-
- Andrew
C. Gallup. Yawning and the thermoregulatory
hypothesis
-
- Brain
Temperature & Autonomic Nervous System for
the study of relaxation
-
- In mammals, yawning is associated with
social and physiological stress, as well as
thermoregulation, but little is known about why
yawning occurs in stressful contexts or how it
is integrated with natural stressors. To
investigate the stress sensitivity of yawning in
birds, we exposed budgerigars (Melopsittacus
undulatus) to a handling stressor that simulated
a predatory encounter. Each bird was captured,
gently held for 4 mm, and then released and
videotaped for I h (experimental). On a separate
day (±24 h), the undisturbed animal was
videotaped for I h (control). The relationship
between handling-induced yawning and body
temperature was assessed in a separate
experiment, in which the underwing temperatures
of the same birds were measured at I min
intervals during a 4 min holding period. After
handling stress, yawning frequency was initially
suppressed, then sharply increased within 20
min. Underwing temperature increased during
handling, and individuals' final temperatures at
minute 4 were negatively correlated with their
latencies to yawn after handling. Thus,
stress-induced hyperthermia may be responsible
for associations between yawns and stress. These
results indicate that yawning may offer a
sensitive, noninvasive measure of stress in
birds.
-
- Yawning is phylogenetically old and
ubiquitous among vertebrates (Baenninger 1987),
but little is known about its physiological
function. It is commonly thought to equilibrate
oxygen and carbon dioxide imbalances in the
blood, but there is no experimental support for
this hypothesis (Provine et al. 1987b). On the
other hand, there is strong evidence to suggest
that yawning is associated with social
(Baenninger 1997) and physiological stress
(Gallup & Gallup 2008), including thermal
challenges in birds (Gallup et al. 2009). In
line with this, yawning may increase brain
arousal, and this arousing function may explain
the behaviour's association with stress
(Baenninger 1997; but see Guggisberg et al.
2007).
-
- Yawning, or behaviour resembling yawning, is
associated with stressful events in several
nonhuman primates, as well as other mammals. In
crested black macaques, Macaca nigra, yawning
occurs during intense agonistic interactions and
other hostile social situations (Hadidian 1980).
Troisi et al. (1990) interpreted yawning by
subordinates in this context as a response to
stress, and those by dominants as a threat
display. In these macaques, yawning also follows
abrupt, startling disturbances, such as thunder,
that may induce low-level, acute stress
(Hadidian 1980). Likewise, in greycheeked
mangabeys, Cercocebus albigena, yawning occurs
in close temporal proximity to alarm calling in
the presence of predators (Deputte 1994). Among
primates, similarities between threat displays
and yawning confounds interpretations of yawning
(Vick & Paukner 2010), making it informative
to study yawning in species without open-mouth
threat displays. Yawning is also associated with
physical stress in laboratory rats (Rattus
norvegicus) in that foot-shock strongly
increases yawning (Moyaho & Valencia 2002).
Taken together, there appears to be a close
relationship between stress and yawning in a
range of mammals.
-
- In budgerigars (Melopsittacus undulatus),
the only birds in which yawning has been
experimentally studied, yawning occurs more
frequently as ambient temperature increases
(22-34'C) towards body temperature (Gallup et
al. 2009). In addition, yawning in budgerigars
is significantly correlated with other avian
thermoregulatory behaviours (e.g. panting, wing
venting; Gallup et al. 2010), suggesting that
yawning is triggered by the need to decrease
elevated body and/or brain temperatures. If
temperature regulation is one general function
of yawning in homeotherms, stress-associated
yawning may be a response to increases in body
temperature induced by external stressors (i.e.
stress-induced hyperthermia; reviewed in Olivier
et al. 2003). For instance, when common eiders,
Somateria mollissima, were handled, their body
and skin temperatures increased within 4 min
following the start of the trial (Cabanac &
Guillemette 2001). Stress-induced hyperthermia
may produce the intimate relationship between
stress, yawning and thermoregulation.
-
- Although yawning is linked to stress across
diverse contexts in mammalian species, no
experimental studies in birds have evaluated the
effect of stress on yawning. Because yawning is
an overt, distinguishable behaviour,
characterizing its relation to stress may make
it a suitable behavioural measure of stress. In
this study, we investigated the stress-yawn
relationship in budgerigars by inducing acute
stress through handling. The degree of handling
(i.e. gentle restraint in a gloved hand
following a quick capture) was comparable to
that experienced by birds during routine
measurements in the laboratory and in the field
(e.g. Cabanac & Guillemette 2001; Carere
& van Oers 2004; Fucikova et al. 2009). We
recorded each bird's yawning and stretching
frequencies both following brief handling
(experimental) and during a similar period with
no preceding disturbance (control). We
hypothesized that yawning would increase in
frequency after the handling session relative to
the control period, but would show no temporal
pattern during the control period. Because of
the temporal association between yawning and
stretching among humans and rodents (reviewed
in: Baenninger 1997; Provine et al. 1987a), we
also recorded stretching. Yawning and stretching
in budgerigars are temporally associated in
nonexperimental settings (M. L Miller, S. M.
Vicario & A. B. Clark, unpublished data).
Interestingly, this temporal relationship is
decoupled as temperatures increase, presumably
when yawns serve a thermoregulatory function
that stretching does not (Gallup et al. 2010).
Thus, if stress-induced yawning is specifically
related to body temperature and brain arousal,
stretching should remain unaffected after a
simulated predatory encounter.
-
-
- RESULTS
-
- Yawning and Stretching
-
- We observed 63 yawns in total (42 by the six
males, 21 by the four females) during 20 h of
observation (10 trials, 2 h/bird). To assess
whether the birds' latency to yawn differed
between control and experimental conditions, the
model included trial condition, trial order and
the interaction between these two factors. For
this full model, there was no difference in
yawning latency between trials categorized by
order (F1,8 = 0.62, P = 0.45, partial Tj2 =
0.07) and no interaction between trial condition
and trial order (F1,8 = 1.50, P = 0.26, partial
j2 = 0.16). After removing trial order from the
model, the latency to the first yawn was
significantly later in the experimental
condition than in the control condition (1322 +
144 s versus 787 ± 189 s; F1,9 = 5.96, P=
0.04, partial 12 = 0.40).
-
- Of the 63 yawns, 34 (3.4 ± 0.54/bird)
occurred during the experimental condition and
29 (2.9 + 0.31/bird) occurred during the control
condition. To investigate whether there were
differences in yawning frequencies between trial
conditions and across time intervals, the model
included trial condition, time interval, trial
order and all interactions between these
factors. In the full model, there was no
difference in total yawning frequencies between
the two trial conditions (F1,8 = 1.02, P = 0.34,
partial = 0.11) or between the three 20 min
intervals (F2,16 = 0.65, P = 0.54, partial i12 =
0.08). There was also no difference in yawning
frequencies between trial orders (F1,8 = 0.02, P
= 0.90, partial j2 <0.01), and no
interactions between trial order and condition
(F1,8 = 2.00, P = 0.20, partial t2 = 0.20),
between trial order and time interval (F2,16 =
0.42, P = 0.67, partial = 0.05), or between
trial order, time interval and trial condition
(F216 = 0.64, P = 0.54, partial i12 = 0.07).
After removing trial order from the model, there
was a significant interaction between time
interval and trial condition (F2,18 = 3.88, P=
0.04, partial Ti 2 = 0.30; Fig. lb). Paired
comparisons between the trial conditions within
each time interval indicated that yawning
frequency was (1) lower in the experimental
condition than in the control condition during
the first 20 min (t9 = 2.23, P = 0.05), (2)
greater in the experimental condition than in
the control condition during the second 20 min
interval (t9 = -2.45, P = 0.04) and (3) not
different during the final 20 min interval (t9 =
-1.05, P = 0.32).
-
- We observed a total of 69 stretches (41 by
males, 28 by females) during the 20h of
observation. The same set of analyses was run to
investigate whether latency to stretch differed
between control and experimental conditions.
Unlike yawning, there was no difference in
latencies to the first stretch between the
experimental and control conditions (1913 ±
307 s versus 1931 +403 s; F1,8 = 0.00, P = 0.97,
partial t2 = 0.00). In addition, there was no
difference in stretching latency between trial
orders (F1,8 = 0.31, P = 0.59, partial = 0.04),
and no interaction between trial condition and
trial order (F1,8 = 1.34, P = 0.28, partial j2 =
0.14).
-
- To investigate whether stretching
frequencies differed between conditions and
across intervals, the full model included trial
condition, time interval, trial order and
interactions between these factors. For this
model, stretching frequencies did not differ
between trial conditions (F1,8 = 0.03, P=0.86,
partial <0.01) or trial orders (F1,8 = 0.79,
P = 0.40, partial = 0.09), and there was no
interaction between trial order and condition
(F1,8 = 0.17, P = 0.69, partial t2 = 0.02), or
between trial order and time interval (F2,16 =
0.45, P = 0.64, partial 2 = 0.05). Unlike
yawning, there was no significant interaction
between time interval and trial condition (F2,16
= 2.05, P = 0.16, partial 2 = 0.20). There was
also no interaction between trial condition,
time interval and trial order (F2,16 = 0.35, P =
0.71, partial 2 = 0.04). When removing trial
order from the model, stretching differed
significantly across the three time intervals
(F2,18 = 4.31, P = 0.03, partial î2 =
0.32). Post hoc corrections showed no
significant pairwise comparisons (all Ps >
0.05).
-
- Temperature Changes with
Handling
-
- During the temperature assessment session,
budgerigar body temperature steadily increased
during the handling session (Fig. 2a). Average
temperature differed across the four I min
intervals (Friedman's test: = 18.04, P
<0.01). All pairwise comparisons between time
intervals showed a significant increase from one
interval to the next (Ps < 0.05), except
between the second and third minute, which was
nearly significant (P = 0.06). A budgerigar's
previously recorded behaviour (yawns or
stretches) was then correlated with this
individual's underwing temperature. Underwing
temperatures at the end of the handling sessions
were strongly and negatively correlated with the
latency to first yawn (Kendall's tau
correlation: b = -0.62, P = 0.03; Fig. 2b). This
indicates that birds with higher body
temperatures following handling yawned sooner
during the experimental condition. Underwing
temperatures at the fourth minute were not
correlated with an individual's total yawn
frequency (b = 0.15, P = 0.61) or with the
number of yawns during any one of the three 20
min intervals (all Ps > 0.05). Increases in
temperature (i.e. difference between the final
and first minute) were not correlated with (1)
yawn latency (b = -0.18, P = 0.53), (2) total
yawning frequencies (b = 0.42, P = 0.16) or (3)
yawning frequencies across each 20 min interval
(all Ps > 0.05). Unlike yawning, stretching
by individuals was not correlated with either
temperature at the fourth minute or change in
temperature (Ps > 0.05).
-
-
- DISCUSSION
-
- These results illustrate that yawning in
budgerigars is affected by handling stress.
Yawns were initially suppressed, but then
increased in frequency after 20 min. As handling
may simulate escape from a predator, initially
suppressing yawns may adaptively reduce
attention-getting movements and/or reduce
conflict with other antipredatory behaviours.
Because acute stress increases body temperature
(e.g. Cabanac & Guillemette 2001), a spike
in yawning after 20 min is adaptive, since
research suggests yawning is a
thermal-stabilizing mechanism that decreases
brain and/or body temperature (e.g. Gallup &
Gallup 2007, in press: Gallup et al. 2009). This
interpretation is supported by the strong
negative correlation between individuals' body
temperatures after handling and their latencies
to first yawn (see Fig. 2b), indicating that
higher body temperatures may trigger birds to
yawn sooner. In contrast to yawning, stretching
did not change in frequency after the stressor.
Stretching frequencies were also unrelated to
the individuals' body temperatures, suggesting
that stretching lacks a thermoregulatory role
(Gallup et al. 2009).
-
- These results are consistent with previous
findings in other species that demonstrate a
temporal association between yawning and stress.
For instance, in South African ostriches,
Struthio camelus australis, yawning did not
occur during intense activity, but did occur
when startling stimuli were recognized as
innocuous, presumably sometime after the
stressor (Sauer & Sauer 1967). When rats
were exposed to a novel environment, yawning
gradually increased, peaking after 30 min
(Moyaho & Valencia 2002). Similarly, when
rats were foot-shocked at fixed, 10 min
intervals, yawning was initially low, but then
gradually increased and peaked by 40 min. On the
other hand, increases in yawning were less
pronounced when rats were foot-shocked at random
intervals (Moyaho & Valencia 2002). This is
consistent with the budgerigar data, because it
shows that yawning occurs during a recovery
period following a stressor: when foot-shocked
at known intervals, rat yawning dramatically
increases, but when randomly footshocked,
yawning does not increase as dramatically,
presumably because the stress state persists.
These data suggest that yawning is related to
the recovery period following a stressor and may
be an adaptive response that increases vigilance
as the environment becomes more predictable (see
Greco et al. 1993).
-
- The appearance of yawns during the second 20
min interval is in accord with the view that
yawning is a thermoregulatory behaviour in
budgerigars (Gallup et al. 2009, 2010). The
increased yawns observed during the second 20
min interval may be explained by temperature
increases that follow handling stress (Olivier
et al. 2003). Similar to the effect of handling
on eider ducks (Cabanac & Guillemette 2001),
handling increased underwing temperature of
budgerigars. This increase in temperature was
substantial and rapid, approximating 2 °C
within 3-4 min. Cabanac & Guillemette (2001)
demonstrated that duck temperature peaked by 10
min of handling, and hyperthermia was maintained
for at least 30 min. Therefore, if the time
course of body temperature is similar in
budgerigars, the spike in yawning during the
second interval may have been a compensatory
mechanism to reduce brain and/or body
temperatures following the simulated capture and
escape. Moreover, latency to yawn was negatively
correlated with skin temperature measured at the
fourth minute. This indicates that birds that
responded to stress with greater temperature
increases needed to yawn sooner, but not at
higher frequencies. In short, increases in
metabolic activity following stress inevitably
cause increases in body temperature. Whether
this increase in temperature is adaptive or a
metabolic by-product is unclear: however,
yawning may provide a means to regain thermal
homeostasis after a stressful event.
-
- Since yawning is an easily distinguishable
behaviour, these results suggest that measuring
yawns may provide a suitable method to detect
and qualitatively measure stress noninvasively.
It is difficult to measure stress without
disturbing an animal, making accurate assessment
of stress difficult. For instance, in laboratory
settings, collecting blood to measure
corticosterone (CORT) levels inherently produces
an emotional response, thereby affecting plasma
concentrations of stress hormones, such as CORT
(Thanos et al. 2008). To appreciate the
application of yawning as a technique to measure
stress, it is important to note the sensitivity
of this relationship. Although the birds used in
this experiment were accustomed to daily human
contact over a period of many years,
-
- the flock continues to respond to human
entry with increased movement and vocalization
(personal observation). In a pilot study,
entering the room to turn on a camera was
sufficient to inhibit yawns during the first 20
min of the control condition (M. L Miller, J. A.
Cusick, D. M. Sutton, A. R. Vogel, A. C. Gallup
& A. B. Clark, unpublished data), which is
why recordings were remotely started in the
control trials. This is not unreasonable, as the
heart rates of laboratory mice increase when a
technician enters the colony and this effect
persists for at least 2 weeks after the first
exposure (Kramer et al. 2004). Monitoring yawns
may provide a sensitive measure of individual
responsiveness to acute stressors.
-
- In summary, these results illustrate a
relationship between yawning, stress and
thermoregulation in birds. This report provides
critical insight into the association between
yawning and arousal. It is the first to show
that yawning is delayed after a simulated
predator's attack and also replicates previous
studies, showing that yawns are strongly
associated with changing body temperature. These
findings also suggest that yawning may provide a
noninvasive measurement of stress in field and
laboratory settings. Follow up studies should
measure other physiological parameters related
to stress (e.g. plasma CORT), and then correlate
these with yawning.
-
-
- References
-
- 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.
-
- Baenninger,
R. 1997. On yawning and its functions.
Psychonomic Bulletin and Review, 4,198-207.
-
- Cabanac, A.J. & Guillemette, M. 2001.
Temperature and heart rate as stress indicators
of handled common eider. Physiology &
Behavior, 74, 475-479.
-
- Carere, C. & van Oers, K. 2004. Shy and
bold great tits (Parus major): body temperature
and breath rate in response to handling stress.
Physiology & Behavior, 82, 905-912.
-
- Deputte, B.
L 1994. Ethological study of yawning in
primates. 1. Quantitativeanalysis and study of
causation in 2 species of Old-World monkeys
(Cercocebus albigena and Macacafascicularis).
Ethology, 98, 221-245.
-
- Fucikova, E., Drent, P. J., Smits, N. &
van Oers, K. 2009. Handling stress as a
measurement of personality in great tit
nestlings (Parus major). Ethology, 115,
366-374.
-
- 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.
-
- Gallup, A.
C. & Gallup, G. G., jr. 2008. Yawning
and thermoregulation. Physiology & Behavior,
95,10-16.
-
- Gallup, A.
C. Miller, M. L & Clark, A. B. 2009.
Yawning and thermoregulation in budgerigars,
Melopsittacus undulatus. Animal Behaviour,
77,109-113.
-
- Gallup,
A. C., Miller, M. L & Clark, A. B. 2010.
The direction and range of ambient temperature
influences yawning in budgerigars (Melopsittacus
undulatus). Journal of Comparative Psychology,
124,133-138.
-
- Gallup, G. G., Jr
& Gallup, A. C. 2010. Excessive yawning
and thermoregulation: two case histories of
chronic, debilitating bouts of yawning. Sleep
and Breathing.
-
- Greco,
M., Baenninger, R. & Govern, J. 1993. On
the context of yawning: when, where, and why?
Psychological Record, 43,175-183.
-
- Guggisberg,
A. G., Mathis, J., Herrmann, U. S. & Hess,
C. W. 2007. The functional relationship
between yawning and vigilance. Behavioural Brain
Research, 179,159-166.
-
- Hadidian, J.
1980. Yawning in an Old World monkey, Macaca
nigra (Primates: Cercopithecidae). Behaviour,
75, 133-147.
-
- Kramer, K., van de Weerd, H., Mulder, A.
& Van Heijningen, C. 2004. Effect of
conditioning on the increase of heart rate and
body temperature provoked by handling in the
mouse. Alternatives to Laboratory Animals, 32,
177-181.
-
- Moyaho,
A. & Valencia, J. 2002. Grooming and
yawning trace adjustment to unfamiliar
environments in laboratory Sprague-Dawley rats
(RatWs norvegicus). Journal of Comparative
Psychology, 116, 263-269.
-
- Olivier, B., Zethof, T., Pattij, T., van
Boogaert, M., van Oorschot, R., Leahy, C,
Oosting, It, Bouwknecht, A., Veening, J., van
der Gugten, J. & Groenink, L 2003.
Stress-induced hyperthermia and anxiety:
pharmacological validation. European Journal of
Pharmacology, 463,117-132.
-
- Provine, R.
R., Hamernik, H. B. & Curchack, B. C.
1987a. Yawning: relation to sleeping and
stretching in humans. Ethology, 76,153-160.
-
- Provine, R.
R., Tate, B. C. & Geldmacher, L L 1987b.
Yawning: no effect of 3-5% CO2. 100% 02, and
exercise. Behavioral and Neural Biology, 48,
382-393.
-
- Sauer, E. G.
F. & Sauer, E. M. 1967. Yawning and
other maintenance activities in the South
African ostrich. Auk, 84, 571-587.
-
- Thanos, P. K., Cavigelli, S. A.,
Michaelides, M., Olvet, D. W., Patel, U., Diep,
M. N. & Volkow, N. D. 2008. A non-invasive
method for detecting the metabolic stress
response in rodents: characterization and
disruption of the circadian corticosterone
rhythm. Physiological Research, 58,
219-228.
-
- Troisi, A.,
Aurelli, F., Schino, G., Rinaldi, E & De
Angelis, N. 1990. The influence of age, sex,
and rank on yawning behavior in two species of
macaques (Macaca fascicularis and M. fuscata).
Ethology, 86, 303-310.
-
- Vick,
S.-J. & Paukner, A. 2010. Variation and
context of yawns in captive chimpanzees (Pan
troglodytes). American Journal of Primatology,
72, 262-269.
-
-
Andrew
C. Gallup. Yawning and the thermoregulatory
hypothesis
|
