What Perfect Pitch Tells Us About the Brain

erfect pitch, also known as absolute pitch, is the ability to identify a musical note from memory. Ask someone with perfect pitch to hum an A-flat and they’ll usually get it within a few Hertz. It’s not only useful for musicians: people with tinnitus have found that their ringing varies in pitch depending on its severity.

Everybody has this ability

At one time it was believed that absolute pitch was genetic and that early musical training was necessary for it to develop. Wong et al. ^[1] tested that theory and found that after 15,327 training trials, about the same number as it takes to learn to play Chopsticks, some adults with no early training could identify 7.08 different pitches (on average) with 90% accuracy.

This means everybody may have this ability without realizing it. I realized this myself when I was a kid. In ninth grade, our school screwed up everybody’s schedule and we ended up with three study hall periods every day. I passed the time by playing back musical tunes in my head.

(I don’t recommend this. Trust me, you do not want a Bach organ piece running through your head over and over for the rest of your life. Or worse: Tom Jones singing “What’s new, pussycat, yowa yowa wowwa.” Torture is not the word for it.)

But my torment taught me that when you play things back in your head as a recording, you also remember the correct pitch. This means if you happened to know that some tune starts with an “A” (as the Bach one does), you can always identify an A when heard. That’s all it takes because you can compare other notes to it.

Auditory cortex

Thus, the brain may have two different ways of remembering sounds: by remembering the relationship among the notes, which means relative pitch, or by ‘recording’ the sound, which automatically gives the correct pitch. It may be that earlier experimenters deliberately disrupted the absolute pitch in their attempts to rule out guessing.

Pitch is mapped onto the auditory cortex, which is how we can hear different pitches. The general term is tonotopic organization, which means that nearby regions of the cortex are topographically mapped to similar sensory signals. In general, low frequencies are mapped to more lateral areas and higher ones to medial areas. The human auditory cortex is different from the one in bats, which have multiple specialized mechanisms for processing FM sweep rate and Doppler shifts.^[2] In humans, individual neurons vary widely in frequency bandwidths.^[3] The human auditory cortex is not just a spectrum analyzer, and a lot of processing is done by surrounding secondary areas and higher-order areas which process more complex sounds.

In fact, there are as many as 15 different cortical auditory areas. They process “what” and “where” information via pathways to other brain areas. Functional imaging including fMRI and diffusion tensor imaging (DTI, a type of MRI) have provided much information on these areas.^[4]

Frequency resolution, or pitch discrimination, is the smallest difference in frequency that can be detected. This has to be measured sequentially to avoid detecting beat frequencies, which piano tuners use to achieve precise tuning. Typical results show that roughly 5 Hz is the best that can be discriminated. This is not just due to the finite number of cochlear hair cells. There’s a fundamental trade-off between sound duration and resolution. Even so, formal music education enhances pitch discrimination both in children and adults.^[5,6] So, even if early training improves it, absolute pitch would be limited to about 20 cents, or roughly 1/5 of a semitone.

Tonal languages don’t help

There’s another myth that native speakers of tonal languages like Finnish and Chinese, in which pitch is a lexical component, are more likely to have absolute pitch. This is a misunderstanding of how those languages work. The 6 (or 8 or 9, there is fierce debate) tones in Cantonese and the 8 in the Minnan dialect, for example, are always relative. A high tone for one person could be a low for another. At most there are three or sometimes four distinct pitches. The direction of the change in pitch is as important than the pitch itself.

The sharp linguistic variations in pitch often give Westerners the impression that Chinese persons are always arguing when in fact they may be merely discussing something calmly.

fMRI and DTI Imaging studies

There are some Youtubers out there digging up old complaints about functional MRI (fMRI), saying it’s some kind of disaster and everything we know about neuroscience is wrong. This is a wild exaggeration. fMRI, if done carefully, can be very useful in studying brain function.

A good example was when fMRI showed that persons with absolute pitch used the left superior temporal sulcus of the brain, which is known to be involved in categorization tasks, when encoding a pitch.^[7] These fMRI findings confirm MRI and EEG results. Humans also use the left posterior dorsolaterofrontal cortex and the auditory association cortex ^[8] when hearing isolated tones. Moulton ^[7] concluded that relative pitch and absolute pitch are complementary. Both are needed, he said, and both require years of training. He wrote:

In this respect, non-AP [absolute pitch] possessors are entitled to view their AP possessor colleagues as musical equals.

Indeed, absolute pitch can be a disadvantage. People with absolute pitch take longer to name tones that are out of pitch, but exact frequency measurement, which would help to identify the listeners’ strategy, is mostly lacking in papers in this field.

A professional piano tuner once told me that perfect pitch is not all or none. He could tell me a note is a B-flat, but without a tuning fork he could not say whether my B-flat is flat or not. Piano tuners describe pitch in terms of cents, where 100 cents is the difference between two adjacent chromatic notes. Adjacent notes differ in pitch by a factor of the twelfth root of 2, or 1.059463, so one cent is a factor of 1.0005778 in pitch, or a 0.2542 Hz difference at 440 Hz. Absolute pitch could would never be this exact because the cochlea doesn’t have this level of resolution. It would also be impractical because analog recordings and acoustic instruments show variations in pitch bigger than this.

Frequencies of musical notes are not set in stone but vary in pitch depending on the temperament being used. If your A is set to 440, your 60 Hz electric house current would be closest to a B1, which is 61.74 Hz. But sometimes A is set to 432, which is 1.8% lower, making the first B 60.6 Hz. Some instruments have even more variation due to inharmonicity, which means that the pitch of an A on one side of the keyboard is not an exact multiple of an A on the other side.

Shibayama et al.^[8] found that absolute pitch only worked in low-conflict contexts like C-major. When hearing something in B-major, they failed, reinforcing the idea that absolute pitch is mostly learned but can be stored in different ways, resulting in people using both absolute and relative pitch simultaneously.

Another factor could be that musical instruments don’t produce pure tones. Even those electronic chromatic tuners often have trouble measuring a pitch. Mine, for instance, is not sure whether my refrigerator is playing a B or a G-sharp. Maybe I need a fridge tuner to come over and take a look at it.

nov 10 2026, 5:12 am