a general response bias of auditory systems (Hartshorne 1956; Searcy 1992).
This hypothesis suggests that song complexity per se is selected, rather than
any message encoded in different song syllables, because it reduces habituation in neighboring males and courting females. Searcy presents some evidence supporting this hypothesis in studies of grackles (Fig. 5.8).These birds
do not have complex repertoires, but females are attracted preferentially
to artificially enhanced repertoires compared to the more monotonous song
of their conspecifics. When females were presented with a song of 32 syllables, they showed more courtship solicitation displays to the song that contained eight successive repetitions of four different syllables to a song of 32
identical syllables. Interestingly, the female response decreased within the
repetition of the same syllable but increased during transition between syllable types, suggesting a pattern of habituation to signals in response to the
simple song but habituation and release from habituation in response to a
more complicated song (Fig. 5.8, top).
Studies of zebra finches and canaries point toward some of the underlying mechanisms of this preference for more complex songs. Both electrophysiological responses (Stripling et al. 1997) of auditory neurons and
expression of an early gene, zenk, implicated in auditory function (Mello
et al. 1992), also show decreased response to repeated song stimuli and
enhanced response during transition between stimulus types (Fig. 5.8;
reviewed in Ryan 1998).
To the extent that habituation to signal monotony was a general phenomenon, we would expect response biases to generate selection for more
complicated signals. The manner in which signal complexity could be
enhanced would be constrained by the types of sounds that are efficacious
in that particular system. For example, although swamp and song sparrows
learn each others’ song in the absence of their own, there is a strong bias
to learn their own song type (reviewed in Marler 1997). Such genetic pre5. Selection on Signals
259
Figure 5.8. (A) Female courtship displays to complex song in grackles show higher
levels in response to multiple-song repertoires than to single-song types. The single
type (circles) contains 32 repeats of the same song. The repertoire (squares) contains four different song types repeated in groups of eight (e.g., 1–8, 9–6, etc.). The
response to repertoire songs shows that there is habituation to repeated songs
within each song type (e.g., 1–8,9–16, etc.) and release from habituation at transition between quartets (song 9, 17, etc.; Searcy 1992). (B) Electrophysiological
responses of units in the zebra finch’s causomedial neostriatum, which borders the
song-control nucleus, shows decreased spike rates to repeated presentation of the
same song and enhanced spike rates in response to a new song (Stripling et al. 1997).
(C) Expression of an immediate early gene, zenk, is higher during transitions from
no song to song (0/S1), or from one song to another song (S1/S2; S2/S1), than during
absence of song (0/0) and repeated stimulation of the same song (S1/S1; S2/S2; from
Mello et al. 1992). (Reprinted with permission from Ryan 1998. Copyright © 1998
American Association for the Advancement of Science.)
᭣
This hypothesis suggests that song complexity per se is selected, rather than
any message encoded in different song syllables, because it reduces habituation in neighboring males and courting females. Searcy presents some evidence supporting this hypothesis in studies of grackles (Fig. 5.8).These birds
do not have complex repertoires, but females are attracted preferentially
to artificially enhanced repertoires compared to the more monotonous song
of their conspecifics. When females were presented with a song of 32 syllables, they showed more courtship solicitation displays to the song that contained eight successive repetitions of four different syllables to a song of 32
identical syllables. Interestingly, the female response decreased within the
repetition of the same syllable but increased during transition between syllable types, suggesting a pattern of habituation to signals in response to the
simple song but habituation and release from habituation in response to a
more complicated song (Fig. 5.8, top).
Studies of zebra finches and canaries point toward some of the underlying mechanisms of this preference for more complex songs. Both electrophysiological responses (Stripling et al. 1997) of auditory neurons and
expression of an early gene, zenk, implicated in auditory function (Mello
et al. 1992), also show decreased response to repeated song stimuli and
enhanced response during transition between stimulus types (Fig. 5.8;
reviewed in Ryan 1998).
To the extent that habituation to signal monotony was a general phenomenon, we would expect response biases to generate selection for more
complicated signals. The manner in which signal complexity could be
enhanced would be constrained by the types of sounds that are efficacious
in that particular system. For example, although swamp and song sparrows
learn each others’ song in the absence of their own, there is a strong bias
to learn their own song type (reviewed in Marler 1997). Such genetic pre5. Selection on Signals
259
Figure 5.8. (A) Female courtship displays to complex song in grackles show higher
levels in response to multiple-song repertoires than to single-song types. The single
type (circles) contains 32 repeats of the same song. The repertoire (squares) contains four different song types repeated in groups of eight (e.g., 1–8, 9–6, etc.). The
response to repertoire songs shows that there is habituation to repeated songs
within each song type (e.g., 1–8,9–16, etc.) and release from habituation at transition between quartets (song 9, 17, etc.; Searcy 1992). (B) Electrophysiological
responses of units in the zebra finch’s causomedial neostriatum, which borders the
song-control nucleus, shows decreased spike rates to repeated presentation of the
same song and enhanced spike rates in response to a new song (Stripling et al. 1997).
(C) Expression of an immediate early gene, zenk, is higher during transitions from
no song to song (0/S1), or from one song to another song (S1/S2; S2/S1), than during
absence of song (0/0) and repeated stimulation of the same song (S1/S1; S2/S2; from
Mello et al. 1992). (Reprinted with permission from Ryan 1998. Copyright © 1998
American Association for the Advancement of Science.)
᭣