Sounds Like a Winner

We’re used to the idea that rhetoric sways voters. But what about another element of language: a candidate’s voice? This week on Hidden Brain, what happens when our political system and ancient biological rules meet. For more information about the research in this episode, visit

The transcript below may be for an earlier version of this episode. Our transcripts are provided by various partners and may contain errors or deviate slightly from the audio.


This is HIDDEN BRAIN. I'm Shankar Vedantam. Which dog do you think is bigger...


VEDANTAM: ...Fido...


VEDANTAM: ...Or Rover?


VEDANTAM: Both of these dogs are making the exact same sound.


VEDANTAM: They're barking, but the vocal quality of the barks easily gives away the size of the dog. The lower pitched dog is a large Saint Bernard while high-pitched Fido is a small Chihuahua.


VEDANTAM: Throughout nature...


VEDANTAM: ...The vocalizations of different animals offer clues.


VEDANTAM: From birds...


VEDANTAM: ...To baboons...


VEDANTAM: ...The sounds you hear can tell you useful things about an animal's size, its intentions, even its role in its social hierarchy.


VEDANTAM: But is any of this true for humans? Does the human voice convey anything important about who's up and who's down? At one level, the answer is no. We determine rank and pecking order using sophisticated language and cultural norms. In politics, we select leaders through elections.


VEDANTAM: But at the same time, humans are also animals. Today on HIDDEN BRAIN - the tension between our modern, complex political systems and ancient biological rules that tell us who's in charge.


CASEY KLOFSTAD: Nonverbal vocal signals that do not necessarily connote language or grammar the way that you and I would understand it as human beings are conveying a great deal of information.


ARNOLD SCHWARZENEGGER: I want to be the people's governor.

UNIDENTIFIED PERSON #1: ...Even my soft voice might be a bit more effective at getting conservative things done.

KLOFSTAD: And what we've done is we have taken that model, and we've applied it to politics and to political candidates.


VEDANTAM: Some years ago, the political scientist Casey Klofstad was hanging out with his wife, Rindy Anderson. She's a biologist who studies songbird vocalizations. They were watching TV. They noticed something that connected her interest in bird sounds with his interest in politics.

KLOFSTAD: Yeah, I can picture us, you know, lying on the couch, you know, mid-evening, you know, and we're just sort of flipping through channels, and we landed on Fox or CNN. I forget which. And you know, my wife brought up the fact that, you know, it seems like the women that are broadcasters tend to have a lower timbre voice. Is there something to this? And at the time, it seemed very anecdotal. Obviously, it's not scientific, but, you know, it led us down this path.


VEDANTAM: The path it led them down was to explore whether there was any connection between the world of animal vocalizations and the world of human vocalizations. Many studies have explored the idea that grunts and roars and squeaks tell us useful things about the size of animals and their social dominance. Casey and Rindy began to look for empirical evidence that connected the human voice with electoral and political outcomes. In one recent study, Casey told me he analyzed the voices of candidates running in the 2012 election for the U.S. House of Representatives.


RODNEY DAVIS: I'm Rodney Davis...

SUSAN DELBENE: I'm Susan DelBene, and I approve this message.

AMI BERA: I'm Dr. Ami Bera, and I approve this message.

JIM RENACCI: I'm Jim Renacci.

DAN MAFFEI: I'm Dan Maffei, and I approve this message.

JUDY BIGGERT: I'm Judy Biggert, and I approve this message.

VEDANTAM: I asked him what he found.

KLOFSTAD: What we found is that both men and women with lower pitched voices were more likely to win. And we found that that is even the case in terms of whether they won or lost or the vote share - the percentage of the votes that they won. You know, cut it any way you want, they were more likely to win. And then we can even control for things like campaign spending, the gender of the candidate, the ideological persuasion of the district that they are running in - things of that nature. We threw the kitchen sink at it, and we tried to explain away the voice-pitch effect, and we could not.


VEDANTAM: At one level, Casey says, this is surprising; at another, it is entirely unsurprising.

KLOFSTAD: Nonverbal vocal signals are endemic throughout the animal kingdom, whether it's human beings or songbirds or anybody in between. Whether it's the hiss of the snake, the roar of the red deer, the bark of a dog, you know, snarl of a dog - things that you see every day that do not necessarily connote language or grammar the way that you and I would understand it as human beings are conveying a great deal of information. And what we've done is we have taken that model, and we have applied it to politics and to political candidates.

VEDANTAM: So when you think about politics - which is, of course, a world in which you live and breathe - the conventional way of thinking about it is that we choose people to be our leaders because they have qualities of leadership. They have a track record of experience. They are skilled communicators. They have great vision about the future. And I think many people would say voice has nothing to do with it. You know, the candidate's voice - high-pitched, low-pitched - how can that possibly matter?

KLOFSTAD: So we, as human beings, make very thin, impressionistic judgments. And I will equivocate to you and say that, yes, that many myriad other things line up to influence who we vote for. And the biggest thing for me, as a political scientist, is partisanship, right? If you're a Democrat, you're going to vote - priori vote for Democrats. If you're a Republican, you're going to vote for a Republican.

But let's say, for example, it's a municipal race that's nonpartisan. You have no prior information. You don't have anything to judge on off the cuff. Now we start to grasp at other straws. So it's not to say that we have the Rosetta Stone, if you will, for electoral politics, but it's one thing that we want to add as a piece to the puzzle. And on top of that, not just voice pitch or whatever, but to acknowledge the fact that we are animals and that we act on animal instincts.


VEDANTAM: When you say we're animals and we act on animal instincts, are you saying that, in some ways, things that might be written deep into our, you know, the algorithms of how our brain works, things that, you know, I call the hidden brain - the unconscious mind, things that we're not aware of - can shape our behavior in the conscious world when it comes to politics?

KLOFSTAD: Absolutely, right? These things are affecting us whether we are conscious of it or not. And you know, Alex Todorov - which I'm sure you're familiar with his work - showed that if you give an image of a candidate for 0.2 seconds, that an individual is willing and able to make a judgment about that individual's competence, and at a rate greater than chance is that rating is able to predict whether that individual wins. So whether it's vision, whether it's voice - whatever - we are judging on a very thin basis.

VEDANTAM: What explains the dichotomy between the way we think about the way we make these decisions and the way we actually might make these decisions? Or I think the thing that a lot of people find surprising is that when the Todorov study finds that these very brief exposure to images or your finding, that very brief, you know, hearing different kinds of voices or pitches, that this has an effect, you know, maybe a small but a measurable effect on electoral outcomes, people are surprised by it because they think, I am just evaluating the candidates and the issues, and I'm coming to a relatively reasoned conclusion about which candidate to support. And your theory is, yes, you might be doing that, but there's also something else that's happening in your brain.

KLOFSTAD: Well, I think you said, I think. And that's the operative phrase, that, yes, indeed, we do cognate most of our decisions generally and most of our political decisions generally. But you know, in this day and age, political decisions are made on a very snap-judgment-basis level. And when elections, particularly nowadays, are won and lost in very thin margins - yes, things like partisanship or other perceptions are important - but when it comes down to these little perceptions of faces, voices, things of that nature - especially if you extract partisanship out of it - if it's a municipal election or if it's a primary, where it's Democrats versus Democrats or Republicans versus Republicans - all of these other things that we bring to the table, as animals, matter.

VEDANTAM: And of course, it actually makes sense. When we're thinking about sort of leadership and politics, we are thinking in some ways about how coalitions work, who's up, who's down, how our side can win, what's going to be best for us, what could harm us? I mean, those questions we might not quite articulate to ourselves in that language, but that is what politics is about.

KLOFSTAD: Absolutely. I mean, you know, nowadays - and we've said this in the coda of many of our papers that, you know, nowadays, obviously politics is about complex ideologies, but at the end of the day, it's about, you know, it's about sticks and stones. And again, not to get too much into the evolutionary psychology waters, but we hold on to that. It is a part of our history, and it plays itself out in modern day decision-making.


VEDANTAM: Coming up - one dilemma facing the growing number of women running for high office.


VEDANTAM: Political scientist Casey Klofstad and biologist Rindy Anderson have jointly explored how the pitch of people's voices might communicate important information in the context of politics. Of course, it's very hard when you're looking at an individual race to disentangle all the effects that might be prompting people to vote for one candidate or another. As Casey points out, forces like partisanship influence voters powerfully. Since there are lots of factors that go into winning an election, how do you tell whether the candidate's voice is playing any role? Casey says the way to do this is through lab experiments that hold the other factors at bay.

KLOFSTAD: The lab allows us to take an anecdote, something from the mass media - and as I like to say, and this is not a very scientific term, but I'm going to use it anyway - let's put some science on it, right? Let's take individuals into the lab and give them two experimentally manipulated voices spoken by the same individual, one that happens to be a little bit higher and a little bit lower, manipulated by my wife and I. And then we say - we don't tell them who's high or low. We just say this is A and B, and vote. And we do that dozens of times with a thousand experimental subjects, and we get the aggregate average. Now, why take it into the lab? Causation. What's going on here? Taking an anecdote and putting science on it. And that is what our objective has been.


VEDANTAM: In the experiment, the researchers played a number of clips. One of them included these two women's voices.



UNIDENTIFIED PERSON #2: I urge you to vote for me this November.


UNIDENTIFIED PERSON #2: I urge you to vote for me this November.

VEDANTAM: I ask Casey what the experiment found.

KLOFSTAD: Voice A was the higher version. Voice B was the lower version. And these were experimentally manipulated, and they were randomized so it wouldn't always be voice A is high, voice B is low. And the idea is, I urge you to vote for me this November - something that's electorally relevant. Something that a typical candidate would say. But it's not liberal. It's not conservative.

VEDANTAM: And when you look at a number of these kinds of stimuli when it comes to women's voices in particular - looking just at women's voices - what does your experiment find in terms of the effect that higher pitches and lower pitches have on, you know, how much people relate and like the candidate?

KLOFSTAD: The effect is we like candidates with lower pitched voices, regardless of the gender of the voter, regardless of the gender of the candidate. But the bias is a little bit stronger when it comes to women. So to put a pin on it, we like low regardless of the gender of the voter, regardless of the gender of the candidate, but the bias is a little bit stronger if we get two female candidates.


VEDANTAM: Casey points to prominent examples in actual politics.

KLOFSTAD: One is with Margaret Thatcher, and there was an entire movie about it, whereby she received vocal training and lowered the timbre of her voice. You can go on YouTube, and you can find before and afters. There's multiple ones of them. And again, these are anecdotes, but this is the idea whereby a woman, who had very raw, strong leadership capacity, was coached up by her agents and by actors and said, lower the timbre of your voice.


VEDANTAM: Margaret Thatcher was the first female prime minister of Britain. You can hear her voice evolve over three decades. Some of this was probably caused by aging, but she also deliberately changed the way she spoke. Here she is in the early 1960s.


MARGARET THATCHER: I think it's going to be an even more organizational method. I'm a great believer in those two things.

VEDANTAM: Here she is about a decade later.


THATCHER: Liberty is fundamental. Liberty, human dignity, a higher standard of living is fundamental.

VEDANTAM: Margaret Thatcher began taking voice lessons from the Royal National Theatre. In May 1979, she became prime minister.


UNIDENTIFIED PERSON #3: Congratulations.

THATCHER: Thank you very much.

UNIDENTIFIED PERSON #3: How do you feel at this moment?

THATCHER: Very excited, very aware of the responsibilities.

VEDANTAM: Her enunciation and delivery became hallmarks of her leadership style.


THATCHER: So what the honorable member is saying is that he would rather the poor were poorer, provided the rich were less rich, that way you will never create the...

VEDANTAM: What do you think is actually happening, as we listen to Margaret Thatcher's lower pitched voice versus her higher pitched voice? What do you think was changing in how listeners were perceiving the candidate? In other words, I think what I'm asking is - we know that more people might be willing to vote for the woman with the lower pitched voice, but why? Is it because they trust her more, they like her more, they respect her more? What's the reason?

KLOFSTAD: So initially, our experiments just asked, you know, vote up or down, A or B, high or low. And you played an example earlier of that. What we did in subsequent experiments is we asked who was stronger, A or B? And who is more competent, A or B? And it turns out that those perceptions are highly predictive of whether you voted for low. So to be succinct about it, why do we like candidates with lower voices, regardless if they're men or women? It's because we perceive them as stronger and more competent.

VEDANTAM: If you are thinking of running for political office or you want to get into that corner office, there is an important caveat to keep in mind before you sign up for voice lessons. It's true that voters appear to prefer candidates with lower pitched voices, but not, it turns out, when those voices are too low.

KLOFSTAD: Yes, we prefer candidates with lower pitched voices because it connotes strength, and it connotes leadership abilities. OK. But if you get too low, for example, vocal fry - (imitating vocal fry) vocal fry, right?

That is too low. Not a very scientific term, but it is too low. It is out of the average. So the general proposition that we've been talking about, with regard to candidates, is that low is good. But if you get too low, and particularly if you're a woman and if you go really low, we don't like it. We find it annoying. And the papers that we have published have shown that folks - especially women, but both men and women - who speak in a vocal fry, they exaggerate the lower timbre of their voice, are seen as less trustworthy and actually less hirable.


VEDANTAM: Vocal fry is only one of the landmines that women face when it comes to how their voices are perceived by others. There is separate research that indicates that when women have high-pitched voices, they are perceived to be more attractive. And we've just heard about research that shows that when women have lower pitched voices, they are perceived to be more competent. Given that getting elected often requires being seen as both attractive and competent, what are female candidate supposed to do?

KLOFSTAD: So women are in a bind. As you mentioned, both men and women - lower pitched voices are perceived as more dominant, more competent, and then when we shift to women, that higher pitched voices are perceived as more attractive, and we know that women who are perceived as more attractive are seen as more electable.

And a number of different studies have shown that, yes, there is a double standard. We want attractive women in office, and yet our perceptions of their voices sort of countermands that. So we are living in a modern society. We have culture. We have new standards of how we want to govern ourselves. And yet there are these things that are buried in our hidden brains.

VEDANTAM: There is a certain amount of unfairness in this whole - in this whole bias that we have about how voices operate. Because I understand that, at a certain animalistic level, this might have made sense millions of years ago. It really doesn't make sense in 2012 or 2016 or 2020.

KLOFSTAD: Well - and I'm reluctant to put, you know, evolutionary explanations on this. But I mean, were we to delve into those weeds - I mean, we're animals. And you know, it was - it was evolutionarily adaptive for us to find leaders, be they men or women - mostly men, I would presume, back in antiquity - with lower voices. Lower voices corresponds with higher levels of testosterone, which corresponds with higher levels of physical and social aggressiveness.

So if we are throwing rocks, and we're hitting each other with sticks, I mean, that's a pretty good selection mechanism. And maybe we've held onto that a little bit.

VEDANTAM: But of course, that doesn't always make sense, right? You don't necessarily want the best rock-thrower being your representative in Congress.

KLOFSTAD: Yeah. And that is the rub, right? Are these folks actually better leaders?

VEDANTAM: Are folks with deep voices actually better leaders? Casey and Rindy asked that question in a recent paper. They analyzed the voices of members of Congress in 2008 and compared the timbre of those voices against the politician's leadership ability.

KLOFSTAD: There was a think tank that came up with a ranking of all of the members of Congress, both in the House and the Senate, in terms of various different characteristics - you know, how good or powerful the committees they were on, how long they had been in office, what their vote share - basically, distilling a lot of different characteristics, objective characteristics, of their leadership ability. And we're going to give you a score.

So I had my students, about a year and a half ago, gather those data, and then go get recordings off of YouTube of those individuals' voice pitch, and we did the correlation. We said, is there any correlation between this "objective," quote-unquote, measure that this think tank did of these individuals and their leadership ability and their voice pitch? What would you think?

VEDANTAM: I would say there would be no correlation whatsoever.

KLOFSTAD: You are absolutely correct. There is nothing - absolutely nothing. And then we did a follow-up study where we did an experiment online whereby we randomly assigned liberal and conservative experimental subjects to listen to liberal and conservative persuasive messages, things like same-sex marriages or gun control, you know, things of this nature. We randomized it all about. No effect there either, that the timbre of the messenger's voice had no effect. So in real life and in the experimental world, we find that voice pitch actually does not have an influence on leadership capacity.


VEDANTAM: Casey Klofstad is a political scientist at the University of Miami. His work on the relationship between the human voice and politics was conducted with his wife, Rindy Anderson. She's a biologist at Florida Atlantic University.

This episode of HIDDEN BRAIN was produced by Thomas Lu and Parth Shah. It was edited by Tara Boyle, Camila Vargas Restrepo and Rhaina Cohen. Our team includes Jennifer Schmidt and Laura Kwerel.

Our unsung hero this week is Sam Turken. During this interview, I was in a studio in Washington, D.C., and Casey was at his home in Florida. We talked over the phone. Sam went to Casey's house to record his end of the conversation. We then combined the two sides of the interview. In audio journalism, this is known as a tape sync, and it's an integral part of how public radio gets made. It takes patience and skill to record a tape sync. If you thought the conversation was clear today, you have Sam to thank.

If you liked today's episode, please share it with your friends. And think about the voices you hear in your workplace and in your community. Ask yourself, whose voices are missing because they don't sound the way important and influential people are supposed to sound? I'm Shankar Vedantam, and this is NPR.

(SOUNDBITE OF MUSIC) Transcript provided by NPR, Copyright NPR.


Subscribe to the Hidden Brain Podcast on your favorite podcast player so you never miss an episode.

google podcast subscribe
spotify podcast subscribe


Go behind the scenes, see what Shankar is reading and find more useful resources and links.