Learning to pause: Fidelity of and biases in the developmental acquisition of gaps in the communicative signals of a songbird

Abstract

The temporal organization of sounds used in social contexts can provide information about signal function and evoke varying responses in listeners (receivers). For example, music is a universal and learned human behavior that is characterized by different rhythms and tempos that can evoke disparate responses in listeners. Similarly, birdsong is a social behavior in songbirds that is learned during critical periods in development and used to evoke physiological and behavioral responses in receivers. Recent investigations have begun to reveal the breadth of universal patterns in birdsong and their similarities to common patterns in speech and music, but relatively little is known about the degree to which biological predispositions and developmental experiences interact to shape the temporal patterning of birdsong. Here, we investigated how biological predispositions modulate the acquisition and production of an important temporal feature of birdsong, namely the duration of silent pauses (“gaps”) between vocal elements (“syllables”). Through analyses of semi‐naturally raised and experimentally tutored zebra finches, we observed that juvenile zebra finches imitate the durations of the silent gaps in their tutor's song. Further, when juveniles were experimentally tutored with stimuli containing a wide range of gap durations, we observed biases in the prevalence and stereotypy of gap durations. Together, these studies demonstrate how biological predispositions and developmental experiences differently affect distinct temporal features of birdsong and highlight similarities in developmental plasticity across birdsong, speech, and music.Research Highlights The temporal organization of learned acoustic patterns can be similar across human cultures and across species, suggesting biological predispositions in acquisition. We studied how biological predispositions and developmental experiences affect an important temporal feature of birdsong, namely the duration of silent intervals between vocal elements (“gaps”). Semi‐naturally and experimentally tutored zebra finches imitated the durations of gaps in their tutor's song and displayed some biases in the learning and production of gap durations and in gap variability. These findings in the zebra finch provide parallels with the acquisition of temporal features of speech and music in humans.

Temporal variability enhances the acquisition of stereotyped communication signals

“…Thus, juveniles were not exposed to song from a live adult during the sensitive period for song learning and were naïve to song before experimental tutoring. When these juveniles were nutritionally independent (i.e., could feed themselves; ∼30–40 days old), they were housed individually in a soundbox for a period of song tutoring (see below) and thereafter until they were mature (∼4 months of age; Derégnaucourt et al, 2005; James and Sakata, 2017; Lipkind et al, 2013; Lipkind et al, 2017; Mets and Brainard, 2018; Mets and Brainard, 2019; Tchernichovski et al, 2001; James et al, 2023).…”

“…The second group of birds was tutored with a species-atypical stimulus where the gap durations randomly varied from 10-50 ms (in 10-ms increments; with an average of 30-ms gaps)("variable gaps" or Vgap; n=32), but with stereotyped sequences of syllables. This degree of variability in gap durations across renditions is >5 times greater than the variability observed in typical zebra finch song (James et al, 2023). The third and final group of zebra finches was tutored with a species-atypical stimulus that consisted of variable (randomized) sequences of syllables ("variable sequencing" or Vseq) but with stereotyped gap durations (30-ms gaps; n=54).…”

“…After documenting general learning in response to our stimuli, we next analyzed how variability in syllable timing or sequencing affected song learning. The duration of silent gaps between syllables is highly consistent from rendition-to-rendition for adult zebra finch song (e.g., Glaze and Troyer, 2006;Norton and Scharff, 2016;James et al, 2023) and is encoded by neurons in auditory and sensorimotor circuits of songbirds (Margoliash and Fortune, 1992;Lampen et al, 2014;Araki et al, 2016;Bouchard and Brainard, 2016). Zebra finches also produce stereotyped sequences of syllables, and behavioral responses and neural responses within circuits for song learning are sensitive to the sequential order of syllables (e.g., Cynx, 1993;Lewicki and Konishi, 1995;Lewicki and Arthur, 1996;Braaten et al, 2006;Brainard, 2013, 2016;Chen and ten Cate, 2015;Cazala et al, 2019).…”

“…For example, mice and rats engage in stereotyped sequences of grooming behavior that vary across strains and species (Spruijt et al, 1992; Geuther et al, 2021), and different species of songbirds produce vocal sequences with different statistical (sequence) dependencies (e.g., Jin and Kozhevnikov, 2011; Warren et al, 2012; Lipkind et al, 2013; reviewed in Kershenbaum et al, 2014; Cohen et al, 2020; James et al, 2020; Searcy et al, 2022). In addition to the sequences themselves, the timing of elements within behavioral sequences can demonstrate varying degrees of species specificity or typicality (e.g., Fox et al, 2020; Liberman et al, 1958; Araki et al, 2016; Feng et al, 1990; Gerhardt, 2005; Hedwig, 2006; James et al, 2023).…”

Music, a piece of many puzzles in developmental science

“…Another compares children's associations between space and pitch across two non-English languages, Dutch and Turkish (Dolscheid et al, 2023). An examination of the biological basis of octave equivalence perception in common marmosets (Wagner et al, 2023) provides a comparative perspective, as does work examining the interplay between biological predispositions and developmental experiences in shaping the temporal patterning of birdsong (James et al, 2023). Contributions like these are essential for a comprehensive developmental science.…”