Istituto di Psicologia del
CNR, Reparto di Psicologia Comparata, Rome,
Vergelijkende Fysiologie, Rijksuniversiteit
Utrecht, the Netherlands, and Cattedra di
Ecologia ed Etologia Animale,Università
di Roma "La Sapienza" Italy
Among primates, yawning is a widespread
behvior, although it occurs at low frequency.
Different hypotheses have been formulated about
its possible functions (Hadidian 1980). Since
the most serious injuries in primates are
usually caused by biting, yawning may be an
intimidating display of the teeth toward
In all the species of nonhuman primates for
which data on sex differences in the frequency
of yawning are reported, males yawn much more
than females (Bertrand 1969; Deputte 1978; Goy
and Resko 1972; Hadidian 1980; HA and DeVore
1965; Redican 1975; Rinaldi 1985; Wolfheim and
Rowell 1972). This différence has been
related to sexual dimorphism in canine size: The
larger size of male canines may make the male
yawn a more effective threat.
The aim of the present study was to test the
hypothesis that the sex differences in the
frequency of yawning are related to sexual
dimorphism in canine size. To this end, the
comparative method was used (Eibl-Eibesfeldt
1974; Maynard Smith 1978). All the primate
species for which delailed data on yawning
behavior are available are sexually dimorphic,
but Homo sapiens is a species with minimal
canine dimorphism (Leutenegger and Kelly 1977).
To date, the only study reporting data on sex
differences in human yawning was done in an
experimental setting (Provine and Hamernik
1986). Thus, a naturalistic study of yawning by
Homo sapiens was needed to test the hypothesis
that a minimally dimorphic species does not show
sex differences in the frequency of yawning.
A further aim of the present study consists
of providing quantitative data on the
spontaneous behavior of humans. Such data are
still rare in human ethology (McGuire and
Fairbanks 1977). A detailed description of human
spontaneous behavior seems to be a necessary
first step to reach a full comprehension of the
causation and function of human behavior.
Subjects of the present study were passengers
of the B line of the Rome underground. We chose
this observational setting for two main reasons:
1) It provides observational conditions that are
extremely favorable (visibility of the subjects
is excellent; moreover, since passengers canne
look through the windows, it is considered
normal to look at other people); and 2) most of
the subjects are unfamiliar to each other, a
situation that might elicit slightly assertive
Data were collected during 94 underground
journeys between May 12, 1986, and March 7,
1987. The duration of the journeys was 15 ±
2.87 minutes (M ± SD). Observations were
made between 8 A. M. and 9 P. M. At the
beginning and at the end of each observational
session, i.e., of each journey, the observer
counted the men and the women present in the
section of the carriage chosen for the
observation. The proportion of men (100 X
[total number of men - total number of men +
total number of women]) was considered to be
an estimate of the relative percentage of the
males and the females who were present on the
underground during the data collection. Journeys
in which the carriage was excessively crowded
were excluded from data collection because of
the poor visibility of the subjects. Since only
adult nonhuman primates are reported to show sex
differences in yawning behavior (e.g., Hadidian
1980), subjects whose visually estimated age was
under 18 years were excluded from data
The sampling method was complete recording
(Slater 1978). Yawning was formally defined as a
gaping movement of the mouth accompanied by a
long inspiration followed by a shorter
expiration (Provine et al. 1987). Two different
types of yawn were distinguished: a) The subject
yawns without any attempt to hide or cover
his/her action; and b) the subject covers,
partially or completely, his/her mouth with a
hand, or yawns incompletely with a partially
open mouth. The observer recorded the sex of the
actor and the type of each yawn.
The relative percentages of males and females
present on the underground during the data
collection were 59.3% and 40.7%, respectively. A
total of 267 yawns by 221 different subjects
were recorded; 57.9% and 42.1% of the subjects
who yawned were males and females, respectively.
The expected distribution of the yawns between
males and females based on their relative
presences on the underground was thus
indistinguishable from the observed one (X2 = 0.
1605, N = 22 1, df = 1 , N. S.).
These data indicate that adult human males
and females, unlike several nonhuman primate
species, do not differ in the frequency of
yawning. These results are congruent with sex
différences in canine size. Even if
correlational data do not demonstrate a direct
cause-effect relationship, these data suggest
that sex différences in frequency of
yawning by primates may be related to the sex
dimorphism in canine size.
Males and females displayed the two types of
yawn in different proportions: 49.2% of male and
32.6% of fernale subjects who yawned showed
uncovered (type a) yawns (X2 = 5.049, N = 208,
df = 1, p < 0.05; subjects who showed both
types of yawn were excluded from this
The most conspicuous form of this display is,
thus, used more by males than by females. Hence,
human males seem still to behave in a more
assertive way, even if they do not appear to
differ from females either in canine size or in
the total frequency of yawning. Alternatively,
human females may be more "polite" than
Several studies have reported a direct
influence of testosterone on yawning by nonhuman
primates (Bielert 1978; Goy and Resko 1972;
Wallen and Goy 1977). The plasma testosterone
concentration of human males is higher during
the early morning than during the evening (e.g.,
Rose et al. 1972). In our study, the frequency
of men's yawns recorded between 8 A.m. and 10:30
A.M. (0.543 ± 0.624 yawns/hour/male
subject) did not differ significantly from that
recorded between 6:30 P.m. and 9 P.m. (0.593
0.766 yawns/hour/ male subject) (Mann Whitney U
test; U = 248.5, NI 22, N2 = 23; sec also
Provine et al. 1987; Baenninger 1987).
Therefore, even if data on diurnal variations
have to be interpreted cautiously, this analysis
seems to indicate that human yawning behavior
has become independent from the influence of
testosterone, given the absence of sex
différences in the frequency of
In conclusion, the present study
demonstrates that human beings, unlike more
dimorphic primates, do not show sex differences
in the frequency of yawning. Data on a wider
variety of contexts would be necessary to
dismiss definitively the possibility of any sex
différence in the frequency of human
yawning. Data from Provine and Hamernik (1986),
though gathered in an entirely different
situation, did not reveal any male-female
différence in the frequency of yawning.
If confirmed, these results would suggest that
sex differences in primate yawning behavior may
be associated with sexual dimorphism in canine
R. Some comparative aspects of yawning in
Betta spiendens, Homo sapiens, Panthera leo, and
Papio sphinx. Journal of Comparative Psychology
101: 349-354, 1987.
Bertrand, M. The Behavioral Repertoire of
the Stuptail Macaque, Basel: S. Karger,
Bielert, F. Androgen treatments of young
rhesusmonkeys. In Recent Advances in
Primatology, Vol. 1, D.J. Chivers, J. Herbert
(Eds.). New York: Academic Press, 1978, pp.
Deputte, B. Etude
du bâillement chez deux espèces de
Cercopithecidae, Cercocebus albigena albigena
GRAY et Macaca fascicularis RAFFLES: Recherche
des facteurs de causalité et de fonction.
Mise en evidence des facteurs
L'Université de Rennes, 1978.
Eibl-Eibesfeldt, 1. Grundriss der
vergleichenden Verhaltensforschung. Etologie,
München: R. Piper & Co. Verlag,
Goy, R.W., and Resko, G.A. Gonadal hormones
and behavior of normal and pseudohermaphroditic
nonhuman female primates. Recent Progress in
Hormone Research 28: 707733, 1972.
Yawning in an old world monkey, Macaca nigra
(Primates: Cercopithecidae). Behaviour 75:
Hall, K.R.L., and DeVore, 1. Baboon social
behavior. In Primate Behavior, 1. DeVore (Ed.).
New York: Holt, Rinehart & Winston, 1965,
Leutenegger, W., and Kelly, J.T.
Relationship of sexual dimorphism in canine size
and body size to social, behavioral, and
ecological correlates in anthropoid primates.
Primates 18:117-136, 1977.
Maynard Smith, J. Optimization theory in
evolution. Annual Review of Ecology of
Systematics 9: 31-56, 1978.
McGuire, M.T., and Fairbanks, L.A. Ethology:
Psychiatry's bridge to behavior. In Ethological
psychiatry, M.T. McGuire, L.A. Fairbanks (Eds.).
New York: Grune and Stratton, 1977, pp.
Redican, W.K. Facial expression in nonhuman
primates. In Primate Behavior, Vol. 4, L.A,
Rosenblum (Ed.). New York: Academic Press, 1975,
Rinaldi, F. Attività di spostamento
del macaco del Giappone (Macacafuscata). Tesi di
Laurea, Università di Roma "La Sapienza,
Rose, R.M., Kreuz, L.E., Holaday, J.W.,
Sulak, K.J., and Johnson, C.E. Diurnal variation
of plasma testosterone and cortisol. Journal of
Endocrinology 54: 177-178, 1972.
Slater, P.J.B. Data collection. In
Quantitative Ethology, P.N. Colgan (Ed.). New
York: John Wiley & Sons, 1978, pp.
Wallen, K., and Goy, R.W. Effects of
estradiol benzoate, estrone, and propionate of
testosterone or dihydrotestosterone on sexual
and related behaviors of ovariectomized rhesus
monkeys. Hormones and Behavior 9: 228-248,
Wolfheim, J.H., and Rowell, T.H.
Communication among captive talapoin monkeys
(Miopithecus talapoin). Folia Primatologica 18: