One of the founding fathers of behavioural ecology, Niko Tinbergen
(1963), proposed that there are four ways of answering the question ‘why?’
in such cases. The first way is in terms of function: the parasite alters host
behaviour because it increases the parasite’s chances of completing its life
cycle. This answer emphasizes the adaptive nature of the phenomenon.
The second way is in terms of causation: the parasite alters host behaviour
because it interferes with the biochemistry or physiology of its host in
ways that have repercussions on host behaviour. This answer shifts the
focus toward the mechanisms of host manipulation. The third way is in
terms of evolutionary history: the parasite alters host behaviour because
it inherited this ability from its ancestors and now shares it with related
parasite species. The fourth way to answer the ‘why’ question would be in
terms of development and would involve the learning processes necessary
for a parasite to achieve host manipulation. Since parasites have no
opportunity to learn about manipulation (they do it as part of their
genetic programming or they die), this fourth type of answer is not
truly applicable to the host-manipulation phenomenon.
After an early enthusiasm for the functional aspects of behaviour, the
recent trends in behavioural ecology suggest a return to an integrated
approach linking function with mechanisms and evolution (Krebs and
Davies, 1997). When studying the behavioural ecology of parasites, it is
only fitting to adopt the perspective of a behavioural ecologist. Thus, here
I shall summarize recent advances in the study of host manipulation by
parasites at the functional, evolutionary and mechanistic levels, and try
to link these levels. I shall not present a comprehensive review of all
examples documented in the literature; these are available elsewhere (e.g.
Moore and Gotelli, 1990; Moore, 1993; Poulin, 1995, 1998a). Instead,
I shall try to provide the conceptual and experimental framework necess-
ary for the rigorous study of parasite manipulation of host behaviour.Manipulation as a Functional Adaptation
Changes in host behaviour following parasitic infection are not necess-
arily beneficial for the parasite. They can be responses of the host aimed at
eliminating the parasite or compensating for its effects. For example,
behavioural fever allows parasitized insects to raise their body temp-
erature and kill their parasites (Boorstein and Ewald, 1987). Alternatively,
they can be pathological side-effects with few or no consequences for
parasite fitness. Here I shall focus on behavioural changes in infected
hosts thought to be cases of adaptive manipulation by the parasite. In
these situations, alterations in host behaviour apparently lead to an
increase in the probability of successful parasite transmission and
completion of the life cycle. Such adaptive manipulation can take many
forms. For instance, parasitoid wasps can change the behaviour of their
insect host to ensure their own survival following their emergence from
the host (Brodeur and Vet, 1994; Eberhard, 2000). Parasitic fungi can make244 R. Poulin