It is possible that guppies do not pay attention to inspectors because they are
observing these interactions out of context, i.e. they are not directly involved in
the inspection bout. In the original study in which guppies did form preferences
for better inspectors (Dugatkin & Alfieri 1991), subjects interacted with the other
inspectors during the inspection bout. Being involved in the inspection bout may
allow the guppies to reliably assess how cooperative other individuals would be
with respect to themselves, making it reasonable to show a preference. In our
study, an observer could only learn how cooperative another inspector was with
respect to some third individual. This type of information may not be relevant to
the observer because it does not provide a gauge of how either inspector would
cooperate with him. As a consequence, simply watching inspection bouts, without
being involved in the interaction, may not be sufficient to elicit preference
formation.
The lack of preference following observation alone may be particularly
relevant in light of the proposed game theory strategies that the guppies may be
following. Several fish species, including guppies, follow a tit-for-tat strategy
when inspecting predators (Milinski 1987; Dugatkin 1988; Dugatkin 1991;
Dugatkin & Alfieri 1991; but see Stephens et al. 1997). Presumably the
cooperative solution to tit-for-tat type interactions develops only in situations
where individuals interact repeatedly, recognize each other, and remember
previous outcomes (Axelrod & Hamilton 1981). Guppies meet these requirements for predator inspections in which they are a participant (Dugatkin 1988;
Dugatkin 1991), and apparently they learn to recognize and prefer to associate
with good inspectors (Dugatkin & Alfieri 1991). In our study, however,
observers could not interact with the inspectors and did not form preferences for
individuals who were better inspectors. This implies that observation alone does
not provide accurate information about another individual’s cooperative
tendencies relative to oneself. These data indirectly support the notion that
cooperation via tit-for-tat is most likely to evolve when individuals are allowed
to interact directly with one another.
It is possible that guppies do not pay attention to inspectors because they are
observing these interactions out of context, i.e. they are not directly involved in
the inspection bout. In the original study in which guppies did form preferences
for better inspectors (Dugatkin & Alfieri 1991), subjects interacted with the other
inspectors during the inspection bout. Being involved in the inspection bout may
allow the guppies to reliably assess how cooperative other individuals would be
with respect to themselves, making it reasonable to show a preference. In our
study, an observer could only learn how cooperative another inspector was with
respect to some third individual. This type of information may not be relevant to
the observer because it does not provide a gauge of how either inspector would
cooperate with him. As a consequence, simply watching inspection bouts, without
being involved in the interaction, may not be sufficient to elicit preference
formation.
The lack of preference following observation alone may be particularly
relevant in light of the proposed game theory strategies that the guppies may be
following. Several fish species, including guppies, follow a tit-for-tat strategy
when inspecting predators (Milinski 1987; Dugatkin 1988; Dugatkin 1991;
Dugatkin & Alfieri 1991; but see Stephens et al. 1997). Presumably the
cooperative solution to tit-for-tat type interactions develops only in situations
where individuals interact repeatedly, recognize each other, and remember
previous outcomes (Axelrod & Hamilton 1981). Guppies meet these requirements for predator inspections in which they are a participant (Dugatkin 1988;
Dugatkin 1991), and apparently they learn to recognize and prefer to associate
with good inspectors (Dugatkin & Alfieri 1991). In our study, however,
observers could not interact with the inspectors and did not form preferences for
individuals who were better inspectors. This implies that observation alone does
not provide accurate information about another individual’s cooperative
tendencies relative to oneself. These data indirectly support the notion that
cooperation via tit-for-tat is most likely to evolve when individuals are allowed
to interact directly with one another.
การแปล กรุณารอสักครู่..
![](//thimg.ilovetranslation.com/pic/loading_3.gif?v=b9814dd30c1d7c59_8619)