Abstract

1. Repertoire size and organization of songs appears to vary greatly across even closely related species of birds. Much of the apparent complexity of song organization can be described by the relatively simple principle of the Markov chain. 2. We examined long sequences of songs in five closely related North American thrushes using two methods of Markov analysis, one based on the maximum likelihood method of HOEL, the other based on information theory. Both methods employed computer simulation of zero, first, second, and higher order Markov chains for comparison with empirical data. 3. Results indicated that each of the five species employed Markov sequencing on more than one level of song organization. Thus six Swainson's thrushes show strict linear first order relationships both in song sequencing and syllables within songs. Two veerys show invariate order within songs, but more variable second order sequencing between songs. A hermit thrush maintained fixed order within songs, but showed first and second and perhaps even higher order song sequencing, and a larger song repertoire. In four wood thrushes, each individual used identical parts in the same section (B or C) of different song types, the usual order of sections being ABC. Sequencing of songs within this species appears to be based on a combination of second or higher order sequencing of individual parts in sections B and C, leading to great richness in repertoire. In three robins, we found syllables rather than songs the primary unit of organization, showing first and higher order sequencing. The resulting songs are extremely variable. 4. A continuum of increasingly complex Markov chaining in both song and syllables within song resembles one based on the genetic affinities of the birds. 5. The process of drift in forming larger repertoires, employing Markov principles as well, is discussed. 6. The usefulness of the Markov model as a common pervasive organizational principle in bird vocalization is stressed.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call