to the so-called “what” and “where” pathways in visual processing.
Although neurons in belt regions can be driven using tonal stimuli, and
there are cochleotopic representations in several regions, many neurons
seem to prefer specific bandwidths for frequency-centered sound bursts,
independent of intensity (Rauschecker et al. 1995). The bandwidth tuning
of cells in the caudolateral (CL), mediolateral (ML), and anterior later
(AL) belt regions varies along an axis that is orthogonal to the cochleotopic
organization.
An additional auditory processing region just lateral to the belt, and
referred to as the “para-belt”, has been identified histologically (Hackett et
al. 1998), but neither its anatomical extent nor physiological response properties have been fully examined. In contrast to the dense reciprocal connections among the core and belt areas, the para-belt does not have direct
access to information in the core but rather appears to be driven primarily
by input from the belt (Kaas and Hackett 2000). The nature of this input is
not fully resolved. However, both the general pattern of connections among
the core, belt, and para-belt regions in primates and the change in tuning
properties across these regions resemble that among field L, NCM, and
cmHV in songbirds described above. Lateral-belt neurons are tuned for
both direction and rate of frequency modulation (Rauschecker 1997), and,
at least in macaques, many of these cells also respond vigorously to conspecific calls or components thereof (Rauschecker et al. 1995). Although
responses to species-specific calls can also be observed in AI, the selectivity is significantly less pronounced than that in the lateral-belt areas (see
Rauschecker 1998). Moreover, early recordings in squirrel monkeys, Saimiri
sciureus, showing large percentages of cells responsive to conspecific calls
and subpopulations therein with modest nonlinear response properties
(Wollberg and Newman 1972; Newman and Wollberg 1973; Winter and
Funkenstein 1973) were mostly done in the superior temporal gyrus and
therefore were likely to lie within the lateral belt or para-belt. Thus, there
appears to be a general pattern of increasingly complex receptive field
properties that coincides with the transition from primary auditory cortex
to postsynaptic regions. This can be observed in humans as well, where the
dorsolateral superior temporal gyrus and planum temporale (putative belt)
are more strongly activated by FM tones than noise, and the superior temporal sulcus (putative para-belt) is more activated by various speech
stimuli, including pseudowords and reversed speech, than by FM tones and
other nonvocal sounds (Binder et al. 1994, 2000; Belin et al. 2000).
4. Summary and Conclusions
We have discussed organizational schemes and mechanisms that underlie
the coding of complex stimuli, including feature detectors and combination
sensitivity, which are likely to provide the basis for representations of
7. Neuroethology of Vocal Communication
367
Although neurons in belt regions can be driven using tonal stimuli, and
there are cochleotopic representations in several regions, many neurons
seem to prefer specific bandwidths for frequency-centered sound bursts,
independent of intensity (Rauschecker et al. 1995). The bandwidth tuning
of cells in the caudolateral (CL), mediolateral (ML), and anterior later
(AL) belt regions varies along an axis that is orthogonal to the cochleotopic
organization.
An additional auditory processing region just lateral to the belt, and
referred to as the “para-belt”, has been identified histologically (Hackett et
al. 1998), but neither its anatomical extent nor physiological response properties have been fully examined. In contrast to the dense reciprocal connections among the core and belt areas, the para-belt does not have direct
access to information in the core but rather appears to be driven primarily
by input from the belt (Kaas and Hackett 2000). The nature of this input is
not fully resolved. However, both the general pattern of connections among
the core, belt, and para-belt regions in primates and the change in tuning
properties across these regions resemble that among field L, NCM, and
cmHV in songbirds described above. Lateral-belt neurons are tuned for
both direction and rate of frequency modulation (Rauschecker 1997), and,
at least in macaques, many of these cells also respond vigorously to conspecific calls or components thereof (Rauschecker et al. 1995). Although
responses to species-specific calls can also be observed in AI, the selectivity is significantly less pronounced than that in the lateral-belt areas (see
Rauschecker 1998). Moreover, early recordings in squirrel monkeys, Saimiri
sciureus, showing large percentages of cells responsive to conspecific calls
and subpopulations therein with modest nonlinear response properties
(Wollberg and Newman 1972; Newman and Wollberg 1973; Winter and
Funkenstein 1973) were mostly done in the superior temporal gyrus and
therefore were likely to lie within the lateral belt or para-belt. Thus, there
appears to be a general pattern of increasingly complex receptive field
properties that coincides with the transition from primary auditory cortex
to postsynaptic regions. This can be observed in humans as well, where the
dorsolateral superior temporal gyrus and planum temporale (putative belt)
are more strongly activated by FM tones than noise, and the superior temporal sulcus (putative para-belt) is more activated by various speech
stimuli, including pseudowords and reversed speech, than by FM tones and
other nonvocal sounds (Binder et al. 1994, 2000; Belin et al. 2000).
4. Summary and Conclusions
We have discussed organizational schemes and mechanisms that underlie
the coding of complex stimuli, including feature detectors and combination
sensitivity, which are likely to provide the basis for representations of
7. Neuroethology of Vocal Communication
367