Guggisberg et
al. (2010) reviewed the evidence for the
origin and function of yawning, and conclude
that theories describing a physiological role
lack support. Instead, they argue research
supports the notion that yawning has a
communicative function. Contrary to the authors'
claim that the social/communication hypothesis
has the "best experimental evidence", there is
in fact no definitive experimental support for
the predictions of this model. Furthermore, the
authors claim to take an evolutionary
perspective, but sufficient examples across the
comparative (non-primate) literature are
missing, and they fail to acknowledge phylogenic
history. Due to the ubiquity of this behavior
across vertebrates, and the regularity of its
occurrence in a number of different
physiological states and social contexts, it is
likely that instead of serving one purpose,
yawning is multifunctional across a number of
species. The most parsimonious explanation for
the origin of yawning suggests that any social
value is a derived feature, while the primitive
feature or function is physiological. The
current paper addresses these concerns, and
identifies a number of other weaknesses in the
social/communication hypothesis as a global
explanation for the origin and function of
yawning.
1. Introduction
This paper discusses the following
shortcomings associated with the review on
yawning by Guggisberg et al. (2010): the lack of
a comparative analysis, the inability to
incorporate instances of excessive or atypical
yawning, an inadequate review of the existing
physiological hypotheses, problems associated
with the description and predictions of the
social / communication hypothesis, the mismatch
of standards and criteria used to evaluate the
various hypotheses, the lack of experimental
support for the social/communication hypothesis,
problems associated with this hypothesis as an
explanation of the phylogenic origin and
function of yawning, and the lack of a
distinction between primitive and derived
features. One major problem with the Guggisberg
et al. (2010) review is their stance on the
origin of yawning as social. Any social or
communicative value of yawning among humans and
nonhuman primates is likely a derived feature,
while the underlying primitive feature or
function is physiological. Due to the ubiquity
of this behavior across vertebrates, and the
regularity of its occurrence in a number of
different physiological states and social
contexts, it is probably the case that instead
of serving one purpose, yawning is
multifunctional across a number of species.
2. Comparative aspects of
yawning
Guggisberg et al. (2010) begin their review
on the origin and functional hypotheses of
yawning by discussing the ubiquity of yawning,
or yawn-like behaviors across vertebrate
classes. Classifying yawning versus open-mouth
gestures or jaw stretching is often difficult,
and no one has systematically defined a complete
action repertoire of yawning that would
accommodate all observed cases. Comparatively,
it is customary to assume that yawning consists
of an extended gaping of the mouth, followed by
a more rapid closure (Baenninger, 1987; Provine,
1986). Among birds and mammals, however, yawning
is accompanied by a deep inhalation of air with
eye closure, followed by a shorter expiration.
Although fish yawns have been described, they
are fundamentally different from those of
terrestrial vertebrates, and have been
characterized to be separate from yawning in
birds and mammals (Rasa, 1971). Other reports on
Siamese fighting fish (Betta spiendens) support
this view, reflecting aggressive social displays
more than spontaneous yawning events
(Baenninger, 1987). So-called yawning in snakes
may also be different from other accounts, as it
is not a spontaneous action, but a motor pattern
following feeding strikes and swallowing that
functions in the stretching and realignment of
the jaw (Dullemeijer and Povel, 1972).
Baenninger (1987) argues that although
amphibians and reptiles open their mouth widely
on occasion, this does not necessarily represent
yawning. Guggisberg et al. (2010) acknowledge
this controversy surrounding the comparative
aspects of yawning initially, but then fail to
further describe the phylogeny of yawning in
relation to various hypotheses throughout the
review.
3. Atypical yawning and medical
conditions
Considering the close association between
atypical yawning and a number of diseases or
medical conditions in humans (Daquin et al.,
2001; Gallup and Gallup, 2008; Walusinski,
2009), it is clear that the study of yawning and
its functions has important implications for
medical research and diagnoses. Likewise, only
after we have achieved a thorough understanding
of the pharmacology and physiology of yawning
can interpretations be made regarding
functionality. One underlying problem with the
review by Guggisberg et al. (2010) is that they
sidestep this literature to focus on normal
yawning. Understanding the pathology of a
behavior in its association with medical
conditions or drug-use can provide intriguing
insights regarding its triggers and functions.
If the authors want to evaluate the various
functional hypotheses in an attempt to account
for all instances of yawning, then it is
imperative that excessive and pathological
yawning be discussed in detail. The authors
briefly mention important work on the
neuropharmacology of yawning (e.g., Argiolas and
Melis, 1998), but fail to integrate it into the
later sections of their review.
4. Review of physiological
hypotheses
Another shortcoming of the Guggisberg et al.
(2010) paper is that it does not sufficiently
review the literature on existing physiological
theories. For instance, the thermoregulatory
hypothesis has far more support than they
acknowledge (reviewed by Gallup et al., 2010).
In addition, recent research has explored the
relationship between brain temperature and
yawning by implanting thermocoupled probes in
the prelimbic cortex of rats (Rattus norvegicus)
to measure changes in brain temperature before,
during and after yawning (Shoup-Knox et al.,
2010). Results show that yawning was preceded in
all instances by rapid increases in brain
temperature (0.12°C), with corresponding
decreases in temperature occurring immediately
after each yawn (Shoup-Knox et al., 2010).
Recent comparative research is consistent with
this, showing a strong negative correlation
between body temperatures and yawn latency in
handling-stressed budgerigars (Melopsittacus
undulatus) (Miller et al., 2010). Although a
detailed review of the association between
yawning and thermoregulation (Gallup and Gallup,
2008) is mentioned in passing, it is otherwise
largely overlooked.
Guggisberg et al. (2010) also inadequately
describe how yawning could provide cerebral
cooling, not taking into account other effects
described by this model (i.e., increased
circulation). In addition, Guggisberg et al.
(2010) attempt to discredit the results of
Gallup and Gallup (2007), which showed that
methods of behavioral cooling (i.e., nasal
breathing and forehead cooling) diminish
yawning, by arguing that it is impossible to
differentiate the effects of temperature and
sleepiness. However, this argument does not take
into account the fact that nasal breathing (in
the absence of mouth closure) reduces yawning
while breathing orally leaves yawning rates
unaffected (Study 1, Gallup and Gallup, 2007).
Nasal breathing, for instance, does not inhibit
sleepiness, and therefore the temperature/sleep
confound described by Guggisberg et al. (2010)
is untenable. According to the reasoning
presented by Guggisberg et al. (2010), the
association between warm temperature and sleep
would predict that yawning should increase when
ambient temperatures are held high. Contrary to
these predictions, experimental research on
budgerigars shows that yawning is reduced at
constant high temperatures, and this is not an
artifact of increased sleep or rest (Gallup et
al., 2009). Therefore, despite the view of
Guggisberg et al. (2010), this research directly
distinguishes between temperature and sleep
effects, and these results are in accord with
the predictions of the thermal window hypothesis
of the thermoregulatory model.
The authors also fail to provide a detailed
description of the vast comparative evidence in
support of the arousal hypothesis reviewed
originally by Baenninger (1997). To date, this
review provides the most comprehensive account
of the comparative research on yawning. Instead
of describing the support for this hypothesis,
the majority of the section on arousal is used
to argue against it (Guggisberg et al., 2010).
Central to this argument is the discussion of an
earlier study by members of the same laboratory
investigating changes in electroencephalogram
(EEG) associated with yawning (Guggisberg et
al., 2007), but the authors fail to describe the
shortcomings of their report. For instance, the
participants used in this study were patients
who suffered from excessive daytime sleepiness
or non-restorative sleep (Guggisberg et al.,
2007). Considering the close connection between
yawning and sleep, it is possible that the
results are unique to this subset of
individuals, and that similar results would not
be found in a normal, and healthy population. In
addition, these changes in EEG were only
measured during a 10-s window (Guggisberg et
al., 2007). Therefore, it is entirely possible
that yawning may increase arousal over a longer
time, and in ways that are not detected by EEG.
An extension and replication of this study is
needed.
5. Problems with the social/communication
hypothesis
As stated, the social/communication
hypothesis is extremely vague. According to this
hypothesis, yawning is a form of nonverbal
communication that signals one's current mental-
or physiological-state. However, descriptions of
the states being communicated are highly
ambiguous. Phrases such as "certain body
states", "mildly to moderately unpleasant", and
"unpleasant but not immediately threatening
states" are vague and imprecise, and the use of
these descriptions demonstrates the scarce
research investigating predictions of the
social/communication hypothesis. For instance,
Guggisberg et al. (2010) speculate that the
reason yawning is suppressed among humans when
being observed (Baenninger and Greco, 1991;
Provine, 2005) is because individuals find the
situation "inappropriate", and that they "hide"
their yawns. But the authors neither provide any
solid evidence for this assertion, nor do they
go on to describe what distinguishes yawns that
are felt inappropriate from those that are not.
There is simply not enough scientific support
for the social/communication hypothesis, and
attempts to construct post hoc arguments in its
favor can easily be dismissed.
Earlier in the review the authors claimed
that if 'the act of yawning does not induce more
autonomic changes than the ones that already
occur hundreds of times throughout the day due
to simple breathing or moving" then it does not
make sense to propose a circulatory function to
yawning when more efficient or effective
behaviors could be employed (e.g., other body
movements). However, the authors fail to
acknowledge that this same argument can apply to
the signaling value of yawning according to the
social/communication hypothesis. For instance,
signals must be detected and distinguished from
background noise, and it is not hard to see how
a more transparent signal would achieve
social/communicative functions with much less
interference. On the other hand, if yawning does
produce a meaningful signal that is easily
interpreted by others (for which there is no
available evidence), then the efficiency or
strength of the signal is not important.
Similarly, since yawning produces significant
changes in circulation (increased heart rate,
vasodilation, and cerebral blood flow) (Greco
and Baenninger, 1991; Guggisberg et al., 2007;
Schroth and Klose, 1992) then it too produces a
physiological outcome. Many of the arguments
made by Guggisberg et al. (2010) in an attempt
to undermine the physiological hypotheses can
also be used to raise serious questions about
the social/communication hypothesis.
Problems continue to arise with the logic
used for the argument that the origin and
function of yawning is social or communicative.
In the most primitive sense, a signal is a
behavior or structure that has evolved to alter
the behavior of another organism (Maynard Smith
and Harper, 2003). Thus, signals need to be
interpreted and recognizable by others, and
mechanisms need to be in place to receive these
signals. Comparative yawning research in support
of these predictions is lacking. There is no
comparative evidence that group members orient
towards or receive this "signal." Similarly,
research on the response of others (i.e.,
altered behavior) to yawning is absent. Although
threat yawns have been described in some
primates (Hinde and Tinbergen, 1958; Redican,
1975; Tinbergen, 1952), the notion that canine
displays during yawns are aggressive signals is
a highly controversial topic (Deputte, 1994;
Dobson, 2010). Furthermore, "threat yawns" are
fundamentally different from spontaneous yawns
in that they are assumed to only occur in
specific social contexts. Therefore, there is
reason to believe such yawns have different
physiological consequences and triggers than
spontaneous yawns, and they may not be useful
for the purposes of identifying the origin of
this behavior.
The social/communication hypothesis falls
short in a number of other important and basic
areas. For one, signal evolution predicts that a
signal will only evolve if the sender transmits
information that is reliable to the receiver.
Since yawning occurs with regularity under a
variety of states such as fatigue, boredom, and
stress, any signal that is displayed remains
nonspecific. That is, unless the information
transferred from yawning is context-dependent,
meaning that receivers can reliably discern the
signal of yawns from individuals in different
environments. There is no research exploring
this question, and in fact results from of an
in-depth study of yawning in Macaca fascicularis
and Lophocebus albigena suggest that aside from
during resting, yawning is not associated with
any particular social context (Deputte, 1994).
Therefore, it is unlikely that yawning outside
of resting could reliably transmit relevant
information regarding a particular situation.
The authors
fail to sufficiently address this seemingly
obvious issue. Instead, they state that all
triggers for yawning "have in common that they
are mildly to moderately unpleasant while not
representing an immediate threat." In addition,
in order for a signal to remain reliable, the
receiver must benefit from having information
about that attribute (Searcy and Nowicki, 2005).
There is no comparative evidence suggesting that
individuals who simply witness others yawning
receive any benefit. On the other hand, if the
physiological consequences of yawning provide
some homeostatic function (i.e., arousal,
thermoregulation), then yawning in response to
perceiving another yawn (contagious yawning) may
provide a benefit to the receiver (Gallup and
Gallup, 2007). But of course this would only
hold true for species that have a capacity for
contagious yawning (an issue discussed below).
Secondly, in the description of yawning as a
social signal, Guggisberg et al. (2010) do not
mention honest versus dishonest signaling, or
the prospect of any cost associated with
yawning. Well-established theories regarding
signal evolution have posited that a signal is
likely to entail some cost to the sender if it
is to remain honest (Zahavi, 1975; Grafen, 1990;
Andersson, 1994). For instance, vulnerability or
production costs may be associated with yawning
during certain circumstances, as a yawn may draw
unwanted attention to the sender, momentarily
decrease vigilance, or conflict with
antipredatory behaviors (Miller et al., 2010).
Although alternative explanations exist for
understanding the evolution of signal
reliability (Searcy and Nowicki, 2005), the fact
that the authors do not discuss this issue, but
proclaim yawning as primarily social, is yet
another shortcoming.
In addition, the social/communication
hypothesis is not consistent with some of the
empirical research on yawning and stress.
Guggisberg et al. (2010) argue that yawns are
triggered by a stressor to communicate an
individual's mental- or physiological-state, but
recent research on budgerigars has shown that
yawning is initially suppressed following the
application of stressors (Miller et al., 2010).
This response more accurately describes a
recovery behavior following the stressor, and
not a communicative signal of the stressor
itself.
6. Support for the social/communication
hypothesis
The majority of the evidence in support for
the social/communication hypothesis hinges on
the contagious nature of this behavior in humans
(Provine, 1986, l989a,b; Platek et al., 2003)
and a select few non-human primates (Anderson et
al., 2004; Paukner and Anderson, 2006; Palagi et
al., 2009), yet this is considered "good"
evidence for the origin and function of yawning
to be communication. Contagious yawning has also
been reported in dogs (Canis familiaris) in
response to human yawns (Joy-Mascheroni et al.,
2008), but a further study shows that dogs do
not yawn contagiously to conspecifics (Harr et
al., 2009). The fact that yawning is triggered
during drowsiness, boredom and stress is also
considered "good" support for the
social/communication hypothesis (Guggisberg et
al., 2010). It is further stated that the
social/communication hypothesis is the most
parsimonious of all models, but this is not the
only model that can account for social effects
of yawning such as its contagiousness or the
different physiological states and social
contexts that can trigger it. In fact, arousal,
state change, and thermoregulatory models do
just as well, and it is easy to argue that
physiological models are more parsimonious. For
instance, considering that yawns are triggered
during low states of vigilance, and produce
changes in underlying physiology (e.g.,
increased facial and cerebral blood flow), the
spreading of this behavior to nearby
conspecifics could enhance group vigilance
(Gallup and Gallup, 2007). This interpretation
actually provides more explanatory value than
the former in that it is sequential in
explaining the initial trigger(s) for yawning,
and then accounts for the infectious nature of
this behavior among a few species.
As if the evidence for the
social/communication hypothesis was
self-evident, Guggisberg et al. (2010) state
that yawning may be used to communicate states
to group-members "in order to enhance behavioral
synchronization." These statements are not based
on scientific evidence. Furthermore, they
continue by stating that the
social/communication hypothesis has the "best
experimental evidence" among all existing
hypotheses. But the fact of the matter is that
the social/communication hypothesis has not been
experimentally tested.Just as the predictions
for physiological models indicate that yawning
acts to alter "up- or downturns of a given body
state", the main prediction of the
social/communication hypothesis states that
yawning synchronizes group behavior. Other than
the act of yawning itself being replicated,
studies of contagious yawning do not provide any
evidence of the sort that group behavior has
been synchronized. Observational studies of
group resting or sleeping (e.g., Deputte, 1994)
cannot be considered synchrony from yawning
without proper controls for circadian effects
(e.g., body temperature, general physiology,
etc.).
The social/communication hypothesis is not
supported by the largest body of experimental
evidence among all propositions for explaining
the origin and function of yawning, and in fact,
the predictions of this hypothesis have never
been experimentally tested. The authors even
contradict their position by stating, "Missing
elements of this model include controlled
studies observing the regulating effect of
yawning on synchronized group behavior" (the
main and foremost prediction of this
hypothesis). According to Guggisberg et al.
(2010), the physiological models fail because
there is insufficient support for the
predictions of each, yet the authors do not use
the same criteria when thinking about the
social/communication hypothesis. Although the
authors argue otherwise, the fact that yawning
is commonly thought of as a sign of boredom or
disrespect in many cultures (Schiller, 2002) can
be taken as evidence against the
social/communication hypothesis. This seemingly
universal perception exemplifies the lack of
informative communicative value to yawning, and
how unconscious and uncontrolled physiological
responses such as yawning can be misinterpreted
in various social contexts. Yawning may have a
social role in some primates (humans included),
but to this point any function remains largely
unclear.
7. Primitive versus derived features: an
evolutionary perspective on yawning
Supporting the view that yawning is
phylogenetically old, the onset of this behavior
occurs quite early in uterine development (11
weeks gestation in humans) (de Vries et al.,
1982). When thinking about yawning, or any
evolutionarily conserved behavior, it is
important to consider primitive versus derived
features. Primitive features of a trait can be
traced to the last common ancestor, while
derived features are present only in more
recently evolved organisms. Guggisberg et al.
(2010) state how the social/communication
hypothesis stems from the view that the trigger
for yawning is physiological. Therefore, the
underlying primitive feature of yawning is
physiological, and any subsequent social or
communicative role is a derived feature. The
divergence of contagious yawns during more
recent evolutionary history is strong evidence
in support of it being a derived feature. On the
other hand, the ubiquitous nature of spontaneous
and isolated yawns represents the primitive, or
underlying feature associated with this
behavior. It is simply inaccurate to describe
the origin of yawning to be social or
communicative. As discussed earlier, behavioral
synchronization has not been experimentally
studied, but if it does exist, it is a derived
feature of yawning. Given the objective of
Guggisberg et al. (2010) to review the available
literature on the origin and function of
yawning, it is troubling that these distinctions
were not discussed.
The authors argue, from an evolutionary
perspective, that the social/communication
hypothesis is sufficient to explain selection
for yawning among all vertebrate species
(Guggisberg et al., 2010). However, they do not
take into account that the first jawed
vertebrates were fishes, and presumably the
first to yawn (Baenninger, 1987), yet these
animals lack yawning contagion and any social
role of yawning described by
social/communication hypothesis. Furthermore,
the authors state "There is therefore no need to
postulate additional physiological functions of
yawning to explain its selection during
evolution." This assertion has many weaknesses.
First, if yawning originally evolved for
communicative purposes, it would imply that the
common ancestor, as well as any subsequent
descendants, should be particularly sensitive to
perceived yawns. There is no support for this.
Second, the communication hypothesis pertains
solely to group-oriented species, yet the
authors do not discuss the persistence of
yawning among solitary vertebrates, or why some
of the most social/gregarious mammals on the
planet (ungulates) have been described to
rarely, if ever yawn (Barbizet, 1958; Craemer,
1924; Heusner, 1946). Statements such as 'We
might have to get used to the idea that yawns
have a primarily social rather than
physiological function" are highly problematic
and misleading. As noted previously,
experimentally supported social aspects of
yawning (i.e., contagion) are limited to a few
primate species, while other primary
social/communicative roles of yawning (i.e.,
behavioral synchronization) have yet to be
established.
8. Conclusions
Given the ubiquitous nature of yawning, it
is likely multifunctional across species. And
when considering the potential
multifunctionality of this behavior, as well as
the presence of derived features in certain
lineages, it should be studied on a species by
species basis. Any differences in form and
function should be anchored to the underlying
physiology and the unique evolutionary histories
and ecological adaptations of that species. As
previously incorporated into the
thermoregulatory model (Gallup et al., 2010),
one potentially fruitful distinction for making
these predictions may come through the
classification of homeothermic (birds, mammals)
versus poikilothermic (fish, amphibians,
reptiles) vertebrates. No one theory is likely
to explain all manifestations or derived
functions of yawning, and since experimental
evidence of contagious yawning is only observed
in a small number of species in one lineage
(i.e., primates), the notion that the origin of
yawning is social/communicative can be easily
dismissed. It is inaccurate to conclude that the
social/communication hypothesis has the best
experimental evidence when in fact the main
predictions have not been tested. Statements
regarding any social function of yawning should
be tempered until much more comparative research
is conducted. The disproportionate focus of
Guggisberg et al. (2010) on human-related
studies may confound and distort the basic
nature of this ubiquitous behavior. The stated
"simplicity and elegance" of the
social/communication model needs to stand up to
rigorous experimental hypothesis testing before
it can be accepted as a viable account of
yawning.
