Absolute pitch is generally considered to reflect a rare musical endowment; however, its characteristics are puzzling and its genesis is unclear. We describe two experiments in which native speakers of tone languages——Mandarin and Vietnamese——were found to display a remarkably precise and stable form of absolute pitch in enunciating words. We further describe a third experiment in which speakers of English displayed less stability on an analogous task. Based on these findings, and considering the related literatures on critical periods in speech development, and the neurological underpinnings of lexical tone, we propose a framework for the genesis of absolute pitch. The framework assumes that absolute pitch originally evolved as a feature of speech, analogous to other features such as vowel quality, and that speakers of tone language naturally acquire this feature during the critical period for speech acquisition. We further propose that the acquisition of absolute pitch by rare individuals who speak an intonation language may be associated with a critical period of unusually long duration, so that it encompasses the age at which the child can take music lessons. We conclude that the potential to acquire absolute pitch is universally present at birth, and that it can be realized by enabling the infant to associate pitches with verbal labels during the critical period for speech acquisition.

The tritone paradox occurs when two tones that are related by a half-octave (or tritone) are presented in succession and the tones are constructed in such a way that their pitch classes (C, C♯♯, D, etc.) are clearly defined but their octave placement is ambiguous. Previous studies have shown that there are large individual differences in how such tone pairs are perceived, and these differences correlate with the listener's language or dialect. We here present findings showing that perception of the tritone paradox can be heavily influenced by speech heard early in life, even for listeners who do not speak their first language fluently. Our findings point to a specific linkage between speech and music, and they also shed light on the issue of critical periods for the acquisition of intonational properties of speech.

It is shown on a number of grounds that the findings described by Repp (1994) are uninterpretable in relation to those of Deutsch (1991) and Deutsch, North, and Ray (1990). First, the geographical correlates with perception of the tritone paradox obtained by Repp were based on data from subject populations that differed substantially from those studied by Deutsch (1991), and Repp's subject populations were also treated differently from each other. Second, Repp obtained substantial differences in perception of the tritone paradox depending on the spectral envelope under which the tones were generated; however, he attempted to correlate the data obtained from only one of these envelopes with pitch ranges for speech. Third, the procedures used by Repp to determine the pitch ranges for speech were problematic and inappropriate in the context of Deutsch's hypothesis.

A previous study (Deutsch, 1991) demonstrated a striking difference in perception of the tritone paradox between subjects who had grown up in two different geographical regions. Specifically, a group of subjects who had grown up in California were compared with a group who had grown up in the South of England. When the Californian group tended to hear the pattern as ascending, the English group tended to hear it as descending, and vice versa. This raises the question of whether regional differences also exist within the United States in the way this pattern is perceived. The present study examined the percepts of subjects who had grown up in Mahoning and/or Trumbull counties in Ohio. Two groups were compared: those whose parents had also grown up in this area and those for whom this was not the case. A highly significant difference between these two groups of subjects was obtained, with those in the latter group producing a distribution of percepts similar to that found among Californians and those in the former group producing a different distribution. From this and other analyses of the data, it is concluded that regional differences in perception of the tritone paradox do indeed exist within the United States and that there is in addition an effect of familial background.

In a study by Deutsch (1991), a large and highly significant difference in perception of the tritone paradox was found between a group of subjects who had grown up in California and a group who had grown up in the south of England: In general, where the Californian group tended to hear the pattern as ascending the English group tended to hear it as descending, and vice versa. The present paper documents some further geographical correlates that are derived from the data obtained by Deutsch (1991). The strength of the relationship of pitch class to perceived height was found to depend on the overall heights of the spectral envelopes under which the tones were generated. However, the direction of this dependence differed significantly depending on the subject population. For subjects showing a "Californian pattern" (i. e., whose overall peak pitch classes were in the range moving clockwise from A#–B to D#–E), this relationship was more pronounced for tones generated under lower spectral envelopes, and so when the tones were perceived as lower in overall height. In contrast, for subjects showing an "English pattern" (i. e., whose overall peak pitch classes were in the opposite region of the pitch-class circle), this relationship was more pronounced for tones generated under higher spectral envelopes, and so when the tones were perceived as higher overall instead. Given the literature on the pitch of speech as a function of linguistic community, these findings provide further evidence that perception of the tritone paradox is related to the processing of speech sounds.

In this paper, a new effect of pitch proximity is reported. Subjects were presented with patterns consisting of octave-related complexes. Each pattern was composed of four tones, which constituted two simultaneous melodic lines, one of which ascended by a semitone while the other descended by a semitone. The subjects judged whether the line that was higher in pitch ascended or descended, and from these judgments it was inferred which line was heard as higher and which as lower. It was found that the patterns were perceptually organized so that the simultaneous tones formed proximal relationships. The findings provide evidence that, just as successions of tones tend to be organized perceptually in accordance with pitch proximity, so simultaneous tones tend to be organized in accordance with this principle also.

The tritone paradox is produced when two tones that are related by a half- octave (or tritone) are presented in succession. Each tone is composed of a set of octave- related harmonics, whose amplitudes are determined by a bell-shaped spectral envelope; thus the tones are clearly defined in terms of pitch class, but poorly defined in terms of height. When listeners judge whether such tone pairs form ascending or descending patterns, their judgments generally show systematic relationships to the positions of the tones along the pitch-class circle: Tones in one region of the circle are heard as higher and those in the opposite region are heard as lower. However, listeners disagree substantially as to whether a given tone pair forms an ascending or a descending pattern, and therefore as to which tones are heard as higher and which as lower. This paper demonstrates that the basis for the individual differences in perception of this musical pattern lies in the language spoken by the listener. Two groups of subjects made judgments of the tritone paradox. One group had grown up in California, and the other group had grown up in southern England. It was found that when the Californian group tended to hear the pattern as ascending the English group tended to hear it as descending, and when the Californian group tended to hear the pattern as descending the English group tended to hear it as ascending. This finding, coupled with the earlier results of Deutsch, North, and Ray (1990) that showed a correlate between perception of the tritone paradox and the pitch range of the listener's spontaneous speaking voice, indicates strongly that the same, culturally acquired representation of pitch classes influences both speech production and perception of this musical pattern.

In the tritone paradox, two tones are presented that are related by a halfoctave. Each tone consists of a set of octave-related sinusoids whose amplitudes are scaled by a bell-shaped spectral envelope; thus the usual cues to height attribution are missing. When listeners judge whether such tone pairs form ascending or descending patterns, judgments are related in an orderly fashion to the positions of the tones along the pitch class circle: Tones in one region of the circle are heard as higher and those in the opposite region as lower. However, listeners differ strikingly in the orientation of the pitch-class circle with respect to height. So far, the basis of the tritone paradox and the reasons for the individual differences in its manifestation have proved elusive. In the present study, a correlation is found between perception of the tritone paradox and the range of fundamental frequencies of the listener's speaking voice. To the authors' knowledge, this is the first demonstration of a close connection between the perception of a musical pattern on the one hand and the listener's speech characteristics on the other.

This article concerns a pattern of tones that provides a curious exception to the principle of equivalence under transposition. When the pattern is transposed by a half octave, the identical pitches are heard but they appear to be reversed in time. Further, the pattern in any one key is heard quite differently by different listeners. The implications of this new pitch paradox are discussed.

The tritone paradox occurs when an ordered pair of tones is presented, with each tone consisting of a set of octave-related components, and the pitch classes of the tones separated by a half-octave. Such a pattern is heard as ascending in one key, but as descending in a different key. Further, the pattern in any one key is heard as ascending by some listeners but as descending by others. It was here found that this phenomenon occurs to a highly significant extent in a general population, and that it is distributed within the population in an orderly fashion. The findings also reveal a surprising ability within the general population to utilize absolute pitch.

A pattern of tones is explored that possesses some remarkable properties. It is heard as ascending when played in one key, yet as descending when played in another key. When the pattern is recorded on tape and then played back at different speeds, it is heard either as ascending or as descending depending on the speed of playback. As a further paradox, the pattern in any given key is heard as ascending by some listeners but as descending by others. Theoretical implications of these findings are discussed.

Ear advantages in dichotic listening are generally held to reflect greater involvement of the hemisphere contralateral to the preferred ear. Rationales for this view rely on the assumption that, when information arriving at the two ears is treated as a single complex signal, an advantage to a component of this signal based on ear of input will not interact with an advantage based on some other attribute, such as frequency. This assumption is shown, for the case of dichotically presented melodies, to be unjustified. Dichotic tone pairs that are presented in a sequential context are more accurately perceived and localized when the higher tone is to the right and the lower to the left, than when the higher tone is to the left and the lower to the right. This anisotropy can, in principle, give rise to patterns of ear advantage which may tend in either direction, depending on factors such as type of material presented, level of difficulty, task requirement, and category of listener. Since its basis is unknown, ear advantages resulting from this anisotropy cannot be assumed to reflect greater involvement of either hemisphere in processing the information.

The pitch of a tone is considered to consist of two components: the rectilinear component of height and the circular component of chroma, or pitch class. In an experiment employing simultaneous sequences of Shepard tones, it is shown that tones in different positions on the chroma circle differ in height when generated under the same spectral amplitude envelope. Further, the directions of these differences remain constant when the envelope is centered at different positions along the spectrum, resulting in clear differences in the overall heights of the patterns.

This article examines conditions under which tones standing in octave relation are treated as equivalent by the perceptual system. According to the two-channel model for the abstraction of pitch relationships proposed by Deutsch (1969, 1982), octave equivalence effects should not operate directly in the processing of melodic intervals. In an experiment to test this prediction, it was found that accuracy in immediate recall of pitch sequences was substantially lower when the tones within a sequence were distributed across two adjacent octaves than when they were all in the same octave. This finding is in accordance with the two-channel model. Its implications for musical processing, together with those of other studies on octave equivalence, are discussed.