Think you can tell how old someone is by listening to them talk? Test yourself with our quiz: “Guess a person’s age from their voice“
THE YEAR was 1927. The BBC had only passed its fifth birthday and radio broadcasting was still a novelty. With radio plays wafting through the airwaves into almost every living room, British psychologist Tom Hatherley Pear wanted “to discover what actually goes on in the minds of different listeners” as they tuned into programmes “presenting the voice and nothing besides”.
So he recruited 9 people – ranging from his 11-year-old daughter to a judge and a minister – to read aloud on air a passage from The Pickwick Papers, in which Dickens describes a comically unsuccessful outing on ice skates. The readings were broadcast on BBC stations across the UK on three consecutive nights in January, and listeners were asked to cut out a form from the Radio Times, fill it out describing their impressions of the speakers, and send it in to Pear’s team.
Advertisement
The experiment captured the public’s imagination: nearly 5000 people responded, some with highly detailed descriptions of the people they had heard. Whether they were right or wrong, what was striking was the strength of impressions they had formed. Many participants happily provided full backstories for the voices in question. Of the minister who read on air, one listener wrote: “I should say he has suffered considerably and is very sympathetic… I would imagine him as being tall and cadaverous, round-shouldered with a long neck and protruding chin.”
Pear had identified something that we perhaps all know instinctively, that the voice can be powerfully suggestive. Whether you are eavesdropping from another room or taking a phone call at work, the way someone speaks can paint a clear picture of a person, their personality and even provide a sense of their history. But often, you won’t be aware of this, nor of the way these impressions are influencing your behaviour.
“There’s so much going on with voice, and we don’t think about it,” says , co-director of the Voice Perception Laboratory at the University of California, Los Angeles. “Your voice is your auditory face.” Kreiman is among a number of researchers who are lending an ear to the secrets of voice. It turns out that some of the most subtle effects are also the most potent; they can influence your sexual allure, your political credentials – even your salary.
Much of the recent work has built on Pear’s original findings to pin down the cues that shape our mental pictures of other people’s physical characteristics. Consider the way we form an impression of someone’s gender from the way they speak. As you might expect, the most obvious clue is pitch – an association with a long evolutionary history. The males of many species tend to have longer vocal tracts than the females, giving a deeper sound that makes them seem bigger to potential rivals and mates. That is probably the result of – with females’ preference for more imposing-sounding mates driving the evolution of the male’s vocal tracts.
In humans, these gender differences can be striking. Men’s vocal tracts are up to 20 per cent longer than women’s, and men also have larger vocal cords – or vocal folds – causing them to speak about an octave lower, on average. Supporting the idea that this is due to sexual selection, studies show that women tend to find men with more attractive, while men prefer higher voices in women. Indeed, one team recorded women’s voices at different stages of the menstrual cycle and found that the pitch rises slightly, and it takes on a subtle kind of shimmer, in the – possibly enhancing a woman’s sex appeal.
Yet we are so accustomed to linking pitch with gender that it can be easy to overlook subtler cues that may also be important. For instance, a study of men undergoing gender reassignment found that two voices speaking at exactly the same frequency can sound either masculine or feminine simply by the way they pronounced the sibilant . As well as variations in pitch, such differences may explain why actors can rarely switch gender convincingly – often to comedic effect (think Julie Andrews in Victor Victoria or Robin Williams in Mrs Doubtfire).
After gender, the next easiest characteristic to read from a voice may be the speaker’s age. As we get older, we tend to speak more slowly, and eventually decreasing muscle tone can mean we produce weaker, breathier sounds when speaking. We hardly have pinpoint precision, though. Kreiman has found listeners in her lab were typically able to estimate a speaker’s age to within about 10 years. There were some notable exceptions, though. One 3-year-old was identified by some listeners as the youngest speaker and by others as the oldest. “It’s actually my niece and she had just had her tonsils out, so she had all the hallmarks of an elderly voice,” Kreiman explains. “She had imprecise articulation because she’s 3, she’s very hoarse, she’s breathy, and she’s talking slowly.”
The sound of a hairstyle
What about reading more specific aspects of someone’s appearance from the way they speak? As Pear’s volunteers showed, our impressions can be surprisingly detailed. Based on the voice alone, one listener described a police detective as “a big, stoutish man, with medium-coloured, rather unruly hair”. Lest you wonder what he meant by this, he continued: “By unruly, I don’t mean a curly, tousled mop, but hair that won’t go the right way and is unaccustomed to brilliantine.”
When forming these pictures, we may be tuning into the way the voice’s sound is transmitted through the body because the shape of the lips, jaw, nose and chest influence where the sound waves resonate and amplify. And to a certain extent, it works. We can approximate height based on voice alone, but our best guesses are still off by about 10 centimetres, Kreiman says. And while we may think voices convey subtle hints about the contours of people’s faces, it is very difficult to pair up photos of faces with the voices that go with them. During the study that found the biggest effect, participants were able (from a selection of people of the same gender) only slightly better than chance.
Things get even murkier when we try to read the same cues for psychological, rather than physical traits. Despite our confidence in our judgements, they are often based on crude biases. And as with other types of bias, the consequences can be troubling.
One of the more telling examples again concerns the pitch of the voice. Besides its role in sexual attraction, pitch can also send out signals of other qualities. For both sexes, . That can be a boon for men with a Barry White growl: an examination of nearly 800 male CEOs of US companies found that other things being equal, those with deeper voices tended to be in charge of larger firms and accrue around , compared with men with higher voices (see diagram).
But the situation is more complicated for women. As with men, speaking in a deeper voice makes women seem more powerful and assertive, but that may be at the expense of their perceived attractiveness. , a political scientist at the University of Miami in Florida, was particularly intrigued by the way this could influence elections: “Is it going to be perceptions of competence and strength that drive our vote choices, or perceptions of attractiveness?” His suggest it is the former. After listening to recordings of female candidates soliciting votes, subjects consistently preferred – and said they would vote for – the women with the deeper voices.
The power of a deep voice was no secret to Margaret Thatcher’s advisers, who helped the once higher-pitched British politician develop a more stately tone. “Thatcher had a completely manufactured voice,” says , a voice coach who has worked with actors on films such as The King’s Speech. Honing a speaking voice may be particularly problematic for female politicians in the US, says Klofstad. “On average, Americans are politically disengaged and we choose leaders based on impressionistic judgments,” he says – so factors like the pitch of someone’s voice matter. Hillary Clinton may want to take note if she is to run for president in 2016, he adds.
Cultural cues
Such implicit associations might even be driving large-scale changes in the way we speak. For example, a 1995 study that compared the pitch of women in Japan and the Netherlands found that . This reflected cultural values, the researchers concluded, noting that traditional gender roles – with an emphasis on the man as the breadwinner and protector – were very highly valued in Japanese culture at the time, but less so in Dutch culture. Studies in Sweden, the US, Australia and Canada have also shown that in those countries since the 1950s – by more than 20 hertz, roughly the equivalent of moving between the “middle” C and B on a piano.
“Women’s voices have grown deeper in Sweden, the US, Australia and Canada since the 1950s”
Perhaps nothing shapes our voice-based judgements of a person the way accents can – even if our assumptions can lead us wildly astray. During Pear’s study, for instance, half of the people who responded were convinced that the police detective reading the Dickens extract was a farmer – presumably because he spoke with a more rural accent.
Our reliance on accents as cues probably had some evolutionary benefit: according to one view, linguistic differences emerge to help us define our cultural identity, so that we can tell our in-group from out-groups. Some evolutionary anthropologists suggest that distinctive vocalisations may have .
It seems that we absorb these associations at a very young age. University of Chicago psychologist recently presented white 5-year-olds with photos of other children of the same age. She found that, when no one was speaking, they tended to prefer those who shared the same skin colour as “friends”. But when the children were given recorded voices of native English speakers or those who spoke with foreign accents alongside the photos, they , regardless of ethnicity. The preference tends to shift towards skin colour as children age but, as Kinzler says, it nevertheless illustrates “the power of the voice as a really important feature of identity”.
The connotations of a particular accent are complex and can influence perceptions of prestige, attractiveness and intelligence. Often, we seem to be more indulgent of these biases, compared with other kinds of prejudice. “If a parent in my lab hears his or her child express race-based attitudes, they become very uncomfortable,” Kinzler says. “But when kids say they like someone who speaks in the same accent as they do, parents don’t have the same kind of negative response.”
Yet like other stereotypes, biases based on accent can unfairly swing important decisions in places such as the courtroom. In one study, based on a voice recording, people were if he spoke in a regional Birmingham accent rather than a more standard English accent. (People who had “Brummie” accents themselves were excluded from the study.) Also, when people have thick accents that may make them harder to understand, we are less likely to trust what they have to say. Researchers asked native speakers of English, and non-native speakers who spoke English with either a mild or heavy accent, to record themselves reading statements of trivia. When listening to these recordings, – even when they were told that all of the trivia was provided by the researchers, not the speakers.
It may seem surprising, but one way to overcome these prejudices may be to try imitating another voice. To some extent, we may already do this naturally. According to one theory, we understand others by going through the motions of their speech ourselves: our brain’s motor cortices kick into action, showing the same kind of activity as if we were actually saying the words.
Simply tuning in to unfamiliar accents may not only help us to decode the new sounds, but also lose our assumptions about the speakers. Patti Adank, now at University College London, and colleagues asked people who had never lived in Scotland, and were not often exposed to Scottish accents, to mimic speakers from Glasgow. Before and after, participants rated the attractiveness of the Glaswegian accent. Assessments of power and competence remained the same, but after mimicking the accent, . “After we asked people to imitate people from Glasgow, they liked them a bit more,” Adank says. She suggests this reflects an in-group/out-group effect. “If you’re being put in an out-group’s perspective, that works as a levelling factor.”
While prejudices remain, many people turn to elocution lessons – and not just to remove traces of a regional accent. Voice training can help transgender individuals, for instance, learn the subtle differences in enunciation between men and women. Others are turning to “voice-lift” therapy to remove the breathy or gravelly tones that can signal age.
Whatever your reason, perfecting a new voice takes time, says Berkery, who worked with the actor Renée Zellweger for two months to help her prepare her English accent for Bridget Jones’s Diary. “For Renée it was a 24/7 occupation,” Berkery says. Put in the time, however, and you will find that your voice is more flexible than you might imagine. “There are physiological limits, but they’re pretty broad,” says Kreiman.
Perhaps the biggest hurdle will be psychological. Our voice has grown with us since we first learned to talk. As Kreiman points out, it is as much a part of our identity as our face. In some small way, changing it means becoming a whole new person.
This article appeared in print under the headline “Voice almighty”
