Abstract:The pitch levels of female speech in two villages situated in a relatively remote area of China were compared. The dialects spoken in the two villages are similar to Standard Mandarin, and all subjects had learned to read and speak Standard Mandarin at school. Subjects read out a passage of roughly 3.25 min in Standard Mandarin, and pitch values were obtained at 5-ms intervals. The overall pitch levels in the two villages differed significantly, supporting the conjecture that pitch levels of speech are influenced by a mental representation acquired through long-term exposure to the speech of others.

Absolute pitch, the ability to name a musical note in the absence of a reference note, is very rare in North America and Europe, so that attempts to characterize its features in the western world have involved small numbers of subjects, informal self-report, questionnaires, or web-based exploration. The study reported here capitalized on the high prevalence of absolute pitch in China to explore its features in detail using direct, on-site testing of 160 subjects in a Chinese music conservatory. As expected, performance levels were extremely high, and there was a large effect of age of onset of musical training, with those who began training by age 5 scoring on average 83% correct not allowing for semitone errors and 90% correct allowing for semitone errors. It was found that errors tended to be on the sharp side. An advantage to white keys over black keys was also found; however this was not due to early experience with the piano, as had been hypothesized by others, since performers on different instruments showed an effect that was as large or larger. Furthermore, the special status for note A that had been hypothesized by others was not found, even for orchestral performers.

Absolute pitch has been hypothesized to be involved in processing lexical tones in tone languages, which associate pitch information with verbal labels. Since possessors of absolute pitch in music utilize the overall pitch of a tone in making identification judgments, the hypothesis was tested that native speakers of Mandarin utilize the overall pitch of a lexical tone as a cue to retrieve its tone label. In a reaction time task, Mandarin syllables in all four tones were presented both in their original forms and also transposed to four different levels of pitch height; subjects listened to each token and judged whether or not its meaning corresponded to the original. It was found that although Mandarin syllables with transposed overall pitch heights were judged to be the same tones as the original tokens, subjects were significantly slower in making judgments for those syllables that were transposed to different levels of pitch height, compared to those that were presented at their original pitch heights. This effect was most extreme for tones 1 and 3. These findings suggest that overall pitch height serves as a cue for identifying lexical tones and further strengthens the link between absolute pitch and lexical tone perception.

No abstract