An animal’s perceptual system must support discrimination among conspecific acoustic signals, and its vocal system must support flexible signal
production. A well-developed perceptual system that operates independently from a flexible vocal-production system is insufficient. A flexible vocal
system that is not guided by socially relevant acoustic input is also inadequate. There must be an audio–vocal interface that allows for adjustments
in signal production with changing acoustic input. Without this neural
interface, there is no way for acoustic input to influence subsequent vocal
production.
Feedback between the study of call function and vocal learning can be
productive. Knowing call function can help to focus research on those vocalizations that are likely to be learned. For example, when acoustic similarity
among group members facilitates recognition and individuals are not
closely related, vocal learning is likely. Similarly, studying the structural
features influenced by the social environment can reveal additional subtleties of call function and where vocal flexibility is adaptive. In addition,
the extent of modification through vocal learning—whether learned acquisition or social modification is involved—gives us insight into the direct
forces that shape vocalizations and points to the perceptual, motor, and
neural mechanisms necessary for such vocal flexibility.
Learned vocalizations are by no means the only interesting ones. Study
of acoustic structure and usage of genetically determined vocalizations
can give great insight into many aspects of a species’ social behavior and
cognition. Vocal learning is but one way to achieve communicative complexity. Contrasting learned with hardwired vocalizations can tell us much
about the conditions favoring cultural inheritance and those favoring
genetic inheritance.
1.3. Testing the Vocal Learning Hypothesis
Observational data can provide evidence in support of vocal learning, but
only carefully controlled experiments can unambiguously demonstrate it,
and experiments are necessary to identify mechanisms and rule out alternative explanations. Important baseline observational data should include
descriptions of normal ontogenetic changes in vocalizations and the social
contexts in which they occur (e.g., Moss et al. 1997). When ontogenetic
change is absent, learned acquisition is unlikely. In addition, interpreting
results from many experimental manipulations requires data on normal
vocal development. Acoustic data accompanied by data on morphological
development are especially valuable, because they facilitate exploration of
how morphological changes affect acoustic structure (e.g., Jones et al. 1991).
Comparisons of acoustic structure among individuals, territorial neighbors,
social groups, and geographic regions help to characterize call function and
reveal when call similarity is favored. When individuals copy neighbors or
other social partners, both microgeographic and macrogeographic patterns
4. Comparative Vocal Learning
141
production. A well-developed perceptual system that operates independently from a flexible vocal-production system is insufficient. A flexible vocal
system that is not guided by socially relevant acoustic input is also inadequate. There must be an audio–vocal interface that allows for adjustments
in signal production with changing acoustic input. Without this neural
interface, there is no way for acoustic input to influence subsequent vocal
production.
Feedback between the study of call function and vocal learning can be
productive. Knowing call function can help to focus research on those vocalizations that are likely to be learned. For example, when acoustic similarity
among group members facilitates recognition and individuals are not
closely related, vocal learning is likely. Similarly, studying the structural
features influenced by the social environment can reveal additional subtleties of call function and where vocal flexibility is adaptive. In addition,
the extent of modification through vocal learning—whether learned acquisition or social modification is involved—gives us insight into the direct
forces that shape vocalizations and points to the perceptual, motor, and
neural mechanisms necessary for such vocal flexibility.
Learned vocalizations are by no means the only interesting ones. Study
of acoustic structure and usage of genetically determined vocalizations
can give great insight into many aspects of a species’ social behavior and
cognition. Vocal learning is but one way to achieve communicative complexity. Contrasting learned with hardwired vocalizations can tell us much
about the conditions favoring cultural inheritance and those favoring
genetic inheritance.
1.3. Testing the Vocal Learning Hypothesis
Observational data can provide evidence in support of vocal learning, but
only carefully controlled experiments can unambiguously demonstrate it,
and experiments are necessary to identify mechanisms and rule out alternative explanations. Important baseline observational data should include
descriptions of normal ontogenetic changes in vocalizations and the social
contexts in which they occur (e.g., Moss et al. 1997). When ontogenetic
change is absent, learned acquisition is unlikely. In addition, interpreting
results from many experimental manipulations requires data on normal
vocal development. Acoustic data accompanied by data on morphological
development are especially valuable, because they facilitate exploration of
how morphological changes affect acoustic structure (e.g., Jones et al. 1991).
Comparisons of acoustic structure among individuals, territorial neighbors,
social groups, and geographic regions help to characterize call function and
reveal when call similarity is favored. When individuals copy neighbors or
other social partners, both microgeographic and macrogeographic patterns
4. Comparative Vocal Learning
141