Comparative developmental psychologists are
interested in the emergence and development of
empathy in both humans and nonhumans. By
empathy, we mean the ability to identify with
another individual's emotions and cognitive
states; it is characteristic of all normal
humans from early childhood. Even today there is
debate among primatologists and psychologists
about whether and to what extent our nearest
evolutionary neighbors, the great apes, share
the capacity for empathy that we humans take for
granted (Gallup 1998; Preston and de Waal 2002;
Povinelli 1998; Povinelli and Vonk 2003;
Tomasello et al. 2003). Although many people who
work closely with these primates are convinced
that they are capable of reflecting about what
other individuals might be thinking, others
express doubts about the extent and level at
which they do this. Much of the controversy
stems from the variable quality of the evidence
presented in support of empathic abilities. The
evidence comes from a range of observational
studies and controlled experiments, and as we
will see from our brief review of the literature
as it concerns chimpanzees, neither source of
data is problem free.
Observational studies of chimpanzees in the
wild and in naturalistic groups in captivity
have led to identification of a range of
phenomena suggestive of the capacity for empathy
in these apes (O'Connell 1995). For example,
some advanced forms of deception may involve
attributing intentions to other individuals and
deliberately altering others' beliefs, emotions,
or attentional states. The evidence for
intentional deception of this type is stronger
for chimpanzees than for any other species of
nonhuman primates (Byrne 1995; Byrne and Whiten
1992). One of the best known examples took place
in a captive group with access to a large
outdoor enclosure. Menzel (1974) described how
one chimpanzee, Belle, would try to feign
disinterest or misdirect another chimpanzee,
Rock, away from some hidden food of which only
she knew the location. Rock in turn began to
feign disinterest in Belle's activities, only to
suddenly wheel round to detect unintentional
cues from Belle about where the food might
be.
Not only does this example suggest that the
chimpanzees understood each other's intentions,
it also involves counterdeception, a phenomenon
that has never been reported in any species of
monkey. Goodali (1986) describes another
noteworthy case of counterdeception in
chimpanzees, witnessed at a feeding box at
Gombe, Tanzania. One chimpanzee knew that food
was available in the box but feigned disinterest
because a more-dominant chimpanzee was in the
vicinity. The latter made as if to leave the
area, but in fact he hid behind a tree. From
there he looked back to see the first chimpanzee
going for the food in the box, the latter
apparently believing that the dominant had left
the scene. The dominant chimpanzee came back and
got the food.
Although impressive, examples such as those
just cited are open to criticism as evidence for
cognitive empathy. The Gombe incident is an
anecdote, a oneoff occurrence that cannot be
reinstated, whereas Menzel's procedure, although
systematic and controlled (showing hidden food
to one chimpanzee and then releasing the entire
group), is not guaranteed to produce an outcome
similar to the scenes acted out by Rock and
Belle. However, a recent study with some
similarities to Menzel's procedure has also
described the emergence of deception by
withholding information and misleading (Hirata
and Matsuzawa 2001). It is to be expected,
however, that tactical deception will occur only
occasionally in any group, otherwise it would
lose its effectiveness. Also, there is always
the possibility that the deceivers have learned
behavioral tactics simply for changing the
behavior of others, with no regard for what the
others might actually be thinking. In short,
although highly suggestive, instances of
possible deception cannot be accepted as the
strongest evidence on which to base claims about
empathy.
Fortunately, systematic observational
studies of social behavior in groups of
chimpanzees have provided other types of
evidence for empathic abilities. For example,
chimpanzee bystanders often approach to offer
reassuring friendly contact to the victim of an
aggressive act; this behavior is much less
frequently observed in macaques (de Waal and
Aureli 1997). Those authors cautiously
hypothesize that the macaque-chimpanzee
difference in consolation behavior may reflect
species differences in empathy and thus call for
further observational studies and experiments to
test the hypothesis.
Some elaborate and complex experiments have
been designed to explore aspects of the capacity
for empathy. In one of their pioneering
experiments, Premack and Woodruff (1978) showed
an adult female chimpanzee, Sarah, video clips
of a human faced with a problem, for example, a
man trying to exit a room through a locked door.
The chimpanzee was then shown several
photographs, one of which depicted the solution
to the problem, which in this case would be a
key. Even without specific training on the task,
garah typically selected the appropriate
photograph, which the authors considered as
evidence that she was able to understand the
actors' intentions. One limitation of this work
is that the correct response might have been
based on simple association between the problem
and the solution, rather than on an
understanding of the actor's predicament.
Other experiments have focused on
chimpanzees' attribution of knowledge to other
individuals, an ability that qualifies as
empathic as it implies identifying with
another's cognitive (nonemotional) state. Most
of these studies have assessed chimpanzees'
processing of others' gaze direction as giving
rise to knowledge. One study showed that
chimpanzees selected the container indicated to
them by a human who had watched while that one
of several containers was baited with food, in
preference to a container indicated by someone
who did not see the baiting procedure (Povinelli
et al. 1990). Other experiments have shown that
chimpanzees can find hidden food through
monitoring and acting on gaze cues from humans
or from other chimpanzees (Itakura and Tanaka
1998, Itakura et al. 1999). If one chimpanzee
can see that a more-dominant individual has
looked at a piece of food, the first is less
likely to make a move for that food, but may
instead go for another piece that is out of view
of the dominant (Hare et al. 2000).
Similar to observations from the field,
evidence from experimental studies of empathy
may be open to alternative interpretations. For
some skeptics, in many laboratory experiments
ecological validity has been too compromised. It
is true that some experiments involve
complicated procedures and humans acting in
quite unnatural ways, such as placing a bucket
over their head and then removing it just before
interacting with the chimpanzee, or pointing to
one object while staring at another object.
Also, as with some of the more naturalistic
studies, replication of results is not
guaranteed; there are examples of very similar
experiments leading to quite different results.
Finally, there is the possibility that over
repeated trials subjects might simply learn to
perform the correct response, without engaging
in mental attribution at all (Heyes 1998). What
is clear is that the debate over empathic
abilities in nonhuman species is in need of new
types of data, preferably from a variety of
approaches and perspectives.
A yawn response during presentation
of a yawn videotape. Ai watches a yawn on the
screen (top left), starts to yawn as the
stimulus yawn ends (top right), continues to
yawn (bottom left), and completes the yawn while
the screen is blank (bottom right) See
the video
2 Yawning and Empathy
Recently, we approached the debate on
empathy in chimpanzees from a new angle. Our
basic question was simple: Do chimpanzees show
contagious yawning, as humans do? Although the
link between contagious yawning and empathy
might not be immediately obvious, it has been
made (Lehmann 1979) and tested (Platek et al.
2003). The latter study examined the
relationship between empathy and contagious
yawning in humans, first by asking participants
(university students) to complete the
Schizotypal Personality Questionnaire (SPQ),
part ofwhich measures empathic tendencies, next
asking them to interpret stories designed to
measure mental state attribution, and then
exposing the participants td a videotape shbwing
a sequence of yawns. The use of video stimuli to
study contagious yawning in humans was pioneered
by Provine (1986), who found that more than 50%
of adults would yawn within a few minutes of
starting to watch a video of someone yawning
repeatedly. Platek et al. reported that 40% of
their subjects yawned in response to the yawn
stimulus videotape, and that people who scored
higher on measures of empathy and mental state
attribution skills were more likely to show the
effect. Furthermore, they found that contagious
yawners were faster than others at recognizing
their own face when it appeared on a computer
screen. Platek et al. (2003) made a case for
contagious yawning being related both to
self-awareness, as measured by self-recognition,
and to empathic tendencies, as measured by the
SPQ and mental attribution tests.
Thus, the scene was set for our
investigation into contagious yawning by
chimpanzees. Chimpanzees are one of the few
species (along with humans, other great apes,
and cetaceans) known to show self-recognition,
for example, in a mirror or on video (Gallup
1970; Anderson and Gallup 1999). Furthermore, as
we have already indicated, there is evidence for
empathic abilities in chimpanzees beyond those
seen in monkeys (de Waal 1996; Preston and de
Waal 2002). Therefore, the following hypothesis
emerged: If self-awareness underlies the
capacity for empathy, as proposed explicitly by
Gallup (1982), and empathy is linked to
contagious yawning (in humans; Platek et al.
2003), then chimpanzees might also be
susceptible to contagious yawning. Until our
interest was kindled, this possibility had never
been considered in the literature. In fact,
there appears to have been a widespread
assumption that, although yawning is ubiquitous
among vertebrates, the phenomenon of contagious
yawning is uniquely human (Baenninger 1987;
Lehmann 1979; Smith 1999).
3 Yawning in Chimpanzees
Chimpanzees appear to yawn in much the same
way as humans do, although there are no
comparative studies of physiological aspects of
the act in the two species. As do humans,
chimpanzees may yawn when they are tired, bored,
or in a mildly disturbing situation (Goodall
1968). It is striking, however, that in all the
books and journal articles describing hundreds
of thousands of hours of observations of
chimpanzees in captivity and in the wild, there
do not appear to be any descriptions of
contagious yawning, by which we mean one
chimpanzee yawning in response to seeing another
chimpanzee yawning.
It is important to note that chimpanzees,
like humans, differ from Old World monkeys in
that yawning by adult males is not a form of
ritualized display expressed in situations of
male-male confrontation. This type of display
was noted by Darwin (1965/1872) and has been
studied in macaques in some detail (Adams and
Schoel 1982; Deputte 1994). Similar to human
yawning, chimpanzee yawning appears to be devoid
of any agonistic signaling function. But, if
chimpanzee yawning is more like that of humans
than that of Old World monkeys, why are there no
accounts of contagious yawning in our great ape
relatives? At this stage two possibilities may
be considered: l) contagious yawning in
chimpanzees does not exist, or (2) contagious
yawning does exist but it has been overlooked by
chimpanzee researchers. This latter possibility
may be true because yawning occurs at times of
the day when it is unlikely to be observed, or
else it occurs less frequently than in humans.
It is clear that only field studies and careful
observational studies of captive groups can
confirm or contradict these two
possibilities.
In the absence of relevant data, we designed
our study to address a third possibility, which
was that if chimpanzees are susceptible to
contagious yawning, the effect should be
observable under experimental conditions
involving exposure to repeated yawn stimuli,
similar to those used to study the phenomenon in
humans.
4 Experimental Demonstration of
Contagious Yawning in Chimpanzees
Inspired by reports that between 40% and 55%
of human adults will yawn shortly after starting
to watching sequences of yawns on a television
monitor, we prepared two "yawn" videotapes, each
showing ten short clips (6-8 s each) of
chimpanzees yawning, with each clip separated by
a blank (blue) screen for 6 to lOs. One
videotape featured chimpanzees from the Primate
Research Institute (PRI) group, therefore highly
familiar to the subjects, while the other
videotape featured unfamiliar chimpanzees from
the Mahale Mountains, Tanzania. Although every
clip was centered around a chimpanzee yawn, the
clips were otherwise varied, and included young
infants yawning, close-up and middledistance
views of yawning, and chimpanzees either sitting
or lying down while yawning (Fig. 1). Two
"control" videotapes were also prepared; these
were similar in structure to the yawn videotapes
but showed chimpanzees displaying a variety of
facial expressions such as grinning or
threatening, but not yawning. Each videotape
lasted approximately 3 min.
The main subjects were six adult female
chimpanzees, ranging in age from 19 to 27 years;
they were all members of a social group housed
at the Primate Research Institute (PRI) of Kyoto
University. Three of the females had juveniles,
aged 3 years, and the latter were also included
as subjects. In fact, the juveniles were
considered important in this research because
the only published study to date of the
development of contagious yawning in human
children failed to find any evidence of this in
children below the age of 5 years (Anderson and
Meno 2003). The chimpanzee adults and juveniles
had all been subjects in a wide range of
noninvasive behavioral research (Matsuzawa 2000,
2001, 2003, 2005), and all were familiar with
the experimental booth. Three sides of the booth
had glass walls, which allowed the chimpanzees
to be filmed from different angles throughout
the session. Access to the booth from the
chimpanzees' outdoor enclosure was via a series
of hatches and tunnels.
Each adult was tested individually except
for the three mothers, who came to the booth
accompanied by their juveniles. Participation in
the sessions was voluntary, in that the
chimpanzees could opt to come in from their
outside enclosure in response to coaxing by one
of the experimenters (T.M.), or they could
decline. The order in which the chimpanzees were
tested therefore depended on period, the
chimpanzee was allowed to rejoin the group in
the outside enclosure. Throughout the session,
apart from the distraction period, the humans in
the room remained motionless and nutral, except
for occasional prompts by T.M. to watch the
video if the chimpanzee seemed to lose attention
(although they are generally attentive to short
movie clips; Morimura and Matsuzawa 2001). Any
occurrences of yawning by the chimpanzees were
noted in real time and verified by later
analysis of videotapes of the sessions, with
100% agreement between the authors.
The results of the experiment were striking
(Anderson et al. 2004). Overall, the adult
chimpanzees yawned more than twice as frequently
during the yawn video trials (exposure and
postvideo periods combined) than during control
trials (totals: 67 and 30, respectively).
Individual binomial tests showed that for two of
the females the difference between the yawn and
control trials was extremely significant, with
many more yawns during the yawn condition.
Figure 2 illustrates the data for the individual
adult females. The equivalent figure in the
original publication erroneously showed
chimpanzee Mari as yawning nine times during the
control videos, whereas in fact she only did so
four times, as illustrated here. These data
indicate that in these adult female chimpanzees,
the video-induced contagious yawning effect was
shown in 33% of the sample. There were no clear
differences in response to the PRI and Mahale
videotapes.
Since the publication of the original
report, re have carried out some additional
analyses on the data. There were no significant
correlations between age of the adults and
frequency of yawning in video trials, control
trials, or both kinds combined. However, it is
noteworthy that the contagious yawning effect
was found in two of the three oldest females,
all three being over 25 years old. Binomial
tests indicated that for both yawn and control
trials, significantly more yawns occurred in the
3-min postvideo period than during the 3-min
exposure period (P < 0.05 and P < 0.01 for
yawn and control trials, respectively). When the
same analysis was run separately for the tapes
showing the PRI and Mahale chimpanzees, the
difference remained significant only for the PRI
tapes (both at P <0.05). Figure 3 illustrates
a yawn in response to a yawn by one of the
females.
A final salient outcome of the experiment
concerns the three 3-year-olds. In spite of
seeing the same videotapes and also seeing their
mothers yawning during the tests, none of the
juvenile chimpanzees ever yawned during the
sessions.
5 Contagious Yawning in Chimpanzees:
Discussion and Implications
Although the experiment just described must
be seen to be preliminary, we hope that it has
paved the way for a new approach to the thorny
issue of how to assess empathic abilities
through experiments with nonhumans. In contrast
to many previous experiments, there is no
complicated procedure in which the chimpanzees
observe humans behaving in unusual ways, nor is
there any experimenter-deemed "correct" response
that the ape might figure out over many
repetitions of the task. Indeed, a case could be
made for our procedure having greater ecological
validity than some of the other studies that are
presented as evidence in the empathy debate.
Although it is true that wild chimpanzees will
not encounter videotapes showing sequences of
clips depicting yawns, neither do humans under
normal circumstances, and exposing subjects to
video stimuli is simply a way of ensuring some
control over the situation.
To discuss the experiment, we think it is
enlightening to consider two comments offered by
the editor of an esteemed psychological journal
to which we submitted the original paper for
publication. In his letter of rejection (the
manuscript was not sent out for review), he
wrote that because only two of the six adult
female chimpanzees showed a positive response,
we could not conclude that there was a
group-level effect. Actually, we have never made
a claim for a group-level effect. It is
noteworthy that in studies in which human adults
are exposed to video yawn stimuli, the
percentage of participants that report yawning
in response to the stimuli varies from 40% to
55% (Provine 1986; Platek et al. 2003). This
result would not be considered a group-level
effect either, but nobody doubts the robustness
of contagious yawning in humans.
Indeed, in one respect we consider our data
to be even more compelling than the human data:
Unlike humans, the chimpanzees had no idea what
was expected of them. In experiments with
humans, the participants are usually instructed
to report their own yawns, for example, by
pressing a counter. This procedure is used to
avoid the inhibiting effect of being observed,
but it almost certainly leads to an inflated
occurrence of yawning, because by the very act
of focusing on their yawning, and being likely
to guess the purpose of being exposed to
repeated yawn stimuli, the participants
themselves are likely to induce the response.
Viewed in this light, a 33% occurrence of
contagious yawning in naïve chimpanzees,
with no such inhibitions and no hypotheses about
the experimenters' aims, is particularly
impressive.
The second comment by the journal editor was
that while on a safari trip he had seen several
lions lying down together and yawning, and that
this challenged the view that chimpanzees'
contagious yawning was related to empathy. Here,
he is confusing two things, namely,
synchronization of activity and the contagion
effect. It is true that after feeding and other
essential activities have been taken care of,
members of a pride of lions may start yawning
during a period of rest (McKenzie 1994), but
this is probably based on shared behavioral and
physiological rhythms, rather than any one
lion's yawns inducing the same behavior in
others. Admittedly, there is no direct evidence
on the issue, but we think it unlikely that
lions are susceptible to contagious yawning. In
fact, it is worth pointing out that the term
"contagious yawning" is something of a misnomer,
as yawns do not necessarily spread throughout an
entire group of humans (or chimpanzees), and the
likelihood of any one individual yawning in
response to seeing someone else yawn probably
varies as a function of a host of variables that
have yet to be clarified. The research by Platek
et al. (2003) has identified differences in
self-recognition ability and empathic tendencies
as a source of variability in humans; whether
the same applies to chimpanzees is certainly
worthy of investigation. Preston and de Waal
(2002) offer an insightful discussion of
different mechanisms underlying what we might
call empathic behavior.
Among the feedback that came from colleagues
upon publication of the article by Anderson et
al. (2004) was the suggestion that somehow the
two adult females who showed the contagious
yawning effect might have done so because they
had a history of particularly close contact with
humans and were highly test experienced. This
argument has little validity. Although there may
be cases of "enculturated" apes showing better
performance in tests in which humans try to get
them to perform in particular ways, this kind of
situation is far removed from our setup. First,
all the chimpanzees live in a group, and they
are much more chimpanzee oriented than
peoplè oriented. Second, the chimpanzees
were spontaneously responding to images of
chimpanzees, not images of humans. Finally, as
we have already ifldicated, there was nô
'task" that the chimpanzee had to figure out,
and no kind of reward involved (other than any
possible intrinsic satisfaction from
yawning).
The discovery of contagious yawning in
chimpanzees has raised many other questions, a
few of which we present here to close our
discussion. When in ontogeny does contagious
yawning start to occur, and what psychological
changes cause it to occur? The fact that none of
the three juveniles ever yawned during the
sessions strengthens the view that the
mechanisms underlying the phenomenon are similar
in chimpanzees and humans, as human children
below the age of 5 years also appear to show no
effect (Anderson and Meno 2003). What stimuli
are effective for eliciting contagious yawning?
From personal experience, we can say that humans
may be induced to yawn by seeing another species
yawning; in fact, we can easily (and sometimes
erroneously) empathize with other animals. Would
chimpanzees yawn in response to seeing another
species yawn? We are currently addressing this
precise question experimentally.
What neural substrates are implicated in
contagious yawning? Recent brain imaging studies
of humans exposed to yawn stimuli have started
to provide some answers to this question. Platek
et al. (2005) reported different patterns of
neural activity in response to watching video
clips of yawning and laughing. In particular,
along with brain regions associated with general
face perception, there was specific activation
in posterior midline cortical regions that have
been associated with other aspects of
self-processing, such as autobiographical memory
and self-monitoring. Schürmann et al.
(2005) reported greater activation in parts of
the superior temporal sulcus in response to
watching yawn sequences than non-yawn mouth
movements. This region is known to be associated
with the perception of biological motion.
Importantly, neither of these recent studies
found evidence of specific activation of the
so-called mirror neuron system during exposure
to yawns, or of prefrontal cortical regions
thought to be important in theory of mind. These
results indicate that whatever underlies
contagious yawning, it does not appear to be
based either on conscious imitation or
higher-level, "conscious" empathy. More studies
on a range of nonhuman species and experimental
conditions should be useful for identifying
which links in the perception-action chain (see
Preston and de Waal 2002) are involved in this
curious and fascinating behavior.
References
Adams DB, Schoel WM,(1982) A statistical
analysis of the social behavior of the male
stumptail macaque (Macaca arctoides). Am J
Primatol 2:249-273
Anderson JR, Gallup GG Jr (1999)
Self-recognition in primates: past and future
challenges. In: Haug M, Whalen RE (eds) Animal
models of human emotion and cognition. American
Psychological Association, Washington, DC, pp
175-194
Anderson
JR, Meno P (2003) Psychological influences
on yawning in children. Curr Psychol Lett: Behav
Brain Cogn 11. See
http://cpl.revues.org/document390.html.
Baenninger
R (1987) Some comparative aspects of yawning
in Betta spiendens, Homo sapiens, Panthera leo,
and Papio sphinx. J Comp Psychol
101:349-354
Byrne R (1995) The thinking ape:
evolutionary origins of intelligence. Oxford
University Press, Oxford
Byrne RW, Whiten A (1992) Cognitive
evolution in primates: evidence from tactical
deception. Man 27:609-627
Darwin
C (1965/1872) The expression of the emotions
in man and animals. University of Chicago Press,
Chicago
Deputte
BL (1994) Ethological study of yawning in
primates. I. Quantitative analysis and study of
causation in two species of Old World monkeys
(Cercocebus albigena and Macacafascicularis).
Ethology 98:221-245
de Waal F (1996) Good natured: the origins
of right and wrong in humans. Harvard University
Press, Cambridge, MA
de Waal FBM, Aureli F (1997) Conflict
resolution and distress alleviation in monkeys
and apes. Ann N Y Acad Sci 807:317-328
Gallup GG Jr (1970) Chimpanzees:
self-recognition. Science 167:86-87
Gallup GG Jr (1982) Self-awareness and the
emergence of mind in primates. Am J Primatol
2:237-248
Gallup GG Jr (1998) Can animals empathize?
Yes. Sci Am 9:65-75
Goodall J (1968) The behaviour of
free-living chimpanzees in the Gombe Stream
Reserve. Anim BehavMonogr 1:161-311
Goodall 1(1986) The chimpanzees of Gombe:
patterns of behavior. Harvard University Press,
Cambridge, MA
Hare B, Call J, Agnetta B, Tomasello M
(2000) Chimpanzees know what conspecifics do and
do not see. Anim Behav 59:771-786
Heyes C (1998) Theory of mind in nonhuman
primates. Behav Brain Sci 21:101-134
Hirata S, Matsuzawa T (2001) Tactics to
obtain a hidden food item in chimpanzee pairs
(Pan troglodytes). Anim Cogn 4:285-295
Itakura S, Tanaka M (1998) Use of
experimenter-given cues during object-choice
tasks by chimpanzees (Pan troglodytes), an
orangutan (Pongo pygmaeus), and human infants
(Homo sapiens). J Comp Psychol 112:119-126
Itakura S, Agnetta B, Hare B, Tomasello M
(1999) Chimpanzees use human and conspecific
social cues to locate hidden food. Dey Sci
2:448-456
Lehmann HE
(1979) Yawning: a homeostatic reflex and its
psychological significance. Bull Menninger Clin
43:123-136
Matsuzawa T (2000) Chimpanzee mind
1995-2000: collection of articles. Research
report by the grant-in-aid from the Ministry of
Education, Culture, Sports, Science and
Technology, Japan
Matsuzawa T (2001) Primate origins of human
cognition and behavior. Springer, Tokyo
Matsuzawa T (2003) The Ai project:
historical and ecological contexts. Anim Cogn
6:199-211
Matsuzawa T (2005) Chimpanzee mind
2000-2005: collection of articles. Research
report by the grant-in-aid from the Ministry of
Education, Culture, Sports, Science and
Technology, Japan
McKenzie AA (1994) The tonsillar evacuation
hypothesis of yawning behaviour. S Afr J Sci
90:64-66
Menzel EW (1974) A group of chimpanzees in a
1-acre field: leadership and communication. In:
Schrier AM, Stollnitz F (eds) Behavior of
nonhuman primates. Academic Press, New York, pp
83-153
Morimura N, Matsuzawa T (2001) Memory of
movies by chimpanzees (Pan troglodytes).J Comp
Psychol 115:152-158
O'Connell SM (1995) Empathy in chimpanzees:
evidence for theory of mind? Primates
36:397-410
Smith EO (1999)
Yawning: an evolutionary perspective. Hum Evol
14:191-198
Tomasello M, Call J, Hare B (2003)
Chimpanzees understand psychological states: the
question is which ones and to what extent.
Trends Cogn Sci 7:153-156
Campbell MW
et al. Do chimpanzees yawn
contagiously in response
to 3d computer animations? 2008
