Yawning is a widespread behavioural response
expressed ina ll classes of vertebrates. There
is, however, little agreement on its biological
significance. One current hypothesis states that
yawning serves
a thermoregulatory mechanism that occurs in
response to increases in brain and/or
bodytemperature. The brain-cooling hypothesis
further stipulates that, as ambient temperature
increases and approaches (but does not exceed)
body temperature, yawning should increase as a
consequence. We tested this hypothesis in a
sample of 20 budgerigars, Melopsittacus
undulatus, through the manipulation of room
temperature.
Birds we reexposed to three separate
conditions (control temperature (22C),
increasing temperature (22&endash;34C), and
hightemperature (34&endash;38C)) in a repeated
measures design, with each condition lasting
21min. The incidence of yawning differed
significantly across conditions (4.20 +-2.39)
yawns per bird in the increasing temperature
condition, compared to 2.05 +- 1.90 and 1.25 +-
0.72 yawns per bird, in the high temperature and
control conditions, respectively). Thesef
indings are consistent with the hypothesis that
yawning serves a thermoregulatory function.
Yawning is characterized by a large gaping
of the mouth, accompanied by a deep inhalation
of air, and a shorter expiration. Although
typically studied in humans, yawning is a widely
expressed, stereotyped phenomenon occurring in
all classes of vertebrates (Baenninger 1987).
but little is known about the function of
yawning in any species. Research has shown that
yawning coincides with a variety of
neurochemical interactions in the brain
(Argiolas & Melis 1998). While the
neurological mechanisms underlying yawning are
not entirely clear, research on yawning under
laboratory conditions has proven valuable in
understanding the physiopathology of certain
diseases, as well as the action of new drugs
(Daquin et al. 2001). However, numerous attempts
at identifying the adaptive or biological
significance of the yawn (reviewed by Smith
1999) have led to little consensus (Provine
2005).
Yawning is under involuntary control, and it
cannot be inhibited or elicited by individual
commahd (Provine 2005). Yawning is also
contagious in humans and some nonhuman primates
(Anderson et al. 2004; Paukner & Anderson
2006). In humans, attempts [o shield a yawn
do not prevent its contagion (Provine 2005). The
spontaneous and uncontrollable nature of yawning
across species lends support for it having
adaptive significance. In humans, yawning occurs
before birth as early as 20 weeks after
conception (Sherer et al. 1991), testifying to
its importance postnatally, as many important
postnatal behaviours begin to appear prenatally
(e.g. breathing movements, swallowing and eye
movements) before they develop any functional
significance (Nijhuis 2003).
Throughout the lives of healthy adult
humans, yawning occurs in a consistent pattern
(Gallup & Gallup 2008), occurring most often
during the first hour after wakening and the
last hour before sleeping (Provine et al. 1987a;
Baenninger et al. 1996; Zilli et al. 2007).
Similarly, variation in yawning among rats
appears to have a circadian pattern (Anias et
al. 1984). In addition, stretching has been
shown to accompany yawning almost 50% of the
time in humans (Provine et al. 1987a).
Researchers have attributed such findings to an
association between yawning and increases in
arousal and activity that accompany transitional
states (Provine et al. 1987a; Greco &
Baenninger 1991; Greco et al. 1993; Baenfinger
et al. 1996). Aside from observational reports,
comparative studies investigating yawning in
nonhumans are few and the ethology of yawning in
nonhuman species remains mysterious. Baenninger
(1987) proposed that yawning may actually serve
different functions in different species.
Nevertheless, the tendency for yawning to
correspond with state changes in humans (Provine
et al. 1987a; Greco et al. 1993; Baenninger et
al. 1996) suggests possible adaptive contexts
for this behaviour across species. New evidence
suggests that yawning may be involved in
thermoregulation (Gallup & Gallup 2007,
2008) and may act as a braincooling mechanism.
This hypothesis has been developed for humans
but suggests one general utility across
endotherms. Based on this theory, the yawn
serves as a cooling mechanism that keeps the
brain and/or body in thermal homeostasis, thus
maintening mental efficiency. Increases in
facial blood flow resulting from a yawn may
operate like a radiator, removing hyperthermic
blood from specific areas, while introducing
cooler blood from the lungs and extremities.
Increases in facial blood flow may alter
cerebral blood flow as well (Heusner 1946;
Barbizet 1958; Zajonc 1985). Consistent with the
radiator hypothesis of human brain evolution
(see Falk 1990), the respiratory and arterial
actions that follow the yawn match those
required to cool the brain effectively. An
increase in cranial blood flow due to yawning
may aid in the dissipation of heat via the
emissary veins. In humans, increased arousal, as
measured by skin conductance, occurs during
yawning (Greco & Baenninger 1991), and
vasodialation has been hypothesized to promote
further cooling. Gaping of the mouth and deep
inhalation of air taken into the lungs during a
yawn can also alter the temperature of the blood
travelling from the lungs to the brain through
convection (Gallup & Gallup 2007). This
hypothesis proposes that it is the temperature
of the air that gives the yawn its utility, not
the air's composition. In fact, variation in 02
and/or CO2 concentrations has no effect on yawn
frequency (Provine et al. 1987b).
The brain-cooling hypothesis leads to
several testable predictions. First, it predicts
that there will be a fairly narrow range of
external temperatures, a 'thermal window', over
which yawning can be triggered (Gallup &
Gallup 2007, 2008). As ambient temperature
rises, it becomes increasingly difficult to
maintain thermal homeostasis, but it also
becomes less effective to lower body temperature
by using environmental heat transfer. The
model's central predictions are that (1) the
frequency of yawning should rise as ambient
temperature approaches body temperature and (2)
yawning should not occur when ambient
temperature reaches or exceeds body temperature,
because its cooling component will no longer
occur. Likewise, when temperatures fall below a
certain point, yawning should cease to be
adaptive and could become maladaptive by sending
unusually cool blood to the brain. This
hypothesis is intriguing because it applies
generally across endotherms and suggests
differences in the importance of yawning for
different species, dependent on both morphology
and environment.
To test the central hypothesis, we
manipulated the ambient temperature experienced
by budgerigars in a laboratory environment while
recording yawning, stretching and gular
fluttering, a thermoregulatory response that
promotes evaporative cooling in birds
experiencing heat stress (Bartholomew et al.
1968). Body temperature is a balance between
heat production and heat dissipation, and
raising the ambient temperature would be
expected to trigger compensatory
thermoregulatory mechanisms. We therefore
hypothesized that the frequency of yawning would
increase in response to rising ambient
temperatures, as opposed to when temperature is
held constant. We chose M. undulatus as our
study species because of its large relative
brain size (Iwaniuk & Nelson 2002) as well
as the fact that its natural habitats include
arid Australia where it would be subject to wide
swings in temperature. In addition, a recent
study found n evidence for contagiûlls
yawning in this species (M. L Miller, S. M.
Vicario & A. B. Clark, unpublished data).
Thus, we were able to investigate the frequency
of yawning within small groups with confidence
that any individual's yawns would not influence
yawning in others.
DISCUSSION
The frequency of yawning was significantly
affected by ambient temperature. As ambient
temperature increased, birds were over twice as
likely to yawn, compared to when temperatures
were held constant (both low and high). Yawning
occurred less frequently at low temperatures
(1.25 + 0.72 yawns per bird), slightly more when
held at high temperatures (2.05 ± 1.90
yawns per bird), and most frequently with
increasing temperatures (4.20 ± 2.39 yawns
per bird). Likewise, the strong quadratic
correlation between yawning frequency and
temperature supports the relationship between
yawning and ambient temperature change. These
data are consistent with the hypothesis that
yawning, like gular fluttering, is connected
with thermoregulation. Stretching, although
often seen with yawning at control temperatures,
was not influenced by ambient temperature
manipulation.
Although the rate of yawning peaked around
30°C, during the increasing temperature
condition, it began to decrease in frequency as
temperature further increased (i.e. 34-38
°C during the high temperature condition).
This trend appeared to be influenced by the
prevalence of gular fluttering; while fluttering
was originally positively correlated with the
incidence of yawning at around 25.6 °c,
this trend was reversed by the time all birds
were engaged in this behaviour (i.e.
35.4°C). As gular fluttering is widely
associated with thermoregulation (Bartholomew et
al. 1968), we argue that this respiratory
mechanism may supplant yawning, especially when
temperature exceeds some critical point around
35.4 oC. That is, yawning may be inhibited when
continuous gular fluttering is required to
prevent hyperthermia. Yawning appears to be an
initial response associated with thermal
homeostasis; as temperature increases and heat
dissipation becomes more difficult, more
effective regulatory mechanisms, such as the
gular flutter, are triggered. This corroborates
the view that yawning serves as a compensatory
rather than primary cooling mechanism (Gallup
& Gallup 2007). Furthermore, as ambient
temperature approaches body temperature, one
would expect yawning to diminish in frequency
(Gallup & Gallup 2007). Although the ambient
temperature in this study never exceeded
budgerigar body temperature (39.5 °C),
attenuation of yawn frequency at 35.4 oc is
consistent with this prediction. At 35.4
°C, the cooling capacity of the yawn (i.e.
difference between ambient and body temperature)
was less than that at lower temperatures.
The incidence of stretching was not affected
by ambient temperature. There was no difference
in stretching among temperature conditions, and
the incidence of stretching did pot vary across
the range of temperatures within this experiment
(P> 0.9), nor was there a correlation between
the incidence of yawning and stretching. Within
the increasing and high temperature conditions,
there was also no observed relationship between
stretching and gular fluttering. Therefore, we
propose that unlike yawning, stretching appears
to be independent of thermoregulation in this
species. In humans, at room temperature,
stretching is accompanied by yawning nearly half
of the time (Provine et al. 1987a), with the
incidence of yawning predicting stretching, but
not vice versa. The yawn/stretch relationship in
budgerigars should be studied at lower ambient
temperatures before a similar relationship can
be dismissed.
This comparative evidence provides novel
insight into yawning as a thermoregulatory
mechanism, revealing that rising ambient
temperature promotes excessive yawning in
parakeets. This effect
could be tested further among an array of
species, including humans. Recent
interdisciplinary research has strengthened this
connection between yawning and thermoregulation
(Gallup & Gallup 2008). A growing body of
medical and physiological evidence implicates
instances of abnormal thermoregulation and heat
stress with symptoms of atypical yawning. For
instance, there is a link between the negative
symptoms of epilepsy, multiple sclerosis and
migraine headaches and increases in the ambient
temperature. More importantly, individuals
suffering from these disorders also yawn
excessively (Gallup & Gallup 2008).
Therefore, applications of this research range
from basic physiological understanding to
improved health and treatment of patients with
thermoregulatory dysfunction.
These findings have significant
ramifications regarding the way in which we
study yawning in humans and other species.
Yawning is widely associated with states of
fatigue, frequently occurring when an individual
wakes or gets ready for sleep (Provine et al.
1987a; Baenninger et al. 1996). Evidence shows
that sleep and thermoregulation appear to be
interrelated, with prolonged sleep deprivation
in rats producing an increase in deep brain
temperature (EverSon et al. 1994). Likewise, it
has been argued that core body temperature and
sleep vary inversely (Gilbert et al. 2004).
Following this rationale, subjective ratings of
sleepiness are correlated with increases in body
temperature (Krauchi et al. 2005). These results
may explain the empirical correlates of yawning
with transitional states of fatigue. Moreover,
the metabolic activity and locomotor changes
associated with awakening may disrupt thermal
homeostasis, and this underlying change in
thermal homeostasis may trigger the association
between yawning and awakening.
The thermoregulatory model complements and
may also help explain models highlighting the
association between yawning and other
transitional states, such as alertness and
arousal (Greco & Baenninger 1991; Walusinski
2006). Gallup & Gallup (2007) proposed that
the cooling component of yawning may actually
facilitate these processes (i.e. mental
efficiency and vigilance) by reinstating optimal
brain temperature. Moreover, this model has
implications for understanding contagious
yawning in humans as well as in nonhuman
primates (e.g. see Anderson et al. 2004; Paukner
& Anderson 2006), as the infectiousness of
the yawn may have evolved to facilitate group
vigilance.
In our study, it was unclear whether the
change in yawn frequency resulted from the
increase in ambient temperature or the change in
temperature (irrespective of direction). If
yawning serves to maintain optimal thermal
homeostasis, yawning frequency should increase
with temperature change. Thus, a decreasing
temperature condition may impose similar effects
on yawn frequency. Our results remain consistent
with the view that yawning is associated with
behavioural state change (Provine et al. 1987a;
Greco et al. 1993; Baenninger et al. 1996). In
addition, we propose that the difference in
yawning frequency among trial procedures may be
due in part to the control condition in the
first trial inadvertently lengthening the
initial acclimation period. As a result, the
second trial procedure (increase-high-control)
may have coupled an already mildly stressful
situation of environmental change with the
manipulation of ambient temperature, increasing
the likelihood of hyperthermia in the first two
thermal conditions (Cabanac & Guillemente
2001).
Because of the potential multifunctionality
of yawning across species (Baenninger 1987), we
suggest that further comparative research is
necessary to more completely understand the
relationship between yawning, ambient
temperature and other factors. For instance, the
thermoregulatory model suggests that there
should be differences in the potential adaptive
significance of yawning between endotherms and
ectotherms, as well as between endothermic
species selected to different degrees for
cooling abilities in challenging thermal
environments.
-Gallup
AC, Miller ML, Clark AB Yawning and
thermoregulation in budgerigars Melopsittacus
undulatus Animal Behav 2009;77(1):109-113
-Gallup AC,
Gallup G. Yawning as a brain cooling
mechanism: nasal breathing and forehead cooling
diminish the incidence of contagious yawning.
Evolutionary Psychology www.epjournal.net;
2007;5(1): 92-101
-Gallup
AC, Gallup Jr GG Venlafaxine-induced
excessive yawning: a thermoregulatory connection
Prog Neuro Psychopharmacol Biol Pyschiatry
2009;33(4):747
-Gallup AC, Gallup
GG Jr, Feo C. Yawning, sleep, and symptom
relief in patients with multiple sclerosis.
Sleep Med. 2010
-Gallup GG, Gallup
AC. Excessive yawning and thermoregulation:
two case histories of chronic, debilitating
bouts of yawning. Sleep Breath.
2010;14:157-159
-Miller
ML, Gallup AC et al. Handling stress
initially inhibits, but then potentiates yawning
in budgerigars (Melopsittacus undulatus). Animal
Behaviour. 2010;
-Platek
SM, SR Critton, et al Contagious yawning:
the role of self-awareness and mental state
attribution Cogn Brain Res 2003; 17; 2;
223-7
Argiolas,
A. & Melis, M. R. 1998. The
neuropharmacology of yawning. European Journal
of Pharmacology, 343,1-16.
Baenninger,
R. 1987. Some comparative aspects of yawning
in Betta splendens. Homo sapiens, Panthera leo,
and Papio sphinx.Journal of Comparative
Psychology, 101, 349-354.
Barbizet,
J. 1958. Yawning. Journal of Neurology,
Neurosurge,y and Psychiatry. 21, 203-209.
Bartholomew, G. A., Lasiewski, R. C. &
Crawford, U. C.Jr. 1968. Patterns of panting and
gular flutter in cormorants, pelicans, owls, and
doves. Condor. 70, 31-34.
Cabanac, A. J. & Guillemente, M. 2001.
Temperature and heart rate as stress indicators
of handled common eider. Physiology &
Behavior, 74, 475-479.
Everson, C. A., Smith, C. B. & Sokoloff,
L 1994. Effects of prolonged sleep deprivation
on local rates of cerebral energy metabolism in
freely moving rats. Journal of Neuroscience, 14,
6769-6778.
Falk, D. 1990. Brain evolution in Homo: the
"radiator" theory. Behavioral and Brain
Sciences, 13, 333-381.
Gallup, A. C.
& Gallup, G. G.Jr. 2007. Yawning as a
brain cooling mechanism: nasal breathing and
forehead cooling diminish the incidence of
contagious yawning. Evolutionary Psychology. 5,
92-101.
Gilbert, S. S., van den Heuvel, C. J.,
Ferguson, S. A. & Dawson, D. 2004.
Thermoregulation as a sleep signally system.
Sleep Medicine Reviews. 8. 81-93.
Greco, M.
& Baenninger, R. 1991. Effects of
yawning and related activities on skin
conductance and heart rate. Physiology &
Behavior, 50, 1067-1069.
Heusner A. P.
1946. Yawning and associated phenomena.
Physiological Reviews, 25, 156-168.
Hoffman, T. C. M., Walsberg G. E. &
DeNardo, D. F. 2007. Cloacal evaporation: an
important and previously undescribed mechanism
for avian thermoregulation. Journal of
Experimental Biology. 210, 741-749,
Iwanluk, A. N. & Nelson,J. E. 2002. Can
endocranial volume be used as an estimate of
brain size in birds? Canadian Journal of
Zoology. 80, 16-23.
Krauchi, K., Cajochen, C. &
Wirz.Justlce, A. 2005. Thermophysiological
aspects of the three-process-model of sleepiness
regulation. Clinical Journal of Sport Medicine,
24, 287-300.
Nijhuis, J. G. 2003, Fetal behavior.
Neurology of Aging, 24, 41-46.
Paukner A.
& Anderson,J. R. 2006. Video-induced
yawning in stumptail macaques (Macaca
arctoides). Biology Letters. 2, 36-38,
Provine, R.
R. 2005. Yawning. American Scientist, 93,
532-539.
Labo rató no
de Comportamento Animal, Universidade de
Brasilia, Brazil
Many attempts have been made to find
physiological stimuli responsible for evoking
spontaneous yawning. Smith (1999) enumerates
many functions that have been suggested for
yawning, ranging from increasing alertness and
increasing oxygen levels in the blood to
evacuation of infectious substances from
tonsils. But Smith (1999) also points out that
those propositions are deficient in presenting
empirical verification to validate them.
Therefore, yawning remains remarkable for being
a behaviour so widespread among vertebrates
whose physiological functions are still unclear
(Baenninger 1987). The contagiousness of the
yawn in certain social animals suggests it has
not only physiological functions, but maybe also
a social role (Deputte 1994). It can signal
tiredness and boredom, and is observable in
animals in stressful psychosocial situations
(Bell 1980; Maestripieri et al. 1992; Baenninger
1997).
The emergence of different theories to
explain the yawn is very positive, and should be
encouraged, for it contributes new elements and
points of view about this issue. Features of the
yawn unidentified up to the present moment can
be extremely valuable in better understanding
this behaviour, as it is not just a simple
reflex of short duration, but has a complex
spatio-temporal organization with facial,
respiratory and other components (Argiolas &
Melis 1998).
Therefore, the hypothesis of yawning as a
compensatory brain-cooling mechanism (Gallup
& Gallup 2007, 2008) has its
merits as a new approach to the study of
yawn. But this hypothesis still deserves more
evaluation, considering that anatomical and
physiological variables associated with
selective brain cooling were not considered. In
humans, for instance, Mekjavic et al. (2002)
demonstrated that inhalation of cold air is not
capable of influencing the temperature of the
brain. Based on this finding, airflow at ambient
temperature that occurs during a human yawn is
unlikely to be enough to cool the brain. But
Gallup et al.'s (2009) experiment used birds, in
which moist surfaces of nasal and buccal
cavities and of the eyes are known as
evaporative heatdissipating organs (Caputa et
al. 1998; Jessen 2001). In this case, yawning as
a brain-cooling mechanism is a hypothesis worth
investigating.
Gallup et al. (2009) interpreted the results
of their study with budgerigars as consistent
with the hypothesis that yawning is connected
with thermoregulation'. It is worth noting,
however, that Gallup et al. (2009) presented no
evidence that yawning maintained or altered the
birds' body temperature. Their results revealed,
in effect, that yawns occurred more frequently
when the ambient temperature was increased. The
allegation that yawning is an initial response
associated with thermal homeostasis also lacks
support from their data.
In animal behaviour studies, as Milinski
(1997) points out, the apparent correlation
between variables can lead the researcher to
draw unjustified conclusions. The results
obtained by Gallup et al.(2009) unmistakably
revealed that yawning occurred more frequently
with the increase in ambient temperature or with
changes in temperature (irrespective of
direction). But, unlike gular fluttering, whose
function as an evaporative-cooling mechanism has
already been established (Lasiewski &
Bartholomew 1966; Bartholomew et al. 1968),
their results did not provide evidence that
yawning is used to alter the temperature of any
part of the body, because no measure of bodily
temperature was taken. And, even the correlation
between variation in ambient temperature and the
occurrence of yawning may not be direct. It is
important to observe that other variables not
considered in their experiment may also have
influenced elicitation of yawning. To
investigate these variables, the experiment
should be designed so as to adopt a more
suitable experimental design.
Some studies suggest that yawns might have a
derived function of social communication (Smith
1999), acting as a signal in social species
(Rasa 1971). In some fish (Microspathodon
chrysurus) and monkeys (Macaca nigra), yawning
occurs in contexts of excitement and tension for
the performers (Rasa 1971; Hadidian 1980). These
findings are in agreement with Sauer & Sauer
(1967), who found that yawning induced
relaxation of tension in groups of excited South
African ostriches. The rapid temperature changes
in the experiment of Gallup et al. (2009) can be
considered a context of high tension for
budgerigars. It is not implausible, then, that
the budgerigars' yawns may have been produced as
a form of social signal among birds in the same
cage. Yawning is not contagious in budgerigars,
so when an individual yawns, it may not trigger
yawns in other individuals within the group, but
the yawning of that individual may have occurred
because the others were present.
According to Gallup (personal
communication), their budgerigars were tested as
a group to reduce stress of being removed from
the group. In fact, Soma & Hasegawa (2004)
demonstrated the budgerigars show decreased
aversion to new places when in groups. However
sound this precautionary measure may have been,
it introduced other variables that would have to
be dealt with. These variables may have no
influence at all in yawning, but that should be
experimentally established. To eliminate this
possibility, budgerigars should have been tested
individually.
Personality differences also could have
influenced the occurrence of yawning. Social
animals respond differently to a situation when
they are alone or with other individuals
(Harcourt et al. 2009). Testing the birds
individually would also have eliminated the
possible influence of this variable.
Yawning could also be credited to stress,
both from the new environment created by the
wooden box cover used in Gallup et al's (2009)
study and from the sudden increase in the
temperature of the surroundings. Gallup &
Gallup (2008) acknowledge and discuss the
connection between yawning and stress. The
experiment with budgerigars would have benefited
immensely from taking this variable into
account, because of its strong influence on the
animal's physiology. To reduce the stress of the
budgerigars in this type of environment,
researchers should use stable temperature
conditions, and for longer periods than in
Gallup et al. (2009), and they should avoid
recording yawns during rapid temperature
transitions. Considering a range of stable
environmental conditions, from below to above
the animal's body temperature (or using as many
conditions as the researcher deems necessary),
would allow researchers to demonstrate that
yawns are responses to environmental temperature
and not responses to variation in environmental
temperature.
In the experiment with budgerigars, yawning
may have been a consequence of social influence
or a consequence of stress from changes in
environmental temperature. It may have also been
a physiological mechanism to prevent brain
tissue from overheating. Because of the possible
influences of all these variables acting
together, it seems that the best course of
action would be to eliminate or at least to
reduce the effect of the variables that are not
under investigation.
After eliminating other intervening
variables, a strong indication that yawning is
related to brain cooling would be the
satisfaction of the model's predictions: (1) the
frequency of yawns rising as ambient temperature
approaches body temperature and (2) yawning not
occurring when ambient temperature reaches or
exceeds body temperature. But even then, the
conclusion that yawning is a compensatory
cooling mechanism would need measures of bodily
temperatures. Differences in temperature between
the brain and the rest of the body are not
uncommon in birds (Burgoon et al. 1987).
Therefore, separate measures of brain and body
temperature should be taken before and after
yawns in each experimental temperature
condition, taking into account circadian brain
and body temperature variations. The maintenance
or the cooling of brain temperature due to yawns
would be a direct evidence of the temperature
regulatory function of yawning, allowing the
researchers to come to an unquestionable
conclusion.
As an effort to elucidate the function of
yawning, the thermoregulatory hypothesis has a
strong appeal, and provides an invaluable
contribution to research. But the limitations of
experiments should lead to more cautious
conclusions regarding the proximate causes of
yawning, if we want to identify its adaptive or
biological significance.
