motivational state. Thus, for example, all of the studies conducted to date
have used audiences consisting of a single individual. Presumably, this is a
relatively rare situation in nature, where several individuals are likely to be
in view of a cockerel finding food; minimally, there will be several individuals within hearing range. How would a cockerel’s food-calling behavior be
affected by the presence of its mate and an unfamiliar female, or its mate
and a cockerel? In addition to audience composition, it would be useful to
look at the interaction between the cockerel’s hunger level and the composition of the audience. If a cockerel is extremely hungry, and only a
limited amount of food is present, does it remain silent even if a female is
nearby? Finally, does the probability of remaining silent change as a function of whether the food is shareable as opposed to nonshareable? Are
cockerels more likely to call if the food is spread out, thereby reducing the
potential costs of competition should other individuals approach?
Showing that animals are sensitive to an audience is a critical component
in investigations of lies of omission. Cockerels are not only sensitive to the
presence or absence of another chicken but are also sensitive to whether
the audience is male or female and, if female, whether they are familiar or
unfamiliar. A crucial question then is whether such sensitivity is preserved
in kind across contexts. If it is, then the system is quite rigid. Marler, Evans,
and their colleagues have investigated this problem in considerable detail,
and the results indicate considerable flexibility rather than rigidity. In striking contrast to the effects of an audience on food-calling behavior, alarm
calls are potentiated equally by hens and cockerels. That is, the rate of
alarm-call production is the same for male and female audiences, and this
is true of real audiences as well as audiences simulated by video playbacks
(Evans and Marler 1991, 1995). The rate of alarm-call production is higher
in the presence of either a cockerel or hen than it is when there is no audience present or when the audience is comprised of a different species, such
as a bobwhite quail. The decrease in alarm-call rate in the presence of bobwhite quail is not due to their smaller size because chickens produce a
higher rate of alarm calls to chicks who are even smaller than quail. Finally,
cockerels call more to a sexually receptive mate than to a broody hen with
and without chicks and also call more when testosterone levels are elevated.
In summary, chickens have the capacity to withhold information in the
context of food and predation. Chickens are capable of committing lies of
omission, but the social consequences of such deception remain unclear.
Cheney and Seyfarth’s work on the alarm-call system of vervet monkeys
in Amboseli National Park, Kenya, represents one of the best-studied functionally referential call systems (Struhsaker 1967; Seyfarth et al. 1980a,
1980b; Cheney and Seyfarth 1981, 1990; Marler 1985; Hauser 1996). These
vervets produce a suite of acoustically distinctive alarm calls in the context
of predator encounters. Of the set produced, the best-studied are those
given to snakes, eagles, and leopards. These three predator types exhibit different hunting strategies, and such differences appear to have led to the
120
W.T. Fitch and M.D. Hauser
have used audiences consisting of a single individual. Presumably, this is a
relatively rare situation in nature, where several individuals are likely to be
in view of a cockerel finding food; minimally, there will be several individuals within hearing range. How would a cockerel’s food-calling behavior be
affected by the presence of its mate and an unfamiliar female, or its mate
and a cockerel? In addition to audience composition, it would be useful to
look at the interaction between the cockerel’s hunger level and the composition of the audience. If a cockerel is extremely hungry, and only a
limited amount of food is present, does it remain silent even if a female is
nearby? Finally, does the probability of remaining silent change as a function of whether the food is shareable as opposed to nonshareable? Are
cockerels more likely to call if the food is spread out, thereby reducing the
potential costs of competition should other individuals approach?
Showing that animals are sensitive to an audience is a critical component
in investigations of lies of omission. Cockerels are not only sensitive to the
presence or absence of another chicken but are also sensitive to whether
the audience is male or female and, if female, whether they are familiar or
unfamiliar. A crucial question then is whether such sensitivity is preserved
in kind across contexts. If it is, then the system is quite rigid. Marler, Evans,
and their colleagues have investigated this problem in considerable detail,
and the results indicate considerable flexibility rather than rigidity. In striking contrast to the effects of an audience on food-calling behavior, alarm
calls are potentiated equally by hens and cockerels. That is, the rate of
alarm-call production is the same for male and female audiences, and this
is true of real audiences as well as audiences simulated by video playbacks
(Evans and Marler 1991, 1995). The rate of alarm-call production is higher
in the presence of either a cockerel or hen than it is when there is no audience present or when the audience is comprised of a different species, such
as a bobwhite quail. The decrease in alarm-call rate in the presence of bobwhite quail is not due to their smaller size because chickens produce a
higher rate of alarm calls to chicks who are even smaller than quail. Finally,
cockerels call more to a sexually receptive mate than to a broody hen with
and without chicks and also call more when testosterone levels are elevated.
In summary, chickens have the capacity to withhold information in the
context of food and predation. Chickens are capable of committing lies of
omission, but the social consequences of such deception remain unclear.
Cheney and Seyfarth’s work on the alarm-call system of vervet monkeys
in Amboseli National Park, Kenya, represents one of the best-studied functionally referential call systems (Struhsaker 1967; Seyfarth et al. 1980a,
1980b; Cheney and Seyfarth 1981, 1990; Marler 1985; Hauser 1996). These
vervets produce a suite of acoustically distinctive alarm calls in the context
of predator encounters. Of the set produced, the best-studied are those
given to snakes, eagles, and leopards. These three predator types exhibit different hunting strategies, and such differences appear to have led to the
120
W.T. Fitch and M.D. Hauser