Conservatory study linked earlier training and Mandarin background with perfect pitch
A conservatory comparison links absolute pitch with both lesson starting age and language background, but cannot untangle language from educational and cultural differences.
Key findings
- Among students starting musical training at ages 4–5, 60% of the Beijing group passed an absolute-pitch test, compared with 14% of the American group, according to Scientific American's contemporary report. [2]
- The published comparison included 88 Mandarin-speaking conservatory students and 115 non-tone-language-speaking students. Earlier training was associated with higher absolute-pitch prevalence in both groups. [1]
- The contemporary report said none of the Rochester students who began training after their eighth birthday had absolute pitch. This sample finding does not establish a universal age limit. [1][2]
Source: [2] Scientific American
Among conservatory students who began musical training at ages 4–5, 60% of the Beijing students passed an absolute-pitch test, compared with 14% of the American students, according to Scientific American's contemporary report. In both groups, students who began training later were less likely to have absolute pitch. [2]
These findings concern the conservatory students studied—not musicians generally. They show associations with starting age and language background, but do not establish that speaking Mandarin caused the difference. [1][2]
What the researchers tested
Absolute pitch, also called perfect pitch, is the ability to name or produce a particular pitch without a reference note. [1][2]
Diana Deutsch and colleagues compared 88 Mandarin-speaking students at the Central Conservatory of Music in Beijing with 115 students at Eastman School of Music in Rochester. The Eastman comparison group consisted of non-tone-language speakers whose parents were also non-tone-language speakers. [1]
Students heard 36 synthesised piano notes and wrote down their names. Intervals between successive notes were larger than an octave, a design intended to minimise relative-pitch cues. Testing took place on site under highly similar conditions. [1]
Participants came from required courses, and every invited student agreed to take part. This avoided self-selection within the target groups, but did not make the samples representative of everyone taking music lessons. [1]
Starting age and prevalence
The published paper reports that earlier musical training was associated with greater absolute-pitch prevalence in both groups. It also reports higher prevalence among the Chinese students at each compared starting-age level. [1]
The shared starting-age groups were 4–5, 6–7 and 8–9. The Beijing groups contained 43, 22 and 12 students respectively; the corresponding Eastman groups contained 21, 31 and 24. The researchers considered only starting-age subgroups containing at least nine students. Some subgroup comparisons therefore rested on relatively small numbers. [1]
The supplied journal excerpt establishes the direction of the relationships but does not include the complete numerical prevalence results. The reported 60% and 14% comparison comes from Scientific American's November 2004 account of results due to be presented at an Acoustical Society of America meeting. That account also says none of the Rochester students beginning after their eighth birthday had absolute pitch. [1][2]
The journal paper appeared in 2006 and notes that portions of the work had been presented in 2004. These accounts should not be treated as independent replications. [1][2]
What language background might explain
Mandarin speakers had higher absolute-pitch prevalence across the compared starting-age groups, but they still showed the association between later training and lower prevalence. [1]
In tonal languages such as Mandarin, pitch height and contour help distinguish word meanings. Deutsch and colleagues proposed that early experience associating pitch with verbal meaning might support later associations between musical pitches and note names. This is a developmental hypothesis, not a demonstrated mechanism. [1]
The comparison involved different institutions in different countries. Grouping students by starting age does not establish that their other experiences were equivalent. Scientific American also noted that the Mandarin-speaking group was ethnically Chinese and that genetic differences could explain some of the effect. The comparison therefore cannot isolate language as the cause of the observed gap. [1][2]
Nor does a comparison involving Mandarin establish the same pattern for every tonal language. [1]
What musicians can take away
The supported conclusion is narrow: among these conservatory students, earlier musical training and Mandarin-speaking background were associated with higher absolute-pitch prevalence. The findings do not show that early lessons guarantee the ability or that later starters cannot acquire it. [1][2]
Finding no qualifying later starters in a sample is not the same as demonstrating a biological deadline. This study compared students with different training histories; it was not a test of whether adults could acquire the skill through a training programme. [1]
Absolute pitch should also not be treated as a measure of overall musicianship. The paper explicitly notes that it is not necessarily accompanied by superior performance on other musical tasks. [1]
Method and limitations
This article compares the supplied excerpt of the 2006 journal paper [1] with Scientific American's 2004 report [2]. Sample sizes, starting-age groups and test procedures come from [1]. The reported percentages of 60% and 14%, and the statement about Rochester students starting after their eighth birthday, come from [2]. These percentages are reproduced as reported, not presented as unrounded values.
No missing age-group prevalence results were reconstructed. No estimates, pooled rates or derived percentages were calculated, and no MyGear user data were used. The supplied excerpt of [3] was not used to draw conclusions about adult learning.
Sources
[1] Deutsch and colleagues, *Absolute pitch among American and Chinese conservatory students: Prevalence differences, and evidence for a speech-related critical period*. https://deutsch.ucsd.edu/pdf/JASA-2006_119_719-722.pdf
[2] Don Monroe, *Speaking Tonal Languages Promotes Perfect Pitch*, Scientific American. https://www.scientificamerican.com/article/speaking-tonal-languages/
[3] Wong and colleagues, *Learning fast and accurate absolute pitch judgment in adulthood*. https://pmc.ncbi.nlm.nih.gov/articles/PMC12325523/
Method
We compared the supplied excerpt of the 2006 journal paper [1] with Scientific American’s 2004 report of the research [2]. Sample sizes, starting-age groups and test procedures come from [1]. The exact prevalence figures of 60% and 14%, and the statement about Rochester students starting after their eighth birthday, come from [2]. The chart reproduces only the reported ages 4–5 comparison from [2]; it does not reconstruct missing results from the journal paper. No estimates, pooled rates or derived percentages were calculated. No MyGear user data were used.
Limitations
This is indicative evidence for the broader language-and-starting-age question, based on an observational comparison of conservatory populations. The published samples contained 88 Beijing and 115 Eastman students, but the shared starting-age subgroups were much smaller: 43, 22 and 12 in Beijing, and 21, 31 and 24 at Eastman. All invited students participated, reducing volunteer bias within the recruited groups, but conservatory admission and attendance strongly limit generalisability. Language background was confounded with institution, country and ancestry; the Eastman analysis also excluded 18 students who themselves, or whose parents, spoke an Asian language. The supplied journal excerpt ends during the methods and does not show the full results or classification threshold. Exact prevalence percentages are therefore attributed to the November 2004 news report, rather than presented as independently verified figures from the 2006 paper. These sources are not independent replications and do not establish present-day prevalence, causation or a universal developmental cut-off.
Based on smaller polls or MyGear's own anonymised data.
Sources
- [1] Absolute pitch among American and Chinese conservatory students: Prevalence differences, and evidence for a speech-related critical period (L) — Journal of the Acoustical Society of America, 2006
"The earlier the age of onset of musical training, the greater the prevalence of absolute pitch"
- [2] Speaking Tonal Languages Promotes Perfect Pitch — Scientific American, 2004-11-09
"In both groups, students who started their musical instruction later were less likely to have absolute pitch"
Keep exploring
Cite this research
MyGear Research (2026). "Conservatory study linked earlier training and Mandarin background with perfect pitch". MyGear, 9 October 2026. https://mygear-music.com/research/perfect-pitch-was-less-common-in-later-starters-and-more-common-in-mandarin-spea
