however, preclude social modification (see below). Learned acquisition is
not restricted to the juvenile phase in all species; adults can acquire new
vocalizations, and this acquisition is often guided by social partners (e.g.,
Slater 1989). This category of vocal learning includes age-dependent learning (Payne and Payne 1997) and selective attrition (Nelson and Marler
1994; Nelson et al. 1996).
Song learning in oscine passerines is an excellent and well-studied
example of learned acquisition. We elect not to describe this process here
because there are a number of excellent recent reviews (e.g., Catchpole and
Slater 1995; Kroodsma and Miller 1996). We refer the reader to these.
Instead, our review focuses on call learning in oscines and on less wellstudied groups.
Social modification occurs when existing vocalizations are altered in
response to social interactions. Frequently, although not necessarily, this
results in increased acoustic similarity between the social partners’ and
target individuals’ vocalizations. Occasionally, vocalizations can become
more distinct, which may facilitate identification of individuals (Janik
and Slater 1997; Janik 1999). In many species, social modification is not
restricted to a particular developmental stage. Vocalizations that are
acquired developmentally can be subject to later social modification. But
even vocalizations that do not require auditory experience for normal production can be socially modified. This category of vocal learning includes
social-dependent learning (Payne and Payne 1997).
Changing frequency characteristics of vocalizations through learning may
require more complex perceptual and motor mechanisms than changing
temporal characteristics. We discuss both frequency and temporal changes
because we consider both to be interesting for understanding vocal development. Other authors have limited their definition of vocal learning to
frequency changes or learned production (e.g., Janik and Slater 1997, 2000).
We elect not to restrict ourselves in this way because we do not think this
is the most critical distinction, although we do differentiate frequency from
temporal changes in our discussion of evidence for vocal learning, and data
on frequency changes are more abundant.
Here, we focus on behavioral and acoustic data. Data on the biophysics
of vocal production are reviewed by Fitch and Hauser (Chapter 3), data on
ecological constraints are discussed by Bass and Clark (Chapter 2), and data
on neural mechanisms underlying call detection are reviewed by Gentner
and Margoliash (Chapter 7).
1.2. What Does Vocal Learning Teach us about
Acoustic Communication?
Vocal learning involves modification of acoustic signals produced in a social
context and requires the coordinated operation of both perceptual and
motor systems. For vocal learning to take place, two things must happen.
140
J.W. Boughman and C.F. Moss
not restricted to the juvenile phase in all species; adults can acquire new
vocalizations, and this acquisition is often guided by social partners (e.g.,
Slater 1989). This category of vocal learning includes age-dependent learning (Payne and Payne 1997) and selective attrition (Nelson and Marler
1994; Nelson et al. 1996).
Song learning in oscine passerines is an excellent and well-studied
example of learned acquisition. We elect not to describe this process here
because there are a number of excellent recent reviews (e.g., Catchpole and
Slater 1995; Kroodsma and Miller 1996). We refer the reader to these.
Instead, our review focuses on call learning in oscines and on less wellstudied groups.
Social modification occurs when existing vocalizations are altered in
response to social interactions. Frequently, although not necessarily, this
results in increased acoustic similarity between the social partners’ and
target individuals’ vocalizations. Occasionally, vocalizations can become
more distinct, which may facilitate identification of individuals (Janik
and Slater 1997; Janik 1999). In many species, social modification is not
restricted to a particular developmental stage. Vocalizations that are
acquired developmentally can be subject to later social modification. But
even vocalizations that do not require auditory experience for normal production can be socially modified. This category of vocal learning includes
social-dependent learning (Payne and Payne 1997).
Changing frequency characteristics of vocalizations through learning may
require more complex perceptual and motor mechanisms than changing
temporal characteristics. We discuss both frequency and temporal changes
because we consider both to be interesting for understanding vocal development. Other authors have limited their definition of vocal learning to
frequency changes or learned production (e.g., Janik and Slater 1997, 2000).
We elect not to restrict ourselves in this way because we do not think this
is the most critical distinction, although we do differentiate frequency from
temporal changes in our discussion of evidence for vocal learning, and data
on frequency changes are more abundant.
Here, we focus on behavioral and acoustic data. Data on the biophysics
of vocal production are reviewed by Fitch and Hauser (Chapter 3), data on
ecological constraints are discussed by Bass and Clark (Chapter 2), and data
on neural mechanisms underlying call detection are reviewed by Gentner
and Margoliash (Chapter 7).
1.2. What Does Vocal Learning Teach us about
Acoustic Communication?
Vocal learning involves modification of acoustic signals produced in a social
context and requires the coordinated operation of both perceptual and
motor systems. For vocal learning to take place, two things must happen.
140
J.W. Boughman and C.F. Moss