Gestalt Factors in the Structure of Birdsong and Human Melody

The influence of Gestalt principles in human auditory perception is well established (Bregman, 1990). Experimental work has implicated the operation of Gestalt principles in auditory streaming and grouping within streams (e.g., Bregman, 1990; Bregman & Dannenbring, 1973; Dorman, Cutting, & Raphael, 1975; Heise & Miller, 1951; Noteboom, Brokx, & de Rooij, 1978). Gestalt-inspired principles have also been applied in a recent music-theoretic approach to the perception of melody (Narmour, 1990). Narmour's model proposes that melodic sequences engage a system of expectancies. The model further suggests that principles governing expectancies operate at several levels; specific Gestalt-like principles operate at a note-to-note level independently of cultural exposure or music training. Empirical support for these principles has been obtained for both perception (Cuddy & Lunney, 1995; Krumhansl, 1995) and production tasks (Thompson, Plaus, & Cuddy, under review). This presentation reports an investigation of the melodic patterns produced in songbirds. In particular we examine the stereotypical song of four songbird species for evidence of conformity to the Gestalt-like principles proposed to operate at a note-to-note level in humans. Analyses consider differences between production patterns formed by songbirds with rich harmonic song versus songbirds with simple harmonic song. Also addressed are the similarities and differences between birdsong and human melody.