absence of food) and one male honest (100% of his calls are produced in
the presence of food). Once a female has been exposed to these two males,
set up a mate-choice experiment and record the female’s preferences. These
experiments, accompanied by others that focus more specifically on the
costs and benefits of honest as opposed to deceptive food calls, will help us
understand how male chickens can get away with such high levels of deceptive behavior.
The piece of this story that has been neglected is the skepticism of the
receiver. Most of our discussion has focused on how individuals can manipulate the behavior of receivers by using functionally deceptive acoustic
signals. However, selection will favor both mechanisms that facilitate deception and those that enable accurate skepticism. Is there evidence of skepticism? The honey bee dance language represents an exquisite example of
a functionally referential signal.As decades of research have revealed, when
honey bees dance, attentive listeners extract information about the location, quality, and distance to food using visual, acoustic, olfactory, and tactile
cues (von Frisch 1967; Gould and Towne 1987; Gould and Gould 1988; Dyer
and Seeley 1989; Michelsen et al. 1992; Seeley 1992). No other signaling
system in the animal kingdom is this precise, with the exception of human
language (Hauser 1996). To assess whether individuals are ever skeptical of
the information conveyed in a dance, Gould (1990) conducted an ingenious
experiment. Using a hive with a long history of experience in one location,
he removed a group of foragers and trained them to move back and forth
from a point on land to a second point on land where a pollen-filled boat
was located. Over time, he increased the distance between these two points
and also moved the boat from a position on land to a position out in the
middle of a lake; throughout the training period, Gould prevented the bees
from returning to their hive. Once the bees reliably traveled to the boat and
back, he allowed them to return to the hive and dance.
When the trained foragers returned to the hive, they danced, indicating
that a rich food source could be found out in the middle of the lake.
Although the bees watched the foragers dance, relatively few of them flew
off to the boat. Why? Because, as Gould argued, food has never been found
out in the middle of this lake, or presumably any lake, and thus the information in the signal was unreliable and inaccurate. The hive members
refused to move, treating the signal skeptically. This interpretation is quite
reasonable when one takes into account the results of a control experiment.
Specifically, Gould trained a second group of foragers to find food in a boat
located on the water but along the edge of the lake. When the foragers
returned and danced, other individuals immediately left the hive and flew
to the pollen-filled boat; presumably, the edge of the lake represents a more
likely place to find bee food.
What we do not learn from Gould’s work is the nature of the information stored in the bee’s brain—the extent to which an individual’s own
knowledge of pollen location can override the social message. For example,
3. Unpacking “Honesty”
115
the presence of food). Once a female has been exposed to these two males,
set up a mate-choice experiment and record the female’s preferences. These
experiments, accompanied by others that focus more specifically on the
costs and benefits of honest as opposed to deceptive food calls, will help us
understand how male chickens can get away with such high levels of deceptive behavior.
The piece of this story that has been neglected is the skepticism of the
receiver. Most of our discussion has focused on how individuals can manipulate the behavior of receivers by using functionally deceptive acoustic
signals. However, selection will favor both mechanisms that facilitate deception and those that enable accurate skepticism. Is there evidence of skepticism? The honey bee dance language represents an exquisite example of
a functionally referential signal.As decades of research have revealed, when
honey bees dance, attentive listeners extract information about the location, quality, and distance to food using visual, acoustic, olfactory, and tactile
cues (von Frisch 1967; Gould and Towne 1987; Gould and Gould 1988; Dyer
and Seeley 1989; Michelsen et al. 1992; Seeley 1992). No other signaling
system in the animal kingdom is this precise, with the exception of human
language (Hauser 1996). To assess whether individuals are ever skeptical of
the information conveyed in a dance, Gould (1990) conducted an ingenious
experiment. Using a hive with a long history of experience in one location,
he removed a group of foragers and trained them to move back and forth
from a point on land to a second point on land where a pollen-filled boat
was located. Over time, he increased the distance between these two points
and also moved the boat from a position on land to a position out in the
middle of a lake; throughout the training period, Gould prevented the bees
from returning to their hive. Once the bees reliably traveled to the boat and
back, he allowed them to return to the hive and dance.
When the trained foragers returned to the hive, they danced, indicating
that a rich food source could be found out in the middle of the lake.
Although the bees watched the foragers dance, relatively few of them flew
off to the boat. Why? Because, as Gould argued, food has never been found
out in the middle of this lake, or presumably any lake, and thus the information in the signal was unreliable and inaccurate. The hive members
refused to move, treating the signal skeptically. This interpretation is quite
reasonable when one takes into account the results of a control experiment.
Specifically, Gould trained a second group of foragers to find food in a boat
located on the water but along the edge of the lake. When the foragers
returned and danced, other individuals immediately left the hive and flew
to the pollen-filled boat; presumably, the edge of the lake represents a more
likely place to find bee food.
What we do not learn from Gould’s work is the nature of the information stored in the bee’s brain—the extent to which an individual’s own
knowledge of pollen location can override the social message. For example,
3. Unpacking “Honesty”
115