The Tritone Paradox

Chapter 5 explores the tritone paradox—a musical illusion that was discovered by the author. Its basic pattern consists of two computer-generated tones that are related by a half-octave (i.e., a tritone). These tones are well defined in pitch class (note name) but ambiguous in pitch height. When one of these tone pairs is played in succession, some people hear an ascending pattern, yet other people hear a descending one. Indeed, a group of people will disagree completely among themselves as to whether such a pair of tones is moving up or down in pitch. Furthermore, any one person hears one of these tone pairs as ascending or descending depending on their note names (such as C–F♯, or G♯–D). How people hear the tritone paradox varies with the geographic location in which they grew up—and so with their native language or dialect. Native English-speaking Californians hear this pattern differently from natives of the south of England. People who are natives of Vietnam hear the pattern quite differently from native English-speaking Californians. The tritone paradox shows, therefore, that the way we perceive music is related to our language, and generally reveals strong effects of our memories and expectations on how we hear music. It also has important implications for absolute pitch (or “perfect pitch”)—the rare ability to name a musical note that is presented in isolation. People make orderly judgments of the tritone paradox, even though they cannot name the notes that they are judging, so they must have an implicit form of absolute pitch.

Correspondence in Perception of the Tritone Paradox and Perfect-Fifth/Perfect-Fourth Intervals

Shepard (1964) found that the pitch height of a pair of octave-related complex tones is perceived in accordance with the principle of proximity around a pitch class circle. However, when these tones form a tritone interval, proximity cannot be used. In the tritone paradox, Deutsch (1986) found that listeners perceive these tones such that half of the pitch class circle is heard as higher in pitch, and the opposite half as lower, with individual differences in which half is heard as higher. In the present experiments, listeners judged the height of octave-related complexes forming tritones and forming intervals of perfect fifths (P5) and perfect fourths (P4). There was a strong relationship between the pitch classes heard higher in the tritone paradox and those heard higher when presented with P5/P4 intervals. Rather than using proximity to judge pitch height with P5/P4 intervals, listeners instead use the same mechanism involved in perception of the tritone paradox.

Speech Patterns Heard Early in Life Influence Later Perception of the Tritone Paradox

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.