Saturday, December 04, 2004

Origins of Absolute Pitch

Absolute pitch, the ability to name or produce a note of an isolated pitch, is very rare in our culture. Less than 1 in 10,000 people of the general population have this rare ability. Characteristics of absolute pitch are a mystery, and its origin is unknown. Today, I will discuss the features of absolute pitch, experiments to explain absolute pitch, and discuss the framework proposed for the origin of absolute pitch based on the experiments.

Evidence shows that people who are unable to name isolated pitches still have a partial form of absolute pitch. An experiment conducted by Terhardt and Ward in 1982, found that musicians who did not have absolute pitch were able to judge whether a song they knew was played in the correct key. Likewise, in 1989, Halpern asked subjects with no musical background to hum the first few notes of familiar songs on different occasions. Their pitches were surprisingly consistent each time. Another experiment in 1994 was conducted by Levitin, in which subjects choose two songs, which had only been performed by one musical group, and were asked to hum, whistle, or sing the melody. Levitin found that 44% of the subjects came within two semitones of the accurate pitch for both songs.

Although absolute pitch is most common among highly accomplished musicians, they usually lack the ability to perform other musical tasks. For example, judgments of musical intervals and registers are not always accurate. Studies have shown that pitch is related to the languages or dialects individuals have been exposed to. The pitch range of a person’s speaking voice is related to that person’s dialect rather than to physiological characteristics, such as height and weight. In 1995, Schlaug, Jancke, Huang, and Steinmetz were the first to document that musicians with absolute pitch tend to exhibit a different brain structure, most commonly leftward asymmetry, than those without absolute pitch.

Lennenberg pointed out that adults and young children acquire absolute pitch in different ways, like learning a second language. If a second language is acquired after puberty, it is frequently spoken with a “foreign accent” and contains grammatical errors. Lennenberg included that there is a critical period, which extends to puberty, and is a crucial time for acquiring speech and language. Studies of second language acquisition have proven Lennenberg’s theory to be true. Individuals who were first exposed to a second language in early childhood were found to be more proficient in that language than children ages 4 to 6 and adults. Relating this to music, absolute pitch can be acquired in adulthood; however this occurs only through extensive training. When young children acquire absolute pitch, they do it unconsciously without specific training. Also, absolute pitch that is acquired as an adult does not have the same ease or preciseness than if it was acquired earlier in life.

There is also evidence that absolute pitch is related to the age of musical training. In 1998, a survey of musicians and music students was conducted by Baharloo, Johnston, Service, Gitschier, and Freimer. Figure 1 shows the results of the survey.

Figure 1

Age training began

4 and under




12 and over

Percentage of those with absolute pitch






Clearly, there is a decline of those with absolute pitch as the beginning training age increases.

The link between absolute pitch and speech becomes even clearer when evidence from tone languages, such as Mandarin, Thai, and Vietnamese are considered. In these languages, words’ meanings are different based on the tones in which they are pronounced. For example, in Mandarin, the word “ma” means “mother” when spoken in first tone, “hemp” in the second tone, and “horse” in the third tone. Likewise, when a person with absolute pitch hears the note G, and identifies it as “G”, the person is also associating a certain pitch with a verbal description.

Which features of pitch are critical to conveying meaning in tone language? One hypothesis is that absolute pitch is treated by tone language speakers as a critical feature of speech. Three experiments were conducted to test this hypothesis.

Seven native speakers of Vietnamese, who had been living in the United States for periods ranging from a few months to 17 years and had received minimal or no musical training, were the subjects of the first experiment. Each subject was tested individually in two sessions. Each session they would read a list of 10 Vietnamese words in a microphone at a rate of one word every two seconds. The words covered the range of all of the tones in Vietnamese speech. All subjects produced pitch difference scores of less than 1.1 semitones, and two of the seven subjects produced pitch difference scores of less than 0.25 semitones. Therefore, the subjects must have been referring to stable absolute pitch patterns when enunciating the list of words.

The purpose of the second experiment was to test the results from the Vietnamese subjects to speakers of a different tone language. It was also conducted to see the differences in pitch if the same words were in enunciated on different days. Fifteen native speakers of Mandarin were asked to speak a list of 12 Mandarin words, which consisted of three words in each of the four Mandarin tones, put together so that the same tone did not occur two times in a row. In contrast to the first experiment, each subject was asked to read out the word list twice in each session with readings separated by 20 seconds. Consistencies were once again achieved. For all comparisons, 1/3 of the subjects produced difference scores of less than 0.25 semitones. The results backed up the hypothesis that the subjects were referring to stable absolute pitch templates when enunciating the words. Although the pitch differences found in comparing the readings from different days were very small, they underestimated the accuracy of the subjects’ absolute pitch pattern.

The first two experiments examined the performance of tone language speakers only. However, the third experiment was designed to see if speakers of an intonation language, such as English, would display the same pitch consistency. The experiment was the same as the first two, except native speakers of English were the subjects, and a list of 12 English words were used. The Mandarin and English speakers showed the same degree of pitch consistency in enunciating their word lists twice in succession, but the English speakers showed less pitch consistency across days.

The results of the three experiments support the hypothesis that absolute pitch is treated by tone language speakers as a critical feature of speech. English speakers were very consistent in the pitches which they articulated on different days. This may be related to the previous studies which showed that those who do not possess absolute pitch can still have a partial form of it. From the studies, we can now expect that speakers of tone language would acquire absolute pitch for music best in early childhood, and this ability would decline with increasing age.

For the rare cases of absolute pitch among people who haven’t been exposed to tone language, Deutsch hypothesized that the critical period for acquiring absolute pitch is unusually long. This unusually long critical period could be genetically determined and could also be associated with an unusual form of brain organization.

The results from the three experiments suggest that parts of the brain underlying absolute pitch originally evolved to help speech, and that it is now involved in processing absolute pitch for both speech and music.

Deutsch, Diana, Mark Dolson, and Trevor Henthorn. "Absolute Pitch, Speech, and Tone Language: Some Experiments and a Proposed Framework." Music Perception Vol. 21, No. 3 (2004): 339-356.