How accurate are our first impressions? With Nicholas Rule, PhD

It's a cliché that you only get one chance to make a first impression, but it turns out that science bears this out. First impressions matter, and they're often surprisingly accurate. From the moment you meet someone at a party, a job interview, or even just crossing paths at the grocery store, your brain is quickly summing them up. Whether you mean to do it or not, you're assigning people to social categories and making judgments about their personality and other qualities. But exactly how accurate are first impressions? What kinds of information can we glean about someone in the first moments of meeting them or even from just a photograph? Can you tell whether someone is extroverted, trustworthy, a good leader, gay or lesbian? What about their political leanings or whether they grew up wealthy? To what degree are our first impressions based on something real, and to what extent are they based on stereotypes or biases? And what are the real-world consequences of our accurate and inaccurate snap judgments? Welcome to Speaking of Psychology, the flagship podcast of the American Psychological Association that examines the links between psychological science and everyday life. I'm Kim Mills. My guest today is Dr. Nicholas Ruhl, a professor of psychology and vice provost for academic programs at the University of Toronto. His research focuses on the science of snap judgments and first impressions. He studies how we perceive other people when we first meet them, when we're accurate and when we're not, and the consequences of those quick judgments. Dr. Ruhl has published hundreds of academic papers, and his research has been covered widely in the mainstream media, including by CNN, The New York Times, NPR and The Washington Post. He is also editor-in-chief of the Journal of Experimental Social Psychology and was recently elected the 2027 president of the Society for Personality and Social Psychology. Dr. Ruhl, thank you for joining me. It's my great pleasure to be here, Kim. So when we first meet someone, how quickly do we start to make judgments and form an impression of them? Does it happen within seconds, milliseconds, and what kinds of judgments are we making in that brief time? It happens instantly, actually, and long before you realize it. So that would be on the order of milliseconds, as you say. The types of judgments that we make about people, they can vary a bit, but generally they follow something that you could almost consider along the lines of a hierarchy of needs. And so the first thing that we'll size up about a person is whether they're going to be a danger to us. Is this someone that's going to hurt me? Is this someone who's going to help me? Are they safe or are they not? After that, the next type of judgment we tend to make is how useful they might be to me, and particularly around whether this is an opportunity for me to mate with a person. And so we make an assessment about their gender, their interests, their age, and other factors that might lead us to make a judgment about whether they are compatible with us in terms of fertility. After that, we proceed on to higher order questions. And so one might be a little bit more complicated, but a version of what I said in the first one, which is, is this person my ally? Can I affiliate with them or are they someone who belongs to a different group from me? And that's slightly different than whether they're dangerous or not, though sometimes those things can be overlapping. Beyond that, we start to get into these really higher order kinds of things. And so some of the judgments that you mentioned is this person extroverted. Is this person a good leader? And those are more social or human dimensions that go beyond sort of the judgments that we would make more basically about animals, other objects, and things like that, from which our judgments of people are largely based. And all of this happens, as I said, extremely quickly within milliseconds before we even have an opportunity to really think about what's happening. And that's a critical piece of what transpires in our interactions with people later on, in other words, how we begin to treat them. Much of your research focuses on how well people can pick up on what you call perceptually ambiguous groups, things that you think wouldn't be obvious from someone's appearance like political affiliation or religion, sexual orientation, social class, even I think you mentioned leadership success. Can you tell us about that work? How accurately are people able to pick up on these traits from a quick meeting or even just from a photo? Well, the short answer is about 65%. And so that's kind of the average overall. But that varies quite a bit depending on what type of judgment you're making, how you're making the judgment. So accuracy for photos is a lot lower. That's on the lower end, right? We have less information there versus interacting with someone and live face to face, you get so much more information than you would get in a two-dimensional photograph. Or even in the case of some of the studies we've run in my lab where we're showing people really just parts of faces, so individual eyes. And yet we find that from just a pair of eyes, black and white photograph, no brows visible that people can make judgments about things like their sexual orientation better than they would be able to do if they were just guessing. And so it's maybe a low bar to say that we're better than a 50-50 chance of guessing whether someone's gay or straight, but systematically over a period of judgments that adds up to something that's pretty meaningful and gives us insight about the mind and brain. And so as I mentioned, it can depend a little bit on the type of judgment that you're making as well. So, for example, some judgments like judging someone's age is shockingly accurate. And so in studies that we've run in my lab, we find that people can judge another person's age within about three years. It's incredible accuracy. And if you sort of quantify that in a different way, people are accurate roughly 80% of the time in judging a person's relative age. Judgments of gender are much higher. They tend to be almost at ceiling, almost 100% accuracy. There are, of course, always exceptions to these things. And I think we can, when you think about the popular media, they often make a light of things like this. There was the famous Pat skit on Saturday Night Live decades ago, which plays on some of these errors that people might make or at least the ambiguity around it. So accuracy, as I say, can vary quite a bit. It really does depend on what kind of judgment you're making. And part of that, too, is sort of the opportunity that you have to make that judgment. As I said, whether it's from a visual perception, an auditory perception, oftentimes auditory judgments are better than visual judgments. Even though we, as a visual species, one quarter of our brain is devoted to just processing visual information, we tend to prioritize visual information about people, but oftentimes other senses actually pick up things a little bit more. And that's one of the more fascinating parts of this area of research. So what kinds of information or cues are we picking up on when we're making these judgments and are we doing it consciously or is it all happening below our actual awareness? Well, it's a combination, right? And so, I mean, there are some judgments where obviously if I ask you to judge whether somebody is, say, extroverted or introverted, you're going to be consciously thinking about it. However, it turns out that you probably already made a decision about that before I asked you. And so depending on the circumstances, it's usually a combination of the two. That said, we do have evidence from a variety of studies showing that a lot of this occurs outside of awareness. So as I said a few minutes ago, before you even start to think about what you're doing, your brain has largely done it. As I mentioned, some of the judgments that we make, they follow this sort of sequence and at first we're sort of getting kind of the broad contours of a person. So their size, their proximity, from there we go on to these sort of finer distinctions, such as things like, well, I just mentioned a moment ago, eyes, but I'll go to eyebrows. So the shape of someone's eyebrows or something that we make a judgment about and that influences the way that we interact with them. So for example, in a series of studies that we did a few years ago, we found that people could reliably judge whether someone was narcissistic from just looking at their eyebrows. And to your question about sort of what are we picking up on, in that case, it turns out that it was the grooming of the eyebrows. So people with more groomed eyebrows tended to have higher levels of grandiose narcissism. And so people somehow know that. People pick up on that and they make an inference about it based on just that very, very tiny bit of information. They then sort of run with that and kind of explode it in a way and make a lot more out of it than perhaps they sometimes should. How do snap judgments interact with stereotypes and bias? I understand that you found, for instance, that people perceive black men, and I assume this is white people, perceive black men as larger and more threatening than same size white men. So how much of our, quote, intuition about people is actually bias? I'm so glad that you asked that because that's such an important part of this whole process of perceiving people. And the answer is quite a lot is bias, actually. A lot of that bias comes from patterns based on our own idiosyncratic experiences, including things that we've observed and a lot of it from things that other people have told us. And we're kind of sponges for information about other people in ways that lead to making assessments of whether someone is going to be safe for us or not. And so generally in psychology, lots of researchers have observed that there's a negativity bias. So we tend to hold on to negative information much more than we do positive information. And so in the case of the study that you just mentioned, we found that it was actually both white and black people who overestimated the size of black men. And one of the reasons that it's important that it applied in both cases is that it underscores the fact that this is really something that we are learning from each other in large part, right? That it corresponds to narratives that we tell in society that black men are stronger, bigger, more threatening, all of these things that we see bear out in these misperceptions of individuals. And so you might be familiar with a very old study from the field in which children who came from less socioeconomically privileged backgrounds, they would perceive the size of a coin as being larger than it actually was because it meant more to them. And we see that same kind of misperception happening with people as well. And that's part of what we found in this work here in particular. So in that study, although both black and white people misperceived the size of black men, they saw them as bigger both in height and they saw them as stronger. They saw them as, as I mentioned, taller, heavier, all of these things. It was only the white people who misperceived them as being more threatening, who converted that into believing that they were going to be more of a threat to them, as they could harm them more. And thus, they believed that greater use of force should be applied to keeping them from harming them. And so that's a critical piece there that, again, intersects with that point that I made about our subjective experiences, our idiosyncratic experiences. And so it's, again, a combination of these different streams of information coming together and informing the way that we approach each other. And so when we talk about accuracy and person perception, we think of accuracy and bias just two sides of the same coin in a way. There are parts of the things that we perceive about each other that are, they're completely based on stereotypes. And then there are elements, sometimes, of those stereotypes that have kernels of truth from which they grew and overgrew. And that's one thing that we call an overgeneralization effect. And so it's those overgeneralization effects that can range from things like misperceiving the size of black men, which has some pretty serious consequences, to misperceiving that someone whose face looks like a cat is going to have the attributes of a lion. And so it really does range from it. It sort of speaks to this more general process that our minds use for understanding sort of who the people are around us. And so it's really, as a scholar of this area, it can be both disheartening and also amusing at the same time, depending on what we're looking at. Well, continuing on the topic of stereotypes and bias, I want to ask you about the question of gaydar, the ability of some people just to judge right away who's gay and who's straight. Is that a real thing? And are some people better at it than others? What do we know about this? We know a tremendous amount about it, actually, which is maybe almost embarrassing to admit. But we've, particularly in my lab and in some others, we've done a lot of research into this. And a part of it is, again, getting back to that notion that it's so unbelievable that you really sort of, once you start to find something, you really have to sort of go down the hole figuring out, well, what's going on. And so people can tell whether someone is gay or straight and more often than not, I should say. Again, there are always exceptions. And some people who seem very much like they would be gay end up being straight. And some people who seem very much like they ought to be straight, you find that they happen to be gay. But on the whole, accuracy is not bad. It's around 60%. And that's based on really limited information, as I mentioned before. So things like just a pair of eyes in a black and white photograph, or even showing someone a picture of a person's face for one 20th of a second. So there's a lot of information there that leads people to make these judgments. And we've sliced and diced it pretty much any way that you can. And generally what we find is that it tends to be constructed on top of gender. And so women who seem, for whatever reason, whether it's the way they look, the way they sound, a combination of those, the way they walk, if they seem more masculine, then people are more likely to think that they are going to be a lesbian and chances are more often than not they are. Likewise, for men who seem more feminine, they might have a more feminine voice. They might have a more feminine way of moving their bodies. They might have more feminine features in their face. People are more likely to think that they're gay and chances are more often than not that they are. So as I said, we know a lot about gay. You asked if some people are better than others and we do find that there is variability in that. So generally, people who themselves are gay or straight do tend to do better. And we think that this is mostly in part just because of their amount of exposure to other gay and straight people. It's not necessarily some kind of special ability that's born into people where they can tell, but rather that we pay much more attention to the people who are like us than the people who are not. When we're doing so, we learn a lot more of those subtle cues. And also, as I mentioned at the outset, there's a motivation around mating that drives a lot of our perception. So gay men are very interested in understanding what men might be available to mate with them for a couple of reasons. One is that they want to mate with those people and the other is that if they make a mistake, the consequences can be very devastating for them on a personal level. And this applies to lesbian women as well. We've done some studies looking at bisexual people. That's often a question that's asked because, of course, there's more than just the gay straight dichotomy. And what we find is that people can also detect whether someone is bisexual from photos of their faces and other cues. And what we generally find is that it's people who tend to see more on the side of being gay who are sort of attributed as being bisexual and such that the dichotomy isn't so much gay versus straight. It's more like non-straight versus straight. And so anyone who's outside of sort of strict heteronormative parameters generally tends to be ascribed a non-heterosexual identity, whether that's gay or bisexual, people don't make a lot of differentiation between the two. We're going to take a short break. When we return, we'll talk about how we judge people's trustworthiness or untrustworthiness and the real-world consequences those judgments can have. What are some of the other real-world consequences of these kinds of judgments? For instance, you've looked at how people's impressions of a person's trustworthiness, just by looking at their face, might affect criminal sentencing decisions. Can you talk about that research as well? Yeah, that's some of the most important research that I think we've done, in part because the consequences are so severe. In that work, it wasn't even only that they got harsher sentences that they actually got the death penalty. And what we found was that people who looked untrustworthy were more likely to have been assigned the death penalty by a judge and a jury than people who looked more trustworthy, even in cases in which we were able to go back and find that the individual had been exonerated of their crimes. And so it's a general bias if we ask participants in the lab to simulate if they were a member of a jury, which they very well could be, because of course juries are taken from the general population, ask them, if this person had committed a crime, what kind of sentence do you think they should get? They're much more likely to harshly punish, to the extent of the death penalty, someone who has a face that just looks untrustworthy versus someone who looks a bit more trustworthy. Of course, the other side of that is that we tend to make mistakes around people who look especially trustworthy. And so Bernie Madoff is a good example. He has a fairly trustworthy looking face. And so it's not as big surprise that he was able to fool a lot of people into investing with him. They trusted him. And these types of individuals, these sort of confidence men, as they're often called, non-men, often are able to leverage their physical appearance and charisma to lure people in in ways such as that. And it's interesting thinking about trustworthy in contrast to judgments of sexual orientation or GEDAR is an important distinction because people tend to think that they're really good at judging trustworthiness. They think, I can tell whether someone is trustworthy. I know whether to trust them. And they have a lot of confidence about that judgment. And people feel that they cannot tell whether someone is gay or straight from looking at them. And yet what we find is that accuracy runs in the inverse of those. People are much better than they expect in judging sexual orientation and much worse than they expect in judging trustworthiness. Judgments of trustworthiness, deception, they tend to be basically transguessing for the most part. And it's that belief that is so uncorrelated or disconnected from the reality of what they're actually able to do that fascinates people like me and the people I work with. How hard is it to change a first impression? So as you get to know someone better, does the first impression stick or does it change as you get more information? It's extremely hard to change a first impression, I'm sorry to say. And there are a variety of researchers that have looked at this question in a number of ways and it is possible to change a first impression. But what we've observed in my lab is that especially with those very quick judgments, they seem to occur persistently. And so for example, if you were to meet somebody and you have a particular impression of them, I can take you, I can train you, and I can make you think the opposite over and over and over until you get it absolutely perfect. And then a few minutes later, you revert right back to that first impression again. And so what we've hypothesized is that this likely occurs every time you see someone, even people that you know. In those first few milliseconds, you still judge them as if there's someone that you've never seen before. Almost something that you've never seen before and you make those same assessments is this person dangerous. Can I meet with this person so on and so forth? But what happens within a few hundred milliseconds is that you recognize the person and you overwrite that. You're then able to say, oh no, that's Kim. I know Kim. I feel good about Kim. I have these experiences with Kim. And so I am not going to be scared of her anymore. And then that goes away. But it underscores again that stubbornness around these first impressions that can be quite discouraging. I could go on with lots of anecdotes about this. But I think we've all had that experience at one time or another where there's someone we know and we think they don't look consistent with who they are at all. And I know them. But chances are that when you first see them, every time you see them, you're thinking the same thing that everybody else is. Kim kind of explains why you might stick with somebody as a partner who really is kind of a jerk, but you're so attracted that you can't stop. You can't change the way your brain is working. Yeah. Attractiveness is one of the most powerful forces in human nature. It's true. Let's switch for a minute and talk about the role of artificial intelligence. As facial recognition and AI become more powerful and really embedded in our lives, what's going to happen as machines start making the same kinds of judgments that humans make? Is there a danger in that? I think there certainly is a danger, but I'll tell you, I'm much more worried about human bias than I am about AI bias. And partly for the reason that we just discussed. What we've observed over many decades now of human perception is that we have a lot of confidence in stating things based on our subjective experiences that can lead us to a lot of trouble. Whether we carry this over to artificial intelligence, it looks like we probably are doing that, and we'll continue to do that. As long as AI aspires to be like a human, it's going to reflect all the good and bad of what it means to be a human. I think there's a particular risk in people perhaps having a sense of objectivity about AI because it seems as if it ought not have bias because it's a machine. Of course, I think that that impression I know I can speak for myself is quickly fading. My own experiences with artificial intelligence have led me to not trust it very much at all. So whether I'll eventually come around to trusting AI the way that I might have previously trusted a machine, that would take a lot of effort by our friends in Silicon Valley. But as I said, the more that AI emulates humans, I think that our trust in AI will continue to follow that same trajectory whereby we'll be able to discern what's trustworthy and what's not. As I said, I worry a little bit about state actors and entities that have a lot of power relying on artificial intelligence as a tool where they haven't necessarily considered the effects of it or its own biases. But that's really no different than, again, based on my experience with artificial intelligence thus far, if they were to take a group of school children and ask them to make these important decisions. Because I would say my experience with large language models has been essentially that they're very much like children telling me what they think I want to hear for the most part. None of us would responsibly pick individuals out of an elementary school to make these important decisions about profiling individuals yet perhaps that may be what we're doing at this moment. So I think we'll need to come a long way. But as I said, between now and then, I'm really worried about the human biases more than the computer biases. I have a work colleague who likens AI to having a not very bright intern working for you. So how should people think about this research when they're considering the impression that they're making on other people? Are there things that we should do to make a good first impression, say, on a job interview or on a date? My advice when I'm asked questions of this type is that you shouldn't think about it. Don't think about it. And I'll tell you why. Automatic human cognitions are not that different from doing any other kind of complicated sequence of steps. And so if you were to, I mean, these days, there aren't a lot of automobiles with stick shift ignitions. But if you can cast your mind back to the days when that was far more common, we've all had this experience of trying to drive a car when we're not familiar with that. And it's like you've got one foot on the clutch and you've got your hand moving and you end up stalling out in traffic. The same thing is going to happen if you overthink it. So the best thing that you could do, the best advice I have is don't think about the impression you're making. Go and be yourself and hope that you've been lucky enough to have the right face, voice, and manner that will line up with the job that you're trying to get or the person you're trying to impress. Because ultimately, if you overthink it, you're just going to, you're just going to fall a little bit on your face. How did you get interested in studying first impressions and snap judgments? When I was growing up, I was befuddled by things that I would see people do that contrasted with the things that they would say. And I grew up in an environment that was quite violent. There was a lot of crime. I grew up across the street from what we would have called a crack house at the time. They were not only dealing drugs, they were running prostitution and all of these things. The altercations that would occur at a place like this would often spill over onto our property as well. And so I needed to develop a high level of vigilance. And in those circumstances, I started to glean pretty quickly that the judgments that people would make based on very small amounts of information were guiding a lot of the things that they would do somewhat instinctively, habitually, and seemingly without a lot of depth of thought. And so based on that experience, I was extremely interested in understanding that dichotomy between, as I said, what people say and what they do. And so when I had the great privilege and opportunity of going away for a post-secondary education to college, I really wanted to study this as long as I could. I was the first person in my family to go away to get an education. I didn't think I was going to be there very long. I expected that I would fail. And so I was really doggedly pursuing this right from the beginning because I figured I didn't have a great window of opportunity to get an answer to these questions. And so I started asking the questions, finding answers, going to the library, irritating my professors with lots of questions and professors at other universities as well. At one time, my first few weeks as a freshman in college, I wrote an email to Noam Chomsky, asking him a question. He wrote back with a kind reply that, in retrospect, I can see he was really humoring me. But I really wanted to pursue it, as I said, as long as I could. And even once I did manage to graduate and get into graduate school, I had that same mentality where I was just going to study this for as long as I could to understand better. And so that was really the genesis of my interest in this work. And I can say how fortunate I am that I've been able to make a career out of answering a question that has meant so much to me, not just intellectually, but personally as well. So looking ahead, what are the big questions that you still want to answer? What are some of the things you're working on right now? I'm really interested in understanding what it is that we don't know about ourselves, right? So again, these contradictions or hypocrisies, you could even call them in people's behaviors, there's all this information that we have in studies in my lab and others have uncovered. In our perception of other people, we're taking in this huge volume of information. Our minds are unconsciously processing the majority of it, and a very small sliver seems to get promoted up to our conscious thought. And that conscious thought is sometimes misdirected as we talked about with bias. It can lead us down paths that we wouldn't ideally like to take. And so the big question that I am still pursuing is kind of that one that I started with, which is what? What is it that we know versus we don't know? How do we reconcile those two things? Why do we say things that are different from what we do? And making sense of all of that, particularly as it applies to our perceptions of each other. Right now, one of the directions that my lab is taking with this and an area in which I'm extremely excited is thinking about the ways in which our perceptions of people extend beyond their bodies and their faces and the things that we are all actively doing in any moment. And rather the way they get sort of translated or transmuted, perhaps more appropriately, into the things with which we interact. And so there's been some work that was done in the starting in the early 2000s on a concept called behavioral residue. And that's the idea that we leave traces of ourselves behind in places that we've been. And so, for example, there was work showing that if you go into someone's office or their bedroom, you can judge their personality traits based on a variety of things. What we're finding recently in my lab and some unpublished work that we're sending out soon for publication is that people also embed qualities of their social categories. So the groups to which they belong, for example, whether they're a man or a woman, in things that they make. And so, for example, this might apply to the things that people curate, the things that they choose, but products that they develop, even including the food that they cook. And so that's really sort of, I think, the frontier for us right now, what we're calling extra personal features. And I'm super excited about that work going forward. Well, this is all great information. I want to thank you for joining me today, Dr. Roll. This was a fun conversation. It was wonderful to be here. Thanks so much, Kim. You can find previous episodes of Speaking of Psychology on our website at speakingofpsychology.org or on Apple, Spotify, YouTube, or wherever you get podcasts. And if you like what you've heard, please follow us and leave a rating. If you have comments or ideas for future episodes, you can email us at speakingofpsychology.org. Speaking of Psychology is produced by Lee Weinerman. Thank you for listening for the American Psychological Association. I'm Kim Mills. Security Program on Spreadsheets. New regulations piling up. An audit dread. It's time for Vantor. Vantor automates security and compliance, brings evidence into one place and cuts audit prep by 82%. Less manual work, clearer visibility, faster deals, zero chaos. Call it compliance or call it calm compliance. Get it? Join the 15,000 companies using Vantor to prove trust. Get started at vantor.com slash calm.