← Back to all episodes
What Is Prosopagnosia?
Dr. Brad Duchaine • 2026-08-04
Dylan Carnahan:Welcome to the Simple Questions podcast. This is your host, Dylan Carnahan. The question for this episode is, what is prosopagnosia? You will learn in this episode how prosopagnosia develops, the social impact of not being able to recognize faces, and coping strategies people use to navigate everyday life. Our guest is a leading researcher on face perception and prosopagnosia, is a professor of psychological and brain sciences at Dartmouth College, and has spent over twenty years studying how the brain processes faces. I introduce to you Dr. Brad Duchaine. Early in the morning when I am typing on my computer doing some, quote, "researching," uh, about neurological conditions, specifically around this time I had learned that I had visual snow syndrome, when I stumbled upon a condition regarding face blindness. Um, one of the things before I get into your personal experience, Dr. Duchesne, is, uh, I just have to ask, which is how do you pronounce this condition? Prosopagnosia? Nosia? What, what's the proper way? Hmm.
Dr. Brad Duchaine:Yeah, different people say different things. I don't know that there's one- ... uniform way to say it, but I say prosopagnosia.
Dylan Carnahan:Okay. Okay. Good, good thing to clarify up front. Um, but yeah, Dr. Duchesne, when did you first hear about this condition?
Dr. Brad Duchaine:Yeah, I know where I was. I, I was taking a undergraduate course on developmental psychology, and our professor just happened to mention prosopagnosia in there. I don't, I don't think we read about it. Um, and I kind of... It's easy to imagine I might not have been interested at all, but in fact it really caught my attention, and it was only years later that I ended up being able to do some work on it. But I found it a revealing window into the organization of the mind that somebody could have problems with face recognition and not have problems with other aspects of visual recognition.
Dylan Carnahan:And, you know, again, it's ... I go- when talking to guests, it's one thing to get exposed to something. It's another thing to actively contribute and be involved with it. What do you think kind of drew you to this?
Dr. Brad Duchaine:Well, so when I was in graduate school, I was very interested in, uh, psychological mechanisms that are specialized for particular purposes, and this is back in the mid-'90s, late-'90s. And so I read a lot about the two best examples of that, language and face recognition. Um, and I didn't have a way to test anybody who had prosopagnosia because nobody in my department, I was in UC Santa Barbara at the time, nobody was doing any work along those lines. But I started running experiments on typical participants that were inspired by work that had been, that had come out of the prosopagnosia literature. And, um, so then I just happened to luck into somebody with prosopagnosia, and things kind of s- things snowballed from there.
Dylan Carnahan:All right. Um, in talking about these mechanisms, how does our brain recognize people?
Dr. Brad Duchaine:Yeah, I mean, so there's a lot of different information we use to recognize people. My lab primarily focuses on sort of how we use the information that comes from the face, but of course, we're also using somebody's body type, their voice, their hair, um, the way that they move even. We use all these sorts of information to recognize people, but the most important sort of information we use for person recognition is the face, and we're able to carry out face recognition because we've got a set of, a, a network of regions primarily within the occipital and temporal lobes that calcu- that compute information about the face. And so you can think about that they're carrying out measurements of the, the different properties of the face. So imagine, you know, you're measuring how wide is the eye, how wide is the, the nose, how, you know, thick are the lips. Imagine you make measurements for 50 different characteristics of a face, and then you take those measurements that you've made, and you compare those measurements to similar measurements that you've made of other faces in the past, and if you get a close enough match, you're gonna... your visual system is gonna say, "Ah, that's Dylan there that I'm looking at."
Dylan Carnahan:And how does this, uh, you know, the, this is, like, the average experience, how does that differ in people with prosopagnosia?
Dr. Brad Duchaine:Yeah. So the people who we study with prosopagnosia, there's probably a wi- there's a w- wide variety of ways that the face processing network can malfunction, and so there's different ways that one can be prosopagnosic, and we don't have a good understanding about what those different subtypes of prosopagnosia are. But, um, some of the leading sort of, uh, hypotheses out there are that there's some people with prosopagnosia who have great difficulty even perceiving a face. So they see a face, and they're not seeing it the way that people with typical face recognition do. And we've run experiments that suggest that for some people with, with prosopagnosia, when they see an upright face, they're seeing that face, they're perceiving it with a similar sort of precision as somebody with normal face processing perceives a face that's been turned 180 degrees. So if you see an, uh, inverted face, and this is a common manipulation that we make in vision experiments, you, you know it's a face, and you can say, "Oh, those are the eyes, and there's the mouth up there," but you don't see it with the same precision that you do if you see an upright face. It's much harder to say, "Is that a man or a woman?" Or to say, "How attractive is that face?" Um, and so our best guess is that a lot of people with prosopagnosia who have perceptual problems, they're just not seeing that upright face with the same precision, um, that the rest of us do. Then there's a second group of people with prosopagnosia who probably have normal perception of faces, but they don't have normal memories for face identity information. Either they don't store that information properly, or they're not able to retrieve that information when they encounter the face again, and so they're not able to compare what they've got stored in memory to what they're currently perceiving.
Dylan Carnahan:So there's kind of the two distinctions, the one being the, a dis- almost like a distortion in perception, and then one being kind of like that indexing, um, aspect of, like, processing that's not quite functioning. Is that correct?
Dr. Brad Duchaine:Yeah. That's right, and undoubtedly this is oversimplifying things. Yeah. There's lots of ways that a system this complicated can go wrong, but sort of as a first pass, I think memory and perception is a, is a good way to think about why things aren't, why people aren't able to recognize faces in prosopagnosia.
Dylan Carnahan:As far as, like, uh, specifically the... someone having an, uh, experiencing this distortion, my understanding is that you're also looking into, there's a condition called PMO that also has implications on that?
Dr. Brad Duchaine:Yeah. So a lot of the work in my lab over the last five, six years is this condition, prosopometamorphopsia, which we abbreviate PMO. And in PMO, people are seeing clear distortions to the face, like imagine the eye might look like it's shifted down its position on the face. And in, in a lot of people with PMO, they even see active change as they look at the face. So we were talking to a guy yesterday on a Zoom call, and eyes were getting larger as he looked at the eyes. He was just looking at a photograph of a face, but the eyes were getting looked larger, and they, they got to really, um, implausible sizes as he looked at them. So that sort of perceptual deficit is different than what I'm thinking of that commonly occurs in prosopagnosia. In prosopagnosia, people are probably not seeing the face. They're not seeing a distortion. They just don't see it with the same precision that most people do, and so their representation is just not as, as good and th- thus not as useful for carrying out computations about the face. I mean, identity, we've talked about facial identity, but there's lots of things we do with facial information. We figure out how somebody's feeling by looking at their facial expression. We look at their, their face to determine whether it's a man or a woman. We figure out how attractive we think that face is. We figure out what somebody's interested in based on their eye gaze and their head position, so where they're looking and what they're actually thinking about. So there's a lot we do, and if, if you've got a, a, a low-quality representation of the appearance of the face, it's just gonna be difficult to do a lot of different tasks.
Dylan Carnahan:It seems like it would be extremely challenging. Again, I didn't even think about all the information that we glean from someone's face, right? Their interest. This is a huge, a lot amount of information, and especially, you know, um, not being able to kind of digest that, as well as, like, you bring up PMO. It's like you have this dynamic thing that's occurring that w- I would imagine would be extremely distracting.
Dr. Brad Duchaine:Absolutely. Yeah. PMO is tough for people and, and more distressing than prosopagnosia for most people anyway because, you know, imagine you're sitting there and you're looking at somebody and their, their face is changing as it look, as you look at them. It's changing in sometimes really sort of stressful ways. I mean, it's, it's an unusual appearing face. It's still a face sort of ... The face is still there in that there's eyes, nose, mouth usually. Although sometimes features disappear. Like the gentleman we were talking with yesterday, part of the lip was disappearing as he would look at it, and it was just replaced by skin. Sometimes people will have part of the face disappear, and they'll actually just see the background sort of through an aperture in the face. Um, and often the skin will be really discolored or the texture will be, um, distorted as well.
Dylan Carnahan:What do we know about the cause of face blindness? Is this ... Is there a precipitous event? Is this something that someone's born with?
Dr. Brad Duchaine:Yeah, so there's different ways that people come to have difficulty with face recognition. So we've known for a long time, so early part of the 20th century, that after brain damage, people sometimes lose their ability to recognize faces. And in fact, the, the term prosopagnosia was first coined in 1947. There was a German neurologist who studied some soldiers who had head wounds from World War II, and they lost the ability to recognize faces, and it was back in the occipital and temporal lobes, particularly the right occipital and temporal lobes. Um, so that's what we call acquired prosopagnosia. So these people had perfectly normal face recognition and then lost it due to some brain damage. Within the last 25, 30 years, we've realized that there are also a lot of people out there, far more people out there, who have difficulty recognizing faces, and yet they've never suffered any brain damage. Um, so these are individuals who failed to develop the mechanisms that are necessary for face recognition, and we call that condition developmental prosopagnosia Um, and there were the first report of developmental prosopagnosia, at least the, the, the name was first used in 1976. If you dig through the literature, you can find a few... I found a few more cases over the years, but really 1976 is when we started doing work on pros- developmental prosopagnosia. There were very few cases all the way up until time I was in grad school in the late '90s. When I got into it, I think there were three or four case studies that had been published. But we've realized that there are a awful lot of people out there who have trouble rec- really serious trouble recognizing faces. Um, and the reason that we've realized that there's lots of people out there is because of the internet. The internet's provided a way for people with prosopagnosia to get in touch with researchers who study face processing
Dylan Carnahan:Yeah. You know, in, in recalling my experiences with visual snow syndrome, right, which is a cluster of symptoms, mainly seeing static, um, it seems like there's a lot of people that, that live with, you know, some of these neurological conditions, and it's like you're standing too close to the elephant, right? That's just your normal experience. Um, and then maybe you encounter something that makes you kind of, you know, uh, inquire a little bit further. Um, so it's interesting to hear kind of how the awareness has, uh, evolved over time.
Dr. Brad Duchaine:Yeah, I mean, so where we see the developmental condition, um, or becoming well-known after the acquired condition is not uncommon, and it makes sense. If you've lived with something your entire life, it's hard to know that what you're experiencing is atypical. And the classic example of this is color deficiency or what people typically call color blindness. You know, there's lots of people who go through life never realizing that they're color deficient, that they're missing a, a particular photoreceptor on their retina, and they're seeing the world differently than others 'cause we just don't spend all that much time talking about it.
Dylan Carnahan:Yeah. I'm curious, Dr. Duchesne, you know, in your research and in your interactions with people, how can people kind of overcome or work with some of these challenges?
Dr. Brad Duchaine:Yeah. So, you know, we all, as I mentioned earlier, we all use in- sor- sorts of information to recognize people, so the face is the best, probably for most people, the best channel for figuring out who somebody is. Um, people with prosopagnosia rely on those other channels a lot more than others. Um, and oftentimes they'll do f- okay if they're in a setting where there's a small number of people they need to recognize. So imagine you're in an office with 12 people and, you know, they're different ages, men, women. Th- there's all sorts of differences. You can imagine that it... You could cover up the, the faces of the people in your office and still figure out who they are quite quickly if they walk into your office and want to talk with you. Um, and so that's sort of the life that a lot of prosopagnosics live. It's okay when it's a small number of people they're trying to recognize, but, you know, if they're at the grocery store and it can be any of a thousand people, that's when they really have great difficulty. And, you know, they'll try to use other cues to recognize people in situations like that, but they're, they're often just not sufficient to do it. And so then you have to fish a bit. You fake it in a conversation, hope that the person provides some information to clue you into who they are, but often that information isn't forthcoming.
Dylan Carnahan:Yeah. It seems like it's, it's easier in context, right? And then once that context broadens, then the difficulty increases.
Dr. Brad Duchaine:Yeah, absolutely.
Dylan Carnahan:Um- Are, you know, in talking about kind of the awareness of face blindness, um, you know, someone may be listening to this or come across content and, and it, it may resonate a little bit, Dr. Duchesne. How can people get a little more clarity on maybe if they have face blindness?
Dr. Brad Duchaine:Sure. Yeah. Um, I, I'd be surprised if there aren't people listening to this who th- are thinking, "Oh, maybe I have face blindness." And our best estimate suggests maybe one out of 100, one out of 50 people out there have enough trouble with face recognition that we would call them prosopagnosic. Um, there are tests online that one can do, so there's a version of the Cambridge Face Memory Test, which is the, one of the best tests of unfamiliar face recognition. There's one that's available, um, at Birkbeck University in London's website. Um, there are also famous face tests out there, where you can gauge sort of how well you can recognize famous faces, um, compared to other people. And w- when we run, when we test people in the lab and we want to determine whether somebody's face recognition is bad enough that we, we would consider them prosopagnosic and include them in our studies, we run both unfamiliar tests of face recognition, where they learn a set of faces and try to recognize them later on, but then we also use familiar face tests, where it's a set of famous faces. Ideally, it's even faces from the person's own life, but of course, those are more difficult to create 'cause you've gotta tailor those for each individual that you test.
Dylan Carnahan:Interesting. I, I'm just curious, um, in talking about kinda testing for that, kinda what is the threshold? Or can you talk about that, that kind of line, so to speak?
Dr. Brad Duchaine:Yeah, sure. So there's no real true dividing line between prosopagnosia and normal face recognition. So this is a arbitrary division that researchers make. I mean, you can think about face recognition ability as a normally distributed ability, and-
Dylan Carnahan:Okay ...
Dr. Brad Duchaine:there have been studies where you've got thousands of people in it, and you do see a pretty normal distribution there. So when we're talking about developmental prosopagnosia, we're talking about people in one of those tails there, the low tail where people are performing poorly on a number of different tests. Where you draw that line, well, that's, you know, arbitrary, as I said earlier. There's no real difference between somebody who's just to the side of that line on, you know, uh, on, in the normal part of the distribution and the person who's in the prosopagnosic distribution. But researchers have found it useful to draw that line when they carry out studies. Um, but, you know, practically speaking, if you're in the second percentile or you're in the third percentile, it doesn't make a difference. We typically... A lot of researchers will draw the line at, at the second percentile, so people where 98% of the population is better than them when it comes to face recognition.
Dylan Carnahan:I'm, I'm curious. Um, what do we still not understand about this condition?
Dr. Brad Duchaine:We don't understand much. Um, I mean, so we have an understanding about which brain areas contribute to face recognition, but we don't know what sort of computations go on in those brain areas. I mean, we have some hints about what's going on in there, and a fair amount of that comes from monkey neurophysiology work where they're able to record from individual neurons within these areas in the brain. Um, but we don't have a good understanding about what it is that these areas are doing in typical participants, and we also, of course, that means we don't have a good understanding about where things go wrong in people with prosopagnosia. We can see if we look at the face network within people with prosopagnosia, we can see in an awful lot of them at the group level, the response is weaker within those parts of the brain in people with prosopagnosia than typical participants. Um, but the differences are fairly subtle. Uh, we're-- I'm working with somebody at Georgia Tech, and we're trying to develop methods where for an individual with prosopagnosia, we can say, "This area is functioning normally. This area is not functioning normally." Um, and we're taking an approach that's a little different than what's been done before, and I'm optimistic about it. But even when we're able to point to a particular part of the brain that's not functioning properly, that still doesn't tell you sort of what's gone wrong in terms of information processing.
Dylan Carnahan:That's so interesting. Um, what have you found surprising in your research?
Dr. Brad Duchaine:Um, well, I think one of the big surprises in work on prosopagnosia is just how many people there are out there. I mean, 'cause when I first started working on prosopagnosia, it was probably 1998, we thought it was extremely rare. And, um, I mean, I thought this w- uh, when I got in touch with a... I was, I was in Santa Barbara and I was testing a high schooler who was in the Los Angeles area, and I thought I've found this really rare neuropsychological participant and I can run lots of studies on him, and like a lot of grad students, I created a webpage talking about my research, and then I started hearing from people. Um, and oftentimes they weren't in California. If they were in California, I'd throw my computers in the car and go test them. But I started realizing, oh, this condition's actually much more common than anybody realized before. Um, and now through our, our prosopagnosia website, we've heard from, I don't know what the number is, 25,000 or something like that. Oh, wow. We hear from people every day. And so these people, there's lots of them out there, as I said, maybe one in 50, one in 100 people have such difficulties with face recognition that it causes them problems in daily life.
Dylan Carnahan:I'm trying to, you know, in going through that journey, you know, from a personal perspective, I mean, how, how does it feel to realize that this is something that is more prevalent, and then understanding kind of the contribution of your research to, you know, this population?
Dr. Brad Duchaine:Yeah, it's satisfying to know that we've I think what we've probably done, we've certainly made some progress on the scientific side in trying to understand what's not working effectively in prosopagnosia, and that's helpful. But that doesn't really do much for people with prosopagnosia, because at this point we don't have interventions that we can use to improve people's face recognition. So what I think we've really done that's helped people, at least in the short term, is just raising public awareness of the condition. And an awful lot of people with prosopagnosia aren't aware of what it is, why it is that they're struggling in social interactions. They know something's wrong, um, but they don't know what They can't pinpoint it. So when they hear about prosopagnosia, it's often a real relief to them. I call it the prosopagnosia epiphany, where you realize, "Oh, I can make sense of all these experiences I've had looking back in my life." Mm-hmm. "It was all due to the fact that I couldn't recognize people's faces. You know, the teacher gave me the, uh, th- my class's homework to hand out to the, the students in the classroom, and I refused to do it. And it's 'cause I didn't know who anybody in my classroom was. I mean, I knew sh- they know so- who some of the people are, but they weren't gonna be able to hand out those pages." And you can understand that everybody's baffled why a student is unwilling to hand out the, the assignments or the homework to the other students, but it, it makes sense when you understand they can't recognize people.
Dylan Carnahan:Yeah. That, that makes a lot of sense. Um, yeah, there's seems like a little bit of, uh, little bit of isolation, right? To, to have this experience in such a way, and I can see too that the people you're interacting with might feel very invalidated, right?
Dr. Brad Duchaine:What, you mean before they know about prosopagnosia?
Dylan Carnahan:Yeah. You know, for example, you're interacting with someone, you don't recognize them, right? And they're, that may have a negative sentiment as a result of that, or maybe you can't pick up on something while you're, you know, interacting with someone.
Dr. Brad Duchaine:No, absolutely. I mean, people take it quite personally if you fail to recognize them when they think you should've recognized them. Um, I mean, I've, you know, I've had conversations with people with prosopagnosia in which they're in tears when they're thinking about, say, lost romantic opportunities. Mm. There was somebody they were very interested in, they look right at them, and they don't react to them in the way that they should've. And it's really hard to recover from a situation like that. People don't understand that, what if you explain, "I didn't know who you were." Um, that, 'cause that indicates to them that you just weren't, they, they weren't all that important to you.
Dylan Carnahan:Mm.
Dr. Brad Duchaine:Similarly, imagine you're, you know, you get into the elevator at your office, and you look right at your boss and you don't acknowledge them. It's not gonna go over well
Dylan Carnahan:Yeah. Um, no, that's very true. I'm just thinking, you know, reflecting on a lot of these social interactions that I may take for granted, right? Mm-hmm. Having that ability. Yeah. Um, what advice do you have for people with prosopagnosia?
Dr. Brad Duchaine:Yeah. I, I... One thing I suggest is, you know, if you know somebody well enough that you feel like you can talk about your difficulties with face recognition to them, do it. Um, because people generally are fairly understanding if you explain, "You know, I have a problem recognizing faces. If I encounter you somewhere else, just say hi to me, 'cause I might not realize who you are." And that, these sort of conversations are getting easier as prosopagnosia is a little better known than it was, say, 25, 30 years ago. Um, there are probably gonna be some people who don't react well to that and are puzzled by that, but I've had prosopagnosics say to me, "You know, if somebody doesn't believe me and can't accept that, well, I don't really even need that person as a friend, frankly."
Dylan Carnahan:Mm-hmm. Um, wow. Th- there's just... Like, this is just extremely interesting, um, as a whole. I greatly appreciate coming on. Um, how can people learn more about you and the work you do?
Dr. Brad Duchaine:Yeah, they can learn about our work. We've got a website which we've had now for 25 years, which is faceblind.org, and on there we talk about, we provide information about prosopagnosia, and that also provides a way for people with prosopagnosia to contact us if they're interested in research participation. Um, and then we've al- I also mentioned the face distortions PMO earlier. If you search for face distortions and, say, you search for my name, Duchesne, or you search for Dartmouth, hopefully our website comes up pretty high. Unfortunately, there's a lot of, say, face distortion apps out there that we're competing with. Um, but, um, yeah, hopefully they can find me if they just provide some specific information.
Dylan Carnahan:Yeah. Um, the competition part, uh, for random apps, I didn't even think about that. Um, you can expect me to list those resources in the show notes, uh, for this particular podcast episode. Um, Dr. Duchesne, thank you for sharing your knowledge and time today.
Dr. Brad Duchaine:Great. Thank you, Dylan.
Dylan Carnahan:That wraps up our conversation with Dr. Duchesne. We talked about how the brain recognizes faces, the difference between prosopagnosia and PMO, and the emotional toll of face blindness. Go to this episode's show notes to see any resources Dr. Duchesne mentioned during our episode. And lastly, subscribe to the Simple Questions podcast to get notified when our latest episodes are released. Thank you for listening, and remember to keep asking questions