Learning: Mechanisms, Ecology, and Evolution • 172.5. What do animals learn?
Early studies of animal behavior emphasized the importance of instinct in
most nonhuman animals and insects in particular (Fabre et al. 1918; Tinbergen
1951). Such views still dominate in many disciplines of ecology and evolu-
tion. Two related events that have been instrumental in changing our current
understanding of learning are the highly successful establishment of fruit
flies as a model system for research on the neurogenetics of learning and the
realization that the genetic and cellular mechanisms controlling learning and
memory are remarkably similar across diverse taxa. Intriguingly, although
evolutionary biologists have studied fruit flies for about a century (Kohler
1994), it was neurogeneticists who critically documented robust learning in
Drosophila melanogaster (Quinn et al. 1974; Davis 2005). The neurogenetic work
on learning in fruit flies has required the development of behavioral protocols
for quantifying learning. Consequently, a variety of ingenious procedures have
been developed in the past few decades, which indicate that fruit flies rely on
learning for all major life activities.
I will focus on fruit flies in this section because they possess a few char-
acteristics that raise the question as to whether they should learn at all. First,
fruit flies D. melanogaster are tiny, short-lived animals. The adult female is
only about 2.5 millimeters long, and the male is slightly smaller. Larval de-
velopment takes about 4 days (Ashburner et al. 1976), and adults in the wild
probably have a life expectancy of only several days. Second, the larval and
adult brains contain approximately 20,000 and 200,000 neurons respectively.
These are remarkably small numbers compared with vertebrates. Finally, fruit
fly life history appears straightforward and does not readily indicate reliance
on learning. In fact, we still do not know to what extent fruit flies depend on
learning in natural settings because all research relevant to learning has thus
far been conducted in the laboratory.
Female fruit flies lay their eggs on decaying fruit containing yeast. Hence,
upon emergence, the larvae just have to commence feeding, which is their
sole task. Even though the larval task seems easy and they possess limited sen-
sory and information-processing abilities, larvae are able to learn. In the first
documentation of learning in fruit fly larvae, Aceves-Pina and Quinn (1979)
allowed groups of third-instar larvae to experience three 30-second pulses of
one odor together with the application of an electric shock. The larvae also
experienced three 30-second pulses of another odor not associated with shock.
The two treatments were alternated and separated by 90-second breaks. In a
subsequent choice test, the larvae exhibited significant avoidance of the odor
associated with shock. Control treatments employing the application of only
odorants or only shocks indicated no effects of these nonassociative treatments