including establishing territories,courting,and communicating between
parent and chick.The brain control of song,at least in passerine birds,is
known to be lateralized to a degree greater than comparable in other
systems in mammals,excepting only the control of language in the
human.It is also the case,however,that peripheral auditory mechanisms
for the reception of sound and the neuromuscular control of sound pro-
duction are quite different in birds and mammals.Auditory sensitivity in
birds is much less acute than in mammals,with an upper cutoff frequency
of the order of 10kHz,and sound sensitivity handled by an acoustic
system much less elaborate than the chain of middle ear bones of
mammals;the bird cochlea is also less elaborate than that in any
mammal.One can therefore recognize systematic differences between
birds and mammals in their handling of auditory information,and neural
organization of their production and analysis of acoustic signals could be
fundamentally different.
But birds remain acceptable models for the study of musicality and its
evolution.Especially interesting is the plasticity of birdsong,in which the
detailed song is demonstrably a product of nature and nurture.In some
species of finches,for example,the adult song as used in courtship,and
other displays are learned as a result of exposure to songs of conspecifics
during critical periods of development.Furthermore,and even more
unusual,some nerve growth apparently takes place seasonally in brain
regions controlling song during the period when courtship and territor-
ial vocal displays occur.The conclusion follows that the detailed circuitry
controlling vocal behavior in birds is built up as a result of,and coordi-
nated with,environmental events.One does not know details of either
the construction or circuitry of the adult system,but this kind of
process is also assumed to take place in the development of human
neural control of speech and language.The process in humans differs
most significantly from that in birds in that the number of neurons con-
trolling human speech probably remains stable during development,and
the neuronal growth that occurs is in arborization,increases in the
number of dendrites and their synaptic connections.It should neverthe-
less be clear why birds are good models for such a process in view of the
evident plasticity at a neural as well as behavioral level in birds and
people.
We are still ignorant of how neural networks are put together in living
brains,so much so that “neural networks”today typically refer to net-
works that exist only as computer programs,often for studies of artifi-
cial intelligence.In the face of this ignorance it is helpful to have animal
models in which actual networks are constructed in real brains.It is likely
that construction of such networks follows similar rules in all metazoans,
and that the information could be applied cross-specifically much as192 Harry Jerison