> 0:08:26.200 hear us? Hear it? Okay? So for some reason I 0:08:26.560 --> 0:08:30.960 keep thinking that I see the American UH physician and 0:08:31.000 --> 0:08:35.439 geneticist Francis Collins everywhere, the guy who worked on the 0:08:35.520 --> 0:08:38.960 Human Genome project. Seen a few pictures. I've never met him. 0:08:38.960 --> 0:08:42.160 I've just seen a few pictures of him around UH, 0:08:42.200 --> 0:08:46.160 and I see like basically an older white guy with 0:08:46.200 --> 0:08:49.160 a mustache and glasses, And I think, is that Francis Collins. 0:08:49.679 --> 0:08:54.400 I don't know why interesting. I mean, I find it. 0:08:55.080 --> 0:08:57.160 There are people that I'm on like heightened alertness for 0:08:57.880 --> 0:09:01.400 mainly like for instance, your us. You know, like I 0:09:01.400 --> 0:09:04.120 think this is true of everyone for the most part. 0:09:04.200 --> 0:09:06.240 You don't want to run into your boss. That's say, 0:09:06.240 --> 0:09:08.559 the grocery store, because the grocery store, first of all, 0:09:08.800 --> 0:09:11.120 is an awkward place to run into anybody. I just 0:09:11.200 --> 0:09:13.480 ran into a coworker or the grocery store the other day, 0:09:13.640 --> 0:09:16.800 the worst, great coworker. Nothing against this person at all, 0:09:16.840 --> 0:09:18.760 but when I saw them, I was like, ah, yeah, 0:09:18.840 --> 0:09:22.400 because it's like, let's have this awkward uh exchange now, 0:09:23.000 --> 0:09:25.560 and let's do it again in one and a half 0:09:25.600 --> 0:09:28.840 minutes on the next aisle, and then let's do it 0:09:28.880 --> 0:09:32.440 another time. And it's just it's a terrible exercise, Andy, 0:09:32.600 --> 0:09:35.439 And then you know your boss. It brings in additional complexities, 0:09:35.440 --> 0:09:37.840 no matter how wonderful your boss happens to be. So 0:09:37.920 --> 0:09:41.000 it's like it results at least in my weird mind 0:09:41.200 --> 0:09:43.959 of you know, me being like hyper alert, like are 0:09:44.040 --> 0:09:46.240 they is? You know, is the is my boss? Here? 0:09:46.280 --> 0:09:48.440 Is a coworker here? I must hide if I see 0:09:48.480 --> 0:09:50.800 them because I want to spare us both the awkwardness 0:09:50.840 --> 0:09:54.080 of running into each other. And that's just around the office, right. 0:09:55.120 --> 0:09:57.839 So yeah, so telling one human apart from another is 0:09:57.880 --> 0:10:01.440 obviously a relevant survivals gill. So it's something that our 0:10:01.480 --> 0:10:05.720 primate brains developed a unique capacity for, especially by means 0:10:05.760 --> 0:10:08.960 of recognizing the visual features of the face, and in 0:10:09.080 --> 0:10:13.559 people who have face blindness or prosopagnosia, this recognition capacity 0:10:13.640 --> 0:10:15.880 has broken down, often due to some kind of brain 0:10:15.920 --> 0:10:20.120 injury or lesion, and uh to the person with severe prosopagnosia, 0:10:20.440 --> 0:10:23.839 human faces can present a problem similar to what we're 0:10:23.840 --> 0:10:25.880 talking about earlier, like looking at a plant you know, 0:10:26.040 --> 0:10:29.559 or looking at similar holly bushes the person with prosopagnosia 0:10:29.600 --> 0:10:32.200 can see. The person can see the face, but the 0:10:32.240 --> 0:10:35.920 faces don't really distinguish themselves from one another in memory 0:10:35.960 --> 0:10:39.400 because of damage to the special recognition power. And as 0:10:39.440 --> 0:10:42.239 a side note, there's another interesting fact about face blindness, 0:10:42.520 --> 0:10:45.720 which is that people who have it also very often, 0:10:45.760 --> 0:10:48.720 not always, but pretty often have a kind of location 0:10:48.840 --> 0:10:52.720 blindness as well. They can become easily lost because they 0:10:52.760 --> 0:10:56.760 don't remember visual characteristics of even familiar locations like the 0:10:56.800 --> 0:10:59.760 building where they work or their house. Yes, I seem 0:10:59.800 --> 0:11:02.880 to have call Oliver Sacks writing about this, um totally. Yes, 0:11:03.080 --> 0:11:06.840 the late author and psychologist who who had face blindness 0:11:06.840 --> 0:11:10.319 as well. Yeah, yeah he did. He wrote about it autobiographically, 0:11:10.400 --> 0:11:11.959 I believe, in a piece for The New Yorker that 0:11:12.120 --> 0:11:13.840 was really good that we talked about in our face 0:11:13.840 --> 0:11:19.360 blindness episode. Um So. Historically, autopsies on the brains of 0:11:19.440 --> 0:11:24.559 people with acquired prosopagnosia were very informative because these brains 0:11:24.600 --> 0:11:29.880 almost always showed lesions on the bottom of a brain 0:11:30.000 --> 0:11:33.960 region known as the occipito temporal cortex. And if you 0:11:33.960 --> 0:11:37.120 want to picture this, it's kind of the rear middle 0:11:37.200 --> 0:11:39.960 underside of the brain, so you think, go down from 0:11:39.960 --> 0:11:42.760 your temples and then back a little bit and on 0:11:42.800 --> 0:11:45.679 the underside of the brain. Uh. This region of the 0:11:45.720 --> 0:11:48.640 brain is also known as the fusiform gyrus, and brain 0:11:48.720 --> 0:11:51.319 imaging like CT scans and m r I on living 0:11:51.360 --> 0:11:55.079 people also confirmed this correlation. Lesions on the fuse form 0:11:55.160 --> 0:11:58.640 gyrus on the underside of the occipito temporal cortex were 0:11:58.679 --> 0:12:03.160 commonly associated with the inability to recognize faces. Meanwhile, real 0:12:03.240 --> 0:12:06.439 time brain imaging like fm r I has also associated 0:12:06.480 --> 0:12:10.000 face processing with increased activity in this part of the brain. 0:12:10.040 --> 0:12:13.000 So if you look at a human face, your fusiform 0:12:13.080 --> 0:12:16.360 gyros tends to get more blood flow, and for that reason, 0:12:16.440 --> 0:12:18.600 this region of the brain has come to be known 0:12:18.600 --> 0:12:22.120 as the fusiform face area. Now, it's really important to 0:12:22.160 --> 0:12:25.240 note that multiple networks of the brain are involved in 0:12:25.320 --> 0:12:28.600 face perception, and we'll talk about some more studies about 0:12:28.600 --> 0:12:31.280 that as we go on, but it appears somehow the 0:12:31.280 --> 0:12:35.360 fusiform gyrus is especially important and that damage to it 0:12:35.400 --> 0:12:38.600 can tend to cause this another way, Uh, that I 0:12:38.600 --> 0:12:40.920 wanted to complicate the idea we were talking about earlier 0:12:40.920 --> 0:12:45.480 that you can usually see faces correctly with prosopagnosia, but 0:12:45.600 --> 0:12:49.040 that you have trouble recognizing. A complication to that is 0:12:49.080 --> 0:12:52.559 like one study I remember seeing video of where there 0:12:52.600 --> 0:12:55.600 was a patient who had an electrode implanted directly to 0:12:55.679 --> 0:12:58.880 stimulate his fuse form gyros, and he was awake and 0:12:58.920 --> 0:13:02.120 could talk about in real time when there was a 0:13:02.160 --> 0:13:04.680 current applied to this part of the brain. He said 0:13:04.720 --> 0:13:08.040 that his vision remained normal except for people's faces, and 0:13:08.080 --> 0:13:10.920 when the current was applied, people's faces would tend to 0:13:11.000 --> 0:13:14.960 kind of metamorphose. That like their features would appear to 0:13:15.040 --> 0:13:18.880 move around and stuff. Oh interesting, like more so than 0:13:18.920 --> 0:13:21.720 just the experience of staring at somebody's face till the 0:13:21.760 --> 0:13:25.080 information starts, you know, loses kind of consistency. Oh is 0:13:25.120 --> 0:13:28.400 that a thing you experience? Uh, yeah, to a certain extent, 0:13:28.559 --> 0:13:30.800 I mean, with even like saying a word until it 0:13:30.840 --> 0:13:33.240 loses meaning. Yeah, yeah, I mean it's kind of the 0:13:33.320 --> 0:13:35.840 effect to of of looking in a mirror too long, 0:13:35.880 --> 0:13:39.000 you know, where you're not really presented with any new data, 0:13:39.400 --> 0:13:42.080 Like you've you've absorbed all the data that is necessary 0:13:42.679 --> 0:13:47.439 to to properly react and uh and situate yourself in reality, 0:13:47.480 --> 0:13:50.480 but then you keep feeding on the same informational source, 0:13:51.600 --> 0:13:53.840 which you know, is kind of like the road to madness, 0:13:54.679 --> 0:14:00.760 especially in situations of sensory deprivation. I've certainly the experience 0:14:00.800 --> 0:14:03.199 where I stare at somebody's face, or stare at say 0:14:03.200 --> 0:14:06.679 a dog's face long enough that like it's it starts 0:14:06.720 --> 0:14:09.280 to it doesn't look any different, but it starts to 0:14:09.400 --> 0:14:12.520 decohere as like the seat of the soul, and instead 0:14:12.559 --> 0:14:16.960 becomes textures of organs. Do dogs have faces? Of course 0:14:17.000 --> 0:14:19.960 they do. I don't know, it don't really what is wrong? 0:14:20.080 --> 0:14:21.720 I don't think. I mean, I don't think of cats 0:14:21.720 --> 0:14:23.920 as having faces either. They just kind of faces. They 0:14:23.960 --> 0:14:27.200 just kind of have the fronts of heads. You know, Um, 0:14:27.400 --> 0:14:31.280 humans have faces. Where does the Cheshire Cats grin live 0:14:31.440 --> 0:14:33.880 if not on its face? Well, it's a cartoon character. 0:14:34.040 --> 0:14:37.360 Cartoon characters have faces because they are they're made in 0:14:37.800 --> 0:14:40.640 at least partially in our likeness. I've just discovered something 0:14:40.840 --> 0:14:43.640 very sinister about you. I guess a pug kind of 0:14:43.680 --> 0:14:47.320 has a face, definitely, as we've we've reread the pug 0:14:47.440 --> 0:14:49.480 enough to where it it is is close to having 0:14:49.480 --> 0:14:53.280 a faces any any dog can really claim to. Now, 0:14:53.320 --> 0:14:56.480 there's another interesting fact about biological face perception. I think 0:14:56.480 --> 0:14:59.160 I mentioned this in the last episode, but just to reiterate, 0:14:59.200 --> 0:15:03.560 the brain, it turns out processes familiar versus unfamiliar faces 0:15:03.680 --> 0:15:06.880 very differently. Like, when I face is familiar, the brain 0:15:07.000 --> 0:15:10.760 is extremely good at recognizing it accurately, even under difficult 0:15:10.840 --> 0:15:15.000 viewing conditions bad light, weird angles, partial view and all that. 0:15:15.560 --> 0:15:19.920 Less familiar faces fail to be recognized under these same conditions. 0:15:19.960 --> 0:15:22.200 So what's going on with the brain here? Well, just 0:15:22.280 --> 0:15:25.440 to reference one specific study by Sophia M. Landy and 0:15:25.440 --> 0:15:28.400 Win rich A fry wall To published in Science in 0:15:28.440 --> 0:15:32.440 two thousand seventeen, called two areas for familiar face recognition 0:15:32.440 --> 0:15:35.760 and the primate brain UH. The authors found quote familiar 0:15:35.800 --> 0:15:41.600 faces recruited two hitherto unknown face areas at anatomically conserved 0:15:41.720 --> 0:15:46.320 locations within the perinal cortex and the temporal pole. So 0:15:46.360 --> 0:15:48.880 in fMRI I, these two areas of the brain, but 0:15:49.120 --> 0:15:53.160 not the rest of the face processing network, responded dramatically 0:15:53.240 --> 0:15:56.720 to familiar faces emerging from a blur, but they didn't 0:15:56.720 --> 0:16:00.040 show any special activity when presented with unfamiliar faces. So 0:16:00.040 --> 0:16:04.160 sounds like the brain also recruits these special additional networks 0:16:04.200 --> 0:16:08.120 in addition to the regular fusiform face area for identification 0:16:08.200 --> 0:16:13.760 when it detects a more familiar face. Now, of course, historically, evolutionarily, 0:16:14.040 --> 0:16:16.920 those familiar faces would be the faces of individuals that 0:16:16.960 --> 0:16:19.200 we are, that are that are part of our society, 0:16:19.280 --> 0:16:22.760 that are part of a close knit group um or 0:16:22.800 --> 0:16:27.040 I guess potentially enemies that you've encountered physically in the past. 0:16:27.440 --> 0:16:30.280 But the modern media version of that is that we 0:16:30.320 --> 0:16:32.840 have all these additional faces as well, like all the 0:16:33.080 --> 0:16:37.400 all the actors we've memorized from watching TV and movies 0:16:37.440 --> 0:16:40.440 and surfing IMDb for example. Yeah, well, I think one 0:16:40.440 --> 0:16:43.479 thing that's important is that when a face is familiar, 0:16:43.520 --> 0:16:45.800 it tends to come with a very complex suite of 0:16:45.800 --> 0:16:49.760 emotional reactions that are implied by the face. You know, 0:16:49.840 --> 0:16:51.760 you see somebody and you know them to be an 0:16:51.760 --> 0:16:54.600 adversary or you know them to be a family member 0:16:54.680 --> 0:16:58.120 or friends that you've got all these complex emotions that 0:16:58.320 --> 0:17:02.720 come out of this emotional response called familiarity. I'd imagine 0:17:02.720 --> 0:17:06.840 the brain's response to unfamiliar faces or less familiar faces 0:17:06.840 --> 0:17:09.600 tends to be more flat, probably, right, Like there's less 0:17:09.640 --> 0:17:13.600 differentiation in the response, right, right, And there's probably a 0:17:13.600 --> 0:17:15.040 lot to be said, And this may be an area 0:17:15.040 --> 0:17:17.840 of separate study, like what happens when you encounter faces 0:17:18.359 --> 0:17:21.400 in real life that you have thus far only encountered 0:17:21.480 --> 0:17:24.159 via media, You know, I mean, it's a it's a 0:17:24.200 --> 0:17:26.720 different scenario, if for nothing else, if nothing else, the 0:17:26.920 --> 0:17:29.199 lighting and the makeup is going to be different. And 0:17:29.240 --> 0:17:32.440 they're so short. Oh that's the thing I'm surprised we've 0:17:32.480 --> 0:17:34.720 never looked into before. There's got to be research on that, 0:17:34.760 --> 0:17:37.920 Like why every you assume that movie stars are seven 0:17:37.920 --> 0:17:40.080 feet tall until you see them in person. Well, I 0:17:40.080 --> 0:17:43.840 think it's because they're standing on apple boxes a lot 0:17:43.880 --> 0:17:47.680 of the times. Um. Now, there's another interesting debate in 0:17:47.720 --> 0:17:50.440 the history of face processing research that we've discussed on 0:17:50.480 --> 0:17:52.760 the show once before. I wasn't able to find a 0:17:52.800 --> 0:17:55.160 resolution here, but but it is sort of a dispute 0:17:55.760 --> 0:17:59.399 among these researchers. So to look at a foundational kind 0:17:59.440 --> 0:18:02.440 of study here, there was a study published in Nature 0:18:02.440 --> 0:18:06.240 Neuroscience in two thousand by Isabel Gauthier. At all in 0:18:06.280 --> 0:18:09.600 the background here was that research had already shown that 0:18:09.960 --> 0:18:12.959 people who had been trained to have an expertise in 0:18:13.280 --> 0:18:17.600 previously unfamiliar objects called greebles will come back to them 0:18:17.600 --> 0:18:21.000 in a second. People who had that expertise would recruit 0:18:21.119 --> 0:18:23.879 parts of the brain that are usually used in the 0:18:23.920 --> 0:18:27.560 processing of faces, such as a fuse form gyrus and 0:18:27.600 --> 0:18:31.680 the occipital lobe. And so greebles are these weird little 0:18:31.760 --> 0:18:35.800 chess piece like objects with abstract kind of goblin ears 0:18:35.840 --> 0:18:39.000 and spikes and stuff. I really like the greebels, you know, 0:18:39.040 --> 0:18:44.159 I was reading about greeple's and uh, greebel's also another 0:18:44.280 --> 0:18:47.320 definition are the and it's pretty closely related, I guess, 0:18:47.600 --> 0:18:49.840 are also the little bits of plastic glued to the 0:18:49.880 --> 0:18:53.200 tops of objects to make them seem more complex. Star Destroyer, Yeah, 0:18:53.240 --> 0:18:57.320 the star Destroyer, I guess, so what the Death Star itself? 0:18:57.440 --> 0:19:01.280 Or a great example of the background on Mystery Sense 0:19:01.320 --> 0:19:05.640 Theater three thousand, at least for a number of seasons there, 0:19:06.040 --> 0:19:08.760 you could if you look closely, you could recognize the 0:19:08.800 --> 0:19:11.800 everyday objects that we're serving as Greebels, such as I 0:19:11.800 --> 0:19:15.240 think a millennium falcon toy was back there as a Greebel. 0:19:15.320 --> 0:19:18.280 But yeah, the more junk that is glued to it, 0:19:19.040 --> 0:19:20.840 the more it looks like it has a lot of 0:19:20.880 --> 0:19:24.000 surface complexity to it. That The board cube is another 0:19:24.000 --> 0:19:26.720 example of this. It's not just a cube, which of 0:19:26.760 --> 0:19:30.280 course it's a model ship, but then they have all 0:19:30.320 --> 0:19:32.119 these little bits on the outside of it and it 0:19:32.119 --> 0:19:34.920 looks even more complicated. Yeah, it's got texture that gives 0:19:34.920 --> 0:19:37.760 it the illusion of functionality. In fact, it's just a 0:19:37.880 --> 0:19:41.119 It's just a surface that hides nothing real behind it. Yes, 0:19:41.400 --> 0:19:43.200 and a similar thing would be true of the greebel's 0:19:43.280 --> 0:19:45.400 used in these studies. So they're like a little imagine 0:19:45.400 --> 0:19:48.800 a little chess piece that's just got different kinds of 0:19:48.880 --> 0:19:51.959 little spikes and features poken out of it. And so 0:19:52.000 --> 0:19:54.480 you can train people on these things and say you 0:19:54.640 --> 0:19:57.760 learn the name for this Greeble versus that greeble, and 0:19:57.760 --> 0:19:59.720 and they'll get names for you know, a group of 0:19:59.760 --> 0:20:02.760 them over time. If people train with objects like this, 0:20:02.880 --> 0:20:05.040 they can learn the names of the different Greebels. They 0:20:05.119 --> 0:20:08.680 look mostly indistinguishable if you haven't trained with them. Even 0:20:08.720 --> 0:20:10.520 though this is again these are like just made for 0:20:10.600 --> 0:20:13.880 the experiment. There's no like pre existing greebel set right, right, 0:20:13.880 --> 0:20:16.240 but you can train people right And so what previous 0:20:16.280 --> 0:20:19.119 research had found is that people who get trained on 0:20:19.160 --> 0:20:22.480 these greebles look at the Greebel's and it seems to 0:20:22.520 --> 0:20:24.600 recruit the parts of the brain that are usually used 0:20:24.600 --> 0:20:27.760 for face processing. This study from two thousand I mentioned 0:20:27.800 --> 0:20:31.879 extended this principle to other areas of visual expertise, including 0:20:32.000 --> 0:20:36.080 birds and cars. So it found that when people had 0:20:36.119 --> 0:20:41.240 acquired an expertise for birds and cars, the brain recruited 0:20:41.320 --> 0:20:44.920 more of the face processing associated networks of the brain, 0:20:45.359 --> 0:20:48.600 such as the fusiform gyros when looking at the objects 0:20:48.600 --> 0:20:52.119 they were experts in. Interesting, okay, and so at some 0:20:52.160 --> 0:20:54.840 points this two thousand study has been used to argue 0:20:54.880 --> 0:20:57.800 that maybe the fusiform face area of the brain is 0:20:57.880 --> 0:21:01.520 more of a visual expertise center than a face center. 0:21:02.320 --> 0:21:04.280 But I think there's also a lot of evidence that's 0:21:04.320 --> 0:21:06.199 going the other way that it has a natural and 0:21:06.240 --> 0:21:09.560 somewhat dedicated role in face perception. Uh That this other 0:21:09.640 --> 0:21:13.120 side saying that it's naturally dedicated to faces is known 0:21:13.160 --> 0:21:17.400 as the domain specificity hypothesis. Uh. So there's stuff going 0:21:17.440 --> 0:21:19.520 back and forth, but just decide. Another one that I 0:21:19.600 --> 0:21:23.080 thought was an interesting follow up to that two thousand study. 0:21:23.160 --> 0:21:27.040 This one was by Yaoda Zoo uh called Revisiting the 0:21:27.160 --> 0:21:30.320 Role of the fusiform Face Area and Visual Expertise, published 0:21:30.320 --> 0:21:34.080 in Cerebral Cortex in two thousand five. It followed up 0:21:34.080 --> 0:21:36.679 from the two thousand study about birds and cars asking 0:21:36.680 --> 0:21:40.200 a reasonable question. The author here says, Okay, if people 0:21:40.240 --> 0:21:43.720 with expertise and birds and cars show increased activation of 0:21:43.760 --> 0:21:45.800 the f f A when they look at birds and 0:21:45.840 --> 0:21:49.639 cars specifically, what if this is quote due to experts 0:21:49.680 --> 0:21:54.200 taking advantage of the faceness of the stimuli. After all, 0:21:54.560 --> 0:21:58.959