Musical Harmony May Be Hardwired in the Brain

By Lisa Raffensperger | November 16, 2012 1:34 pm

Man with parabolic ear attachments device Anyone who’s ever watched a horror film will know that the sound of two clashing notes evokes a visceral response in most people. Among Western listeners there’s a strong preference for consonance, which exists even from infancy; consonance is the pleasing mixture of two tones, while dissonance is their clashing. (For a good example of both, see this video.) It’s controversial whether the same preferences exist in other cultures, but new research indicates the preferences might be wired in our brains.

The prevailing theory of music in the brain is that dissonant combinations share frequencies that are a bit too close. When these frequencies are perceived by the cochlea, the part of the inner ear that translates sounds to nerve impulses, they can’t be well distinguished. Because similar frequencies are processed next to one another on the cochlea, their nerve signals can interfere with one another. The perception is a grating effect, called “beating.”

But neuroscientists have found that beating is not actually what we dislike when we hear discordant tones, as they reported this week in PNAS. Their approach was to study a unique subset of people called amusics. Amusia is a congenital disorder; people with the condition have a hard time detecting the direction of pitch changes and can’t tell some notes apart. Thus, many of the sensations of music are lost on people with this condition. The researchers played various combinations of tones to the amusics, many of which normal listeners would find quite unpleasant, but the subjects reported no displeasure. The one exception, however, was beating. Amusics could indeed detect beating and found it grating, just as other people do. When two notes of close frequency were played to the same ear, subjects rated that as jarring; when, in contrast, they were separated one note to one ear, the beating was lessened and so too was the subjects’ dislike of the sound.

Thus beating is unpleasant to all of us, but doesn’t appear to have anything to do with our appreciation for music, since amusics dislike it, too. That appreciation appears instead to be hardwired in our brains, reliant on our ability to accurately detect the gaps in frequency between two notes, rather than just the dislike of beating. The finding helps resolve a key question of how and why we appreciate pleasing sounds. And that should be music to your ears.

Photo courtesy of jeanbaptisteparis via Flickr

MORE ABOUT: audio, brain, harmony, hearing, music, PNAS
  • Callum Hackett

    As convincing as this study may seem, it actually has a number of serious flaws that have been pointed out in various places by researchers in the field, though these have been frustratingly ignored in favour of pretending that this study offers an actual answer to a difficult question (time and again, people acknowledge complexities in physics and biology, but when you bring up the neuroscience of art, everyone acts like the latest fad explains everything).

    One of the most obvious problems with the study that you acknowledged is cultural difference – the styles and nature of music around the globe are so enormously diverse that the results found here would likely not hold if the participants were selected from a wider field. But let’s pretend for a moment that that’s actually not a problem. There is, in fact, an even more fundamental issue with the study in that it uses the Western dodecpahonic scale as a basis for testing – the 12 notes that we all know as C-C#-D-E-F-F#-G-G#-A-A#-B. Given that we’re all brought up learning that this is *the* basis of music, many of us are led to the mistaken belief that this is somehow an inevitable musical construct, but it has only been around since the turn of the 18th century.

    So why does it undermine the study? Well, it’s right that some of these intervals – such as C-G, a fifth – have frequencies related by simple mathematical ratios which allow for the overlap of harmonics. However, the scale is actually not optimised for this. Each semitone is related to the last by the twelfth root of two, and that means that some of the intervals approach simple ratios, but are actually out by a bit. Why? Because compositional technique was developing towards the end of the Baroque period to allow for a greater variety of key modulations, but the tunings of the instruments severely restricted this ability (if you wanted to transpose a tune, you couldn’t just change your fingering, you had to retune the instrument or get a different one). So equal temperament (the twelve tone scale we know and love today) was eventually introduced to solve this problem, but it could only do so by compromising the frequency relationships of notes when compared to earlier Pythagorean tuning and just intonation which we were fundamentally based on simple mathematical ratios for the entire scale without adjustments.

    Thus, under these researchers’ hypothesis, those earlier scales which preceded the ones we use today ought to be the most consonant to a listener. But what do we find? They’re reported as sounding a little off and out of tune, and that’s because the participants haven’t been inculcated with these scales since birth – the conclusion that harmonics are the key must be fallacious. I suspect that the apparent connection between harmonics and consonance is a correlation, not a cause, itself related to widespread (and arbitrary) harmonic progressions which pervade Western music, and which people are brought up listening to and so are programmed to find enjoyable.

  • Lisa Raffensperger

    That’s incredibly interesting, Callum – I hadn’t come across that in my research, but an interesting bit of physics and history too. Thanks.

  • Cathy

    I’d also like to add in that the “beating” of two notes is well detected by trained musicians; indeed, it is how one musician tunes his or her instrument to match another’s. The concertmaster usually tunes to an electronic device, and the rest of the ensemble or orchestra tunes to that instrument. The ears of a musician are so sensitive to this tuning process that he can pick out his own frequency among the dozens of notes around him. And the tuning is done to minimize the “beating”, or in small groups, attempt to eliminate it entirely. At that level it is no longer just an unpleasant sound, but also a musician’s tool.

  • chiangraiken

    The article starts out by mentioning ‘western listeners’ have a strong preference for ‘consonance.’ Does that infer that ‘eastern listeners’ have less of a preference? I’m a western musician but have resided in Thailand for 14 years. Every day, when I’m in and around town, I hear loud and annoying sounds (I may have a slight bit of tinnitus). Even so, I’ve noticed that Asians never seem to be bothered by ‘consonance’ (as you call it). There can be times when two sound systems in the same space are blaring two different tunes concurrently, and/or there may be gross distortion in the loud sound(s) and not one Asian within earshot will raise an eyebrow. Perhaps a study of that could be an off-shoot of the interesting article above.

  • John Lerch

    How about the possibility that dissonances are just too hard to sing? Since I usually sing bass (although I can sing alto and tenor), I usually have the primary note of the chord or at least that other chord whose name I forget with the 5th below (the 3,4,5 chord rather than the 4,5,6 chord) so everyone has to key off me. But the altos frequently have a dissonance with us basses and they have a hard time knowing that they’re right even when they’ve learned it really well.
    They should have tested for the pure tone dissonance between 2 tones an octave part–C2,B3 e.g.

  • Neil

    One of the joys of modern electronic instruments is that one is not compelled to stick with 12-note octaves – it may be impossible to play, but one can certainly program music on a 19- or 31-note scale. See:

    for the maths and background respectively.

  • geack

    @1. Callum,
    Not questioning anything in your post – but wasn’t all of that exactly what the work with atonics is intended to counter? They are the closest thing we’ll find to people who lack cultural music conditioning, and their consistently higher discomfort with dissonance in one ear rather than split between both ears supports the idea of a bio/neuro basis for the preference. This appears to be an example of research that makes a well-planned attempt to avoid the superficiality which you’re bemoaning in your first paragraph.

    I have absolutely no direct expertise in the subject, but the research as described appears to address your criticism. Can you expand on your thoughts?

  • Cardillac

    interesting concepts, but how would they fit into music before there wasn’t any harmony/chords/tonality (start with Gregorian Chant)-

    not to mention all of this before the “Well-tampered-with” Klavier became the standard of Western music-

    social conditioning determines what sounds we perceive as either pleasant or unpleasant


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


80beats is DISCOVER's news aggregator, weaving together the choicest tidbits from the best articles covering the day's most compelling topics.

See More

Collapse bottom bar