When Humans Hear Music, Monkeys May Hear Noise
When Humans Hear Music, Monkeys May Hear Noise

When Humans Hear Music, Monkeys May Hear Noise

The auditory cortices of humans and rhesus monkeys respond very differently to harmonic tones.

Oct 1, 2019
Katarina Zimmer

ABOVE: © ISTOCK.COM, LUXIANGJIAN4711

If there’s one thing Bevil Conway has learned from studying the visual cortices of rhesus macaques, it’s that they’re remarkably like those of humans. The visual cortex is anatomically highly similar in the two species, and macaques and humans show comparable behavioral and neural responses to colors and images. When a macaque opens its eyes, “I’m pretty sure he’s seeing what I’m seeing,” says Conway, a neuroscientist at the National Institutes of Health (NIH) in Bethesda, Maryland. But does the same hold true for what he hears? 

The question came up in 2014 over a beer with Sam Norman-Haignere, then a graduate student with Josh McDermott and Nancy Kanwisher at MIT, where Conway headed a lab at the time. Norman-Haignere told Conway about his groups’ recent collaborative finding that a particular patch of the human auditory cortex is more sensitive to harmonic tones—notes that have an easily discernible pitch, such as those played on a piano—than to noisy sounds, such as those made by a drum. When a dozen people listened to harmonic tones in a functional magnetic resonance imaging (fMRI) scanner, this patch of the auditory cortex flared up with heightened neural activity—substantially more than when the participants listened to noise. 

Musing whether there was an equivalent region in monkeys that would also respond selectively to harmonic tones over noise, Conway and Norman-Haignere made a bet. “I thought, I’m pretty sure that monkeys are going to have that,” Conway recalls. Given that the visual systems of humans and macaques are so similar, he figured that their auditory cortices would be as well. Norman-Haignere wasn’t so sure. Humans rely on harmonic tones not only in music, but also to articulate vowels in spoken language. Because monkeys lack these cultural aspects, the animals may not need a brain region devoted to perceiving harmonic sounds, reasoned Norman-Haignere, who is now a postdoc at Columbia University.

Along with McDermott and Kanwisher, the two researchers set out to repeat Norman-Haignere’s earlier experiments, this time comparing the human brain’s response to that of the macaque (Macaca mulatta). To find out if the two species’ auditory cortices respond selectively to harmonic tones, the researchers used computer software to synthesize two sets of sounds. The first set of stimuli was harmonic, based on short sequences of up to a dozen notes differing in melody and the range of audio frequencies—that is, the range from the lowest note to the highest note in the sequence. The second, control set of stimuli was noisy, created from a scrambled kind of sound much like static on the radio or TV. The team designed the noise to match the frequency ranges of the harmonic stimuli and thus ensure that the sound sets were comparable. 

Three macaques and four people took their turn in an fMRI machine as the scientists played the sounds inside the scanner and monitored each participant’s auditory cortex. As observed in the earlier experiments, the human auditory cortex showed significantly greater activity in response to the harmonic stimuli than to the noise. But to Conway’s surprise, the same brain region in macaques showed no significant differences in response to the two sets of sounds. In fact, if anything, the monkeys’ brains often had a greater response to noise than to harmonic tones. 

Andrzej krauze

The contrast between macaques and humans couldn’t have been clearer, Conway says, “but then we had to spend about two years full-time collecting data in order to convince ourselves that the result was real.” 

Initially, the team wondered whether the monkeys’ lack of selective response to sounds with a specific pitch could be simply due to their unfamiliarity with synthetic sounds. So the researchers repeated the same experiment, this time playing recordings of real macaque calls and computationally engineered, pitch-less, or noisy, versions of those vocalizations. Even then, the fMRI results revealed that human auditory cortices were much more sensitive to the pitch quality of those sounds than macaque brains were. “I think I owe Sam a beer,” Conway remarks. 

What makes the finding all the more surprising is that macaque brains are perfectly capable of detecting differences in audio frequency. Like humans, macaques have a “tonotopic map” in their auditory cortex, a pattern of brain regions in which each region responds to a specific range of frequencies, although the team’s fMRI data, consistent with earlier research, indicate that this map is arranged slightly differently in macaques than it is in humans. The monkeys just don’t show a preference for sounds with a specific pitch in the way that humans do. 

To Conway, the findings may explain why previous researchers have so far failed to find neurons that respond selectively to sounds with pitch in macaques, and have had a hard time training macaques and other monkey species, such as capuchins, to perform certain auditory tasks that humans excel at. “The explanation has historically been that the monkeys’ memory cortex works differently, but our findings say no, maybe their memory systems are just fine, but the auditory cortex works differently.” 

The heightened responsiveness of the human brain to pitch may well have to do with the origins of spoken language as well as music. “Harmonic structures are very important to humans,” says Charles Snowdon, a primatologist who recently retired from the University of Wisconsin–Madison. “We’re producing very clear harmonics [with] every vowel sound that we use, and we can’t speak without producing vowels.”

Macaques do use some harmonic vocalizations, for friendly calls and chatter in particular, but they use them far less frequently than do humans. “The macaques make a greater use of noisy signals and noisy vocalizations in their own communication and therefore it makes sense that they would have significant amounts of brain tissue devoted to processing noise,” Snowdon says. Humans can make vocalizations that are noisy too—such as growling—but they are rarely used, “because we have language instead,” he adds. 

But whether humans are unique among primates in having a preference for harmonic sounds—what Conway and his colleagues call “pitch bias”—will require further research in other primate species, Snowdon says. Conway and his colleagues plan on examining tamarin monkeys (genus Saguinus) next, which make many vocalizations that have harmonic structure. 

For now, Conway is left to wonder what a macaque’s experience of the acoustic world may be. What would music sound like to a macaque? It’s impossible to say, but he imagines a recent performance by Hong Kong musical artist Samson Young might offer one indication. Young had an orchestra play a “muted” version of Russian composer Tchaikovsky’s Fifth Symphony. “He taped all of the strings on the instruments so they couldn’t make any pitches, and all that’s recorded is the noise,” Conway says—the swishing and tapping of bows against tape. “I think that’s probably what it sounds like to the monkey.” 

Katarina Zimmer is a New York–based freelance journalist. Find her on Twitter @katarinazimmer.