STRUCTURES of EXPOSURE describe one aspect of a process called "auditory scene analysis" (which Albert Bregman has described extensively in a book of the same name). More on that later, but a crucial way that we infer things about our environment, using sound, is to separate, or segregate, complex auditory stimuli into "streams." In music, the segregation of streams occurs in both obvious and non-obvious ways. Obvious examples include the very notion of counterpoint itself: composers aim to produce separate and independent streams of information, occurring simultaneously, and they are aided in that task by an elusive array of techniques, including the careful deployment of dissonance and consonance, "well-formed" melodic shapes, and a variety of oblique, contrasting, and similar motion.

But we also have the classic example of a solo instrumental work in which no two notes are played simultaneously, but which somehow suggests a number of contrapuntal voices. In such cases, all the usual rules of counterpoint apply, but our perceiving brain must separate streams from one another without the usual aid of knowing that simultaneously sounding notes must come from different sound-producing sources: in the solo instrumental work, where notes occur only one at a time, the segregation of streams requires us to infer separation from cues like articulation, pitch difference, and the abstract relationships between successions of notes. Each of these "cues" is a potential basis for a structure of exposure, and the presence of such cues is potentially responsible for the perception of unique event-streams, sometimes in non-obvious ways.

I emphasize the term "structures of exposure" here because of how much impact each of these cues can have on rhythm perception, and because the impacts of these cues are fundamentally different. In musical examples, and in reports on empirical studies (my own and others'), I will show that many 'rhythms' involving a number of notes -- say, anywhere from 7 to 12 -- there are, in fact, a number of emergent streams, each of which can be reduced to a smaller number of notes. In view of the kinds of emphasis afforded by articulation, loudness, pitch, timbre, duration, and other patterns, a number of rhythmic phenomena can be understood as consequences of a basic rhythmic structures that are made of event groups that come in threes. Thus "structures of exposure" are indirectly responsible for our choice to emphasize the "properties" of timespan pairs.


  • Any arrangement of sonic events in which cues (e.g. pitch, loudness, duration, and timbre) interact to segregate a 'stream' of events as a distinct rhythm.

  • The cues that can afford such distinctions produce this rhythmic distinction in different ways, and have different impacts on the perception of meter and pulse (Carson 2007).