Chorusing, synchrony, and the evolutionary functions of rhythm

Abstract

A central goal of biomusicology is to understand the biological basis of human musicality. One approach to this problem has been to compare core components of human musicality (relative pitch perception, entrainment, etc.) with similar capacities in other animal species. Here we extend and clarify this comparative approach with respect to rhythm. First, whereas most comparisons between human music and animal acoustic behavior have focused on spectral properties (melody and harmony), we argue for the central importance of temporal properties, and propose that this domain is ripe for further comparative research. Second, whereas most rhythm research in non-human animals has examined animal timing in isolation, we consider how chorusing dynamics can shape individual timing, as in human music and dance, arguing that group behavior is key to understanding the adaptive functions of rhythm. To illustrate the interdependence between individual and chorusing dynamics, we present a computational model of chorusing agents relating individual call timing with synchronous group behavior. Third, we distinguish and clarify mechanistic and functional explanations of rhythmic phenomena, often conflated in the literature, arguing that this distinction is key for understanding the evolution of musicality. Fourth, we expand biomusicological discussions beyond the species typically considered, providing an overview of chorusing and rhythmic behavior across a broad range of taxa (orthopterans, fireflies, frogs, birds, and primates). Finally, we propose an "Evolving Signal Timing" hypothesis, suggesting that similarities between timing abilities in biological species will be based on comparable chorusing behaviors. We conclude that the comparative study of chorusing species can provide important insights into the adaptive function(s) of rhythmic behavior in our "proto-musical" primate ancestors, and thus inform our understanding of the biology and evolution of rhythm in human music and language.

Keywords: chorusing; coupled oscillators; evolution of communication; isochrony; music perception; rhythm; synchronization; timing.

FIGURE 1

Two-dimensional space representing levels of inquiry in evolution and behavior. Asking why and how a given species exhibits proto-musical behaviors entails a number of more specific questions, whose answers might have a clear functional or mechanistic perspective or, sometimes, be an inextricable combination of both perspectives. While much of the previous literature has focused on the mechanisms underlying rhythm perception and production, the evolutionary functions of rhythm have received less attention.

FIGURE 2

Visual depiction of the proposed definitional framework. (A) The solo tree hierarchically categorizes temporal patterns produced by a single individual. Categorization is accomplished by starting at the top and following the black lines down according to which branch provides a better fit at each level. Visual examples are shown in green at each level as a guide. Each example depicts a progression of events (spikes) in time (x-axis; left to right) that satisfies the conditions for inclusion in a particular category. For space reasons, the tree is only filled out for periodic patterns; for further description of aperiodic patterns see the main text. (B) The chorus tree hierarchically categorizes temporal patterns produced by multiple individuals. The format is the same as in (A), with the exceptions that two patterns (red, displayed at the top, and blue, displayed at the bottom of each pattern pair) are necessary to show examples of category membership, gray boxes are used to highlight groups of categories (labeled according to the names following the roman numerals I–III), and light gray boxes are used to highlight subgroups of categories (labeled according to names in parentheses). In the light gray box labeled “period ratios”, the ratios given beneath each visual example relate the periods of the corresponding example patterns in milliseconds. For space reasons, the tree is only filled out for coupled simultaneous patterns; for further description of uncoupled choruses and coupled alternating choruses see the main text.

FIGURE 3

Calling onsets of 3 individuals over two time periods, following a “selfish chorusing” phase-shift rule. In the first period (left) individual B (blue) is surrounded by much silence, hence in the second period (right) B will postpone its call and signal between R (red) and G (green). Likewise, G will anticipate its call and signal between B and R. Individual R’s optimal strategy is to keep its call onset unchanged. Figure reproduced and modified from (Ravignani, 2014).

FIGURE 4

Agent based simulation of 3 individuals calling over six time periods (based on the model in Ravignani, 2014). Phases’ onsets of calls are depicted on top as time series. Time periods are separated by vertical lines, with the equivalent polar representation below each period. During the first time period, individuals listen to each other, so to shift their calls’ onset in the second period, repeating this strategy over time. The blue and green individuals will alternate as leader and follower over time periods; the red individual’s best strategy is, in this example, to keep its call onset constant over time periods. The agents will have reached almost perfect synchrony by period 6.