← Can You Learn It?

What Is Absolute Pitch?

A working definition: absolute pitch (AP) is the ability to identify the musical note of an isolated tone — or to produce a named note — without any external reference. Most musicians can’t do this. A small minority can do it almost effortlessly.

The fruit-market analogy

The clearest description in the literature comes from a 2005 review by Daniel Levitin and Susan Rogers. Their analogy is not about what having AP feels like — it is about what happens when someone with AP hears a familiar piece played in the wrong key:

“[AP possessors hearing a familiar piece in the wrong key] often become agitated or disturbed. To get a sense of what it is like, imagine going to the produce market and finding that, because of a temporary disorder of visual processing, the bananas all appeared orange, the lettuce yellow, and the apples purple.” — Levitin & Rogers, 2005, p. 27 Levitin 2005

Note what the analogy is actually about: it describes the discomfort of hearing music transposed, not the experience of possessing AP. And the same review is explicit that AP possessors do not perceive pitch categorically the way we perceive colour — that idea, it says, “was subsequently refuted”: category structure affects their ability to label a pitch, not their perception of it. The difference is in naming, not in hearing.

Spectrum or yes/no? The field is split

The popular picture is “has it / doesn’t have it.” The measured performance is clearly continuous. Whether the underlying trait is continuous is a different question, and the papers in this collection sit on both sides of it.

Miyazaki’s 1988 study was designed to separate AP possessors from non-possessors cleanly, and it did — though even there a third, “imprecise” group emerged that his own abstract doesn’t mention. The same study documented the “white-key advantage”: identification is more accurate for the white keys of the piano than for the black ones Miyazaki 1988. Two decades later, Bermúdez and Zatorre tested 51 musicians with a self-paced task and found “a range of behavior from perfect to random performance … including an important number of intermediate-level performers” — in declared tension with the bimodal reading of Miyazaki’s data. They add, honestly, that “this remains a controversial point” Bermúdez & Zatorre 2009.

The strongest evidence for gradation is unglamorous: the “non-AP” comparison groups in the literature systematically score about twice chance. Bairnsfather’s review puts the pooled non-AP mean at 17.0% on raw scoring (24.5% in the studies that give credit for semitone errors) against a chance level of 8.3%, and notes that “above-chance participants are sometimes included in the ‘non-AP group’, which reduces the discriminatory power of studies.” Plenty of people have some of this ability and are counted as having none.

And the picture got even more complicated in 2025. A systematic review by Bairnsfather and colleagues looked at 160 studies covering more than 23,000 participants and concluded that the field broadly agrees on what AP is — and diverges on nearly everything about how to measure it. Only 59% of the studies stated any accuracy cutoff at all. Among those that did, cutoffs ran from 20% to 100% (mean 77% on raw scoring, 71% where near-misses got partial credit — two scales the authors warn “are not directly comparable”). Their own summary of the pooled data: “pitch-naming ability is a dimensional trait, with scores lying along a spectrum from chance to ceiling, regardless of participant classification into AP and non-AP groups” — while adding that whether the phenotype is discrete has not been established, and proposing taxometric analysis to find out Bairnsfather 2025.

So when someone says “I have absolute pitch,” it’s reasonable to ask how absolute. Within natural variation, plenty of musicians possess strong pitch memory or partial AP without meeting the strictest definitions.

How rare is it?

The most-cited estimate is that AP appears in roughly 1 in 10,000 people in Western populations — a figure that predates modern testing and that none of the studies collected here measures directly. What has been measured: in the 1998 survey by Baharloo and colleagues, 15% of 612 trained musicians reported having AP (the 40% sometimes quoted is only the subgroup who began training at age 4 or earlier). Even in an elite musician sample, the large majority do not have AP after years of immersive training Baharloo 1998.

That’s a key data point: musical training is not enough on its own. The authors concluded that early training appears necessary but not sufficient, suggesting a gene-environment interaction.

What about the brain?

Brain imaging studies of people who already have AP report anatomical and functional differences — repeatedly, though not always convergently. Zatorre 1998 set out to replicate the planum temporale asymmetry and did not: it found a larger left PT against a reference sample, but “did not find an exaggerated asymmetry per se”.

• Asymmetry in the planum temporale. Schlaug and colleagues scanned 30 professional musicians, 11 of whom said they had AP — the paper states plainly that “no formal test for perfect pitch was applied.” Those 11 showed about 2.5 times more leftward asymmetry of the planum temporale, an auditory region, than the musicians without AP; the musicians without AP did not differ from non-musicians at all Schlaug 1995.
• Specific white-matter pathways. Loui and colleagues used diffusion tensor imaging and found thicker tracts between auditory regions (posterior superior and posterior middle temporal gyrus) in AP musicians in both hemispheres — but only the left tract’s volume tracked how good each person’s AP actually was. Their words: “local hyperconnectivity in both temporal lobes may be one of the prerequisites,” with early training shaping the left more than the right Loui 2011.
• Automatic labeling. A PET study by Zatorre and colleagues found that AP musicians spontaneously activate left dorsolateral frontal cortex when hearing tones — areas associated with verbal labeling. AP, in this view, isn’t a separate brain system; it’s the automatic engagement of a general labeling network when processing pitch Zatorre 1998.

An important caveat: every one of these studies looked at people who already had AP. They show what AP looks like in the brain — not whether the brain can be changed to produce it.

Genes, environment, and language

Three threads run through the evidence on origins.

Genetics matter. A 2009 genome-wide linkage study by Theusch, Basu, and Gitschier found a significant signal on chromosome 8q24.21 (LOD 3.464) in the 45 families of European ancestry — in the 19 East Asian families no peak reached even the suggestive threshold, and there was no signal at 8q24.21 — though the abstract notes that one of the four regions, 7q22.3, was also evident there. The locus heterogeneity in the paper’s title comes from somewhere else: within the European sample alone there were more independent peaks than chance allows (p = 0.0042). The contrast with the East Asian families is a null result on 19 families — compatible with the genes differing between populations, but not a demonstration of it. Either way, AP isn’t a single-gene trait Theusch 2009. There’s a heritable predisposition, but it isn’t a switch.

Early training matters. The Baharloo finding that AP tracks age of musical onset was reproduced with a stricter, scored test Deutsch 2006. Takeuchi and Hulse propose, explicitly as speculation, that some kind of pitch labelling has to happen during childhood for AP to crystallize — in Takeuchi and Hulse’s formulation, “AP may only develop if early musical training includes the association of pitch names with particular absolute pitches” Takeuchi & Hulse 1993.

Language matters too. This is one of the most striking findings in the field. Diana Deutsch and colleagues showed that speakers of tone languages — languages where pitch changes word meaning — have far higher rates of AP. In one study, about 60% of Chinese conservatory students who began musical training between 4 and 5 years of age met the strict AP criterion, compared with about 14% of US-trained students from that same age range (both read from the paper’s Figure 1A; the study reports prevalence per age-of-onset band, not overall) Deutsch 2006. A separate study found that tone-language speakers carry a stable absolute pitch template for the words they say: asked to read the same list on two different days, Mandarin speakers were as consistent across days as across 20 seconds, while English speakers got significantly worse across days. All seven Vietnamese speakers tested stayed within 1.1 semitones of themselves, two of them within a quarter of a semitone Deutsch 2004.

The authors are careful about what this does and doesn’t show. The two student bodies differ in language and in ancestry, and the design cannot separate the two — the paper says so: “it may alternatively be proposed that the differences between the two groups obtained in the present study were due to genetic factors… Indeed, both critical period and genetic factors might be involved.”

If language exposure during childhood already trains a kind of pitch precision, the line between “trained” and “born with it” gets fuzzy. That is the honest state of this thread: fuzzy, not resolved.