vocal behavior, such as calling behavior in relation to complex social interactions (Margoliash et al. 1994), or other reward/reinforcement contingencies involving vocal perception in adults. Such speculations are consistent
with emerging homologies between the AFP and cortico-basal ganglia
loops in mammals (Luo and Perkel 1999).
3.3.5. Songbird Auditory System: Physiology
In European starlings, Ov is tonotopically organized, with best frequencies
decreasing ventrally (Bigalke-Kunz et al. 1987). Neurons throughout the
auditory telencephalon also show tonotopic organization (Leppelsack and
Schwartzkopff 1972; Rubsamen and Dorrscheidt 1986), although in much
more complex patterns. In starlings, roughly 11 different regions can be
identified on the basis of the direction of the tonotopic gradient and tuning
curve bandwidth (Haüsler 1997; Capsius and Lepplesack 1999), and similar
patterns are observed in zebra finches (Gehr et al. 1999). These tonotopically defined regions appear to respect anatomically defined regions of the
field L complex.
Relatively few studies have examined responses in the telencephalic
regions using complex acoustic stimuli. Neurons in L1 and L3 have lower
response rates to tone bursts than those in L2 and show greater selectivity
to species-specific vocalizations (Leppelsack and Vogt 1976; Bonke et al.
1979a; Müller and Leppelsack 1985; Theunissen and Doupe 1998). This
selectivity is borne out by the complexity of the spatial-temporal receptive
fields (STRFs) for many neurons within field L. Indeed, more reliable estimates of the STRF are derived from responses to conspecific vocalizations
than tone pips (Theunissen et al. 2000; cf. Schäfer et al. 1992). This general
pattern of increasing response selectivity from field L2 to the higher-order
areas continues into NCM and cHV (Müller and Leppelsack 1985), suggesting that these regions are involved in the extraction of complex features. Early data from white-crowned sparrows are consistent with this in
showing a small subset of neurons in the NCM that are selective for
specific directions of FM in a common trill element of conspecific song
(Leppelsack 1983). Recent preliminary data (Grace and Theunissen 2000;
Gentner and Margoliash 2001) support the idea that cHV in particular is
involved in the extraction and/or representation of complex features in
showing highly selective responses in this region to specific features in
behaviorally relevant conspecific songs.
Neurons in NCM are broadly responsive to conspecific stimuli and
respond to the repeated presentation of conspecific song in a stimulusspecific manner (Chew et al. 1995; Stripling et al. 1997). The repeated presentation of a single conspecific song elicits a rapid modulation in the initial
firing rate of NCM neurons (Stripling et al. 1997). If the same song is
repeated on the order of 200 times, this initial modulation of the firing rate
is no longer observed when that same song is presented on subsequent trials.
7. Neuroethology of Vocal Communication
363
with emerging homologies between the AFP and cortico-basal ganglia
loops in mammals (Luo and Perkel 1999).
3.3.5. Songbird Auditory System: Physiology
In European starlings, Ov is tonotopically organized, with best frequencies
decreasing ventrally (Bigalke-Kunz et al. 1987). Neurons throughout the
auditory telencephalon also show tonotopic organization (Leppelsack and
Schwartzkopff 1972; Rubsamen and Dorrscheidt 1986), although in much
more complex patterns. In starlings, roughly 11 different regions can be
identified on the basis of the direction of the tonotopic gradient and tuning
curve bandwidth (Haüsler 1997; Capsius and Lepplesack 1999), and similar
patterns are observed in zebra finches (Gehr et al. 1999). These tonotopically defined regions appear to respect anatomically defined regions of the
field L complex.
Relatively few studies have examined responses in the telencephalic
regions using complex acoustic stimuli. Neurons in L1 and L3 have lower
response rates to tone bursts than those in L2 and show greater selectivity
to species-specific vocalizations (Leppelsack and Vogt 1976; Bonke et al.
1979a; Müller and Leppelsack 1985; Theunissen and Doupe 1998). This
selectivity is borne out by the complexity of the spatial-temporal receptive
fields (STRFs) for many neurons within field L. Indeed, more reliable estimates of the STRF are derived from responses to conspecific vocalizations
than tone pips (Theunissen et al. 2000; cf. Schäfer et al. 1992). This general
pattern of increasing response selectivity from field L2 to the higher-order
areas continues into NCM and cHV (Müller and Leppelsack 1985), suggesting that these regions are involved in the extraction of complex features. Early data from white-crowned sparrows are consistent with this in
showing a small subset of neurons in the NCM that are selective for
specific directions of FM in a common trill element of conspecific song
(Leppelsack 1983). Recent preliminary data (Grace and Theunissen 2000;
Gentner and Margoliash 2001) support the idea that cHV in particular is
involved in the extraction and/or representation of complex features in
showing highly selective responses in this region to specific features in
behaviorally relevant conspecific songs.
Neurons in NCM are broadly responsive to conspecific stimuli and
respond to the repeated presentation of conspecific song in a stimulusspecific manner (Chew et al. 1995; Stripling et al. 1997). The repeated presentation of a single conspecific song elicits a rapid modulation in the initial
firing rate of NCM neurons (Stripling et al. 1997). If the same song is
repeated on the order of 200 times, this initial modulation of the firing rate
is no longer observed when that same song is presented on subsequent trials.
7. Neuroethology of Vocal Communication
363