Chapter 2: Understanding the Proximate and Ultimate Causes of Bird
Song
2.1 Work on the development of singing behavior in male white-crowned sparrows
has demonstrated that the birds must learn to sing a particular dialect of the full
song of their species. This finding eliminates the genetic differences hypothesis for
the song differences between white-crowns in Marin and in Berkeley. Would we be
right, therefore, to conclude that the genetic information present in the cells of
white-crowned sparrows is irrelevant for the development of the bird’s singing
behavior? In this regard, what importance do you attach to the finding that whitecrown males apparently can learn their species’ song far more easily than the song
of other sparrows? What about the finding that white-crown males that hear whitecrown song only during a 40-day period early in life can nevertheless generate a
complete song, although they do not start singing until 3 to 4 months after exposure
to a tutor tape?
To say that the behavioral differences between two individuals are not genetic does not
mean that the behavior itself can develop without the genetic information possessed by
the two individuals. Both Marin and Berkeley birds have genes, which could be identical
in both individuals, and that are vital to the development of their brains. The genes in
question influence the structure and capacities of their brains, an essential contribution if
their brains are to grow and acquire the ability to learn things. The fact is that all male
white-crowns have learning biases that steer them toward the acquisition of a local dialect
of their own species. Thus, the gene–environment interactions that take place during
development produce a very special kind of brain with limited, but adaptive, properties.
2.2 A natural experiment sometimes occurs in Australian woodlands when galahs (a
species of parrot) lay eggs in nest holes in trees that are then stolen from them by
pink cockatoos (another parrot). Thus, the cockatoos become foster parents for
baby galahs. The young foster-reared galahs produce begging calls and alarm calls
that are identical to those produced by galahs cared for by their genetic parents.
However, the adopted galahs eventually give contact calls very much like those of
their adoptive cockatoo parents, as you can see from Figure 2.6 of the textbook.
(The birds produce these signals to help maintain contact with others when
traveling in flocks [1037].) Someone claims that these observations show that galah
begging and alarm calls are genetically determined, whereas contact calls are
environmentally determined. Explain why this claim is wrong (you may find it
helpful to read ahead in Chapter 3). Then defend the superficially similar statement
that the differences between the alarm calls given by adopted galahs and by their
cockatoo foster parents are the result of genetic differences between them. What
other behavioral differences are the result of differences between the social
environment of the adopted galahs and that of certain other individuals?
All behavioral traits require both genetic and environmental inputs in order to develop.
Therefore, none of the calls of a galah can be purely genetic or purely environmental.
Instead, the brain and the vocal apparatus of the parrot could only develop under the
influences of both genetic information and the material and experiential environment.
The differences, however, between the alarm calls of two individuals could be the
product of differences either in the birds’ DNA or their environments. The difference
between adopted galahs and the cockatoos in their alarm calls appears to be genetic given
that the environment of foster reared galahs is essentially the same as that for cockatoo
nestlings. On the other hand, the difference in the contact calls given by galahs that have
been adopted by cockatoos and those that are reared by their genetic parents can be
attributed largely to differences in the acoustical and social environments of the two
categories of galahs. Because galahs that have been adopted by cockatoos do not have the
same experiences as galahs reared by their parents, they undergo a slightly different set of
gene  environment interactions, which lead to the development of a different
vocalization.
2.3 Naturally, female starlings possess the same ZENK gene males do. In the female
brain, the ventral caudomedial neostriatum, or NCMv, responds to signals sent to it
from auditory neurons that fire when the bird is exposed to sounds, such as those
made by singing male starlings. When captive female starlings are given a choice
between perching next to a nest box where they can hear a long starling song versus
perching next to another nest box where a shorter starling song is played, they
spend more time at the long song site (Figure 2.10 of the textbook). What proximate
hypothesis could account for the song preferences of female starlings? What
prediction can you make about the activity of the ZENK gene in the NCMv of female
starlings exposed to long versus short songs? How might you check your prediction?
What would be the scientific point of collecting the data necessary to evaluate your
prediction?
The brain cells that process acoustical stimulation have the genetic information needed to
produce one pattern of activity in response to long songs and another in response to short
songs and to provide stronger rewarding sensations when the birds hear long songs. This
difference in neural activity should be related to differences in activity in the ZENK gene.
This prediction could be checked by sacrificing samples of female starlings shortly after
they had listened to tapes of long or short songs; the NCMv could then be assayed for the
ZENK gene product. The point of collecting these data would be to test the hypothesis
that cells in the NCMv possessed a gene or genes that contributed to the physiological
mechanism underlying song preferences by female starlings.
2.4 In the greater spear-nosed bat, unrelated females roost together in groups. The
members of any given group tend to produce similar calls, but different groups have
different calls. If you transfer some young bats into a new group and, at the same
time, move some of their siblings into still another group, the shifted bats eventually
change their calls to resemble those in their new groups [128]. What proximate
hypothesis and what ultimate hypothesis on the acquisition of acoustical
communication are supported by these results?
The proximate hypothesis would be that the differences between the calling patterns of
different groups was due to differences in the social influences and experiences operating
in the different groups. The ultimate hypothesis is that social integration of group
members is facilitated by the ability of individuals to mimic the signals given by others in
their group, perhaps especially the calls produced by dominant group members.
2.5 At the proximate level, two hypothesis have been advanced to account for the
ability of some yearling white-crowned sparrows to match the dialect of neighboring
males. The late acquisition hypothesis states that yearlings have a developmental
window similar to that of very young birds, which enables them to listen to and
learn directly from their immediate neighbors, overriding (if need be) any dialect
learning that took place earlier in life. In contrast, the selective attrition hypothesis
argues that fledglings can memorize a number of dialect versions of their species
song early in life; then, after settling next to some older males, these birds gradually
discard certain of their learned variants until they are left with the one that best
matches the dialect of their neighbors. In light of these two alternative hypotheses,
what significance do you attach to Figure 2.20 of the textbook? Outline the entire
scientific process here, from the question that motivated the work to the scientific
conclusion.
The causal question: what causes young male white-crowns to sing like their neighbors?
The two hypotheses generate different predictions with the late-acquisition hypothesis
predicting that young males should be able to learn a dialect novel to them long after the
early period of song acquisition documented by Peter Marler. The selective attrition
hypothesis, in contrast, states that the birds have actually learned several songs early in
life but that later, when confronted by a rival with a particular song type, the young birds
will come to drop the nonmatching songs and sing the matching type more often. Figure
2.20 provides direct evidence of this latter pattern, which in turns gives support to the
selective attrition hypothesis, which is the scientific conclusion of this work.
2.6 William Searcy and a team of researchers played taped songs to captive female
song sparrows that had been given hormone implants shortly after being taken to
the laboratory [1078]. The recorded songs came from male song sparrows that lived
in the population from which the females had been taken as well as from males
living various distances (18, 34, 68, 135, and 540 kilometers) from the female
subjects. Songs from males living 34 or more kilometers from the populations from
which the females came were not nearly as effective in eliciting the precopulatory
display as songs from local males; in contrast, songs from males living only 18
kilometers away were about as sexually stimulating as local songs. These data have
relevance for more than one ultimate hypothesis on song learning by male sparrows.
What are the hypotheses, and what importance do these findings have for them?
One hypothesis is that females prefer songs of males in their natal neighborhood because
these males will supply locally adapted genes to their offspring if chosen as mates.
Alternatively, females may prefer the songs of males that most closely match those that
the females heard early in life because adult males capable of singing good matches to
the songs that the males heard early in their lives must have had “good” brains as
youngsters, which made it possible for them to mimic the songs they were hearing
accurately. Males of this sort may either be genetically superior or be in superior
condition with the capability of providing superior parental care. Both hypotheses
generate the prediction that females will be more stimulated by males living in their natal
area or close to it (since these nearby males will probably sing the same or a similar
dialect). The findings support both hypotheses.
2.7 What features of language learning in humans are similar to song learning in
birds? Do these similarities suggest certain hypotheses on the proximate bases of
human language learning, especially the genetic and developmental components?
Do comparisons with birds also suggest some interesting hypotheses on the adaptive
value of learned language for members of our species? After you have attempted to
answer these questions, go to the ISI Web of Science, if it is available at your college
or university library, and try to find references on the shared proximate and
ultimate causes of vocal learning by humans and birds. You may find it useful to
know that those who have written knowledgeably on this subject include Peter
Marler and Fernando Nottebohm.
Among the shared features are the apparent window of language learning that makes it
much easier for an infant or young person to learn a language than an older child or adult.
Another is the stunning ease with which these complex acoustical signals can be learned
by very young infants. Another is the acquisition of local dialects by both birds and
humans, a function of the capacity of members of both species to match the
communication sounds they hear precisely. And infants, like young white-crowns, go
through a pre-language babbling phase in which they practice the sounds they can make
and match them to their memories of the sounds they have heard others make, a process
that requires the child to be hearing itself (not deaf). To do these things, humans, just like
white-crowned sparrows, must possess genetic information that helps make the
development of a learning-capable brain possible. The human brain is one, like the whitecrown’s, that is very strongly influenced not just by the acoustical environment, but also
by the social environment, with human infants highly attuned to the sounds produced by
their primary caretaker, usually their mother. One ultimate hypothesis that might apply to
both birds and humans is that the skill with which a person uses his group’s speech could
be important in mate choice, with females evaluating male language ability
unconsciously as an indicator of the developmental history of potential mates. Language
acquisition takes place during a very demanding period of growth, and persons
handicapped by inadequate diets may exhibit life-long deficits in speech and body
condition, which should make them less likely to elevate the fitness of a partner.