- seCtIon FoUR: eVoLUtIon
of what another chimp can see. They might have
an evolved tendency to look where someone else
is looking. To find out, careful experiments are
needed.
Chimpanzees beg for food from humans and from
each other. In an ingenious series of experiments,
Povinelli (1998) and his colleagues used this
behaviour to find out whether chimpanzees know
what someone else can see. First they tested the
chimps to make sure that they begged for food
from an experimenter out of their reach, and did
not beg for inedible items. Then two experiment-
ers offered them food; one had a blindfold over
her mouth and the other had one over her eyes.
The chimps came into the lab, paused, and then
begged for the food, but they were just as likely to
gesture to the person who could not see them as
the one who could. This was even true when one
experimenter had a bucket over her head. Some-
times, when their begging failed to elicit any food,
they begged again, as though puzzled at getting
no response.
They seemed to pass one test: when one person turned her back the chimpan-
zees were less likely to gesture to her. However, when both experimenters sat
with their backs to the apes and one looked back over her shoulder, the chimpan-
zees gestured randomly to both. They seemed oblivious to the fact that there is
no point begging to someone who cannot see you. This is dramatically different
from the behaviour of human children, who can understand this before they are
three years old.
More recently, researchers have devised tests for theory of mind that do not
require cooperation with humans. One experiment investigated whether subor-
dinate chimps understand what their dominant counterparts know about food
they are competing for. The researchers found that the subordinates can distin-
guish between cases where the dominant has or has not seen the food hidden
or moved: subordinates will go for food unseen by the dominant competitor, but
stay away when they know it has been seen (Hare, Call, and Tomasello, 2001).
Experiments have also distinguished between knowing about others’ knowledge
and about their beliefs. Two chimps take turns choosing from a row of buckets,
some of them containing food. In the first condition (the knowledge–ignorance
test), one chimp sees its competitor observing one of two pieces of food being
hidden and then choosing one of three buckets. Can the chimp use its knowl-
edge of what its competitor knows to determine which bucket might still contain
food? In the second condition, the chimp sees its competitor misled by an exper-
imenter pretending to put food in one bucket. Can it predict the competitor’s
choice (by identifying their false belief about the location of the food)? Six-year-
old children pass both tests, but the chimps fail the false-belief test (Kaminski, Call
and Tomasello, 2008). Some have speculated that apes can represent relations
between agents and information that is true from their perspective, but cannot
‘Unless one needs to
discuss behaviour, or to
catch a Hollywood spy,
submentalising may be
the smart option’
(Heyes, 2017, p. 2)
FIGURE 10.11 • Deception and theory of mind
are closely linked. Only a
creature capable of attributing
mental states to others would
hope to get away with illicit
activity by hiding behind a rock.