Experimental
evidence of contagious yawning in budgerigars
(Melopsittacus undulatus).
Gallup AC, Swartwood L, Militello J, Sackett
S.
Psychology Department, State
University of New York at Oneonta,
USA
Abstract
Experimental evidence of contagious yawning
has only been documented in four mammalian
species. Here, we report the results from two
separate experimental studies designed to
investigate the presence of contagious yawning
in a social parrot, the budgerigar
(Melopsittacus undulatus). In Study 1, birds
were paired in adjacent cages with and without
visual barriers, and the temporal association of
yawning was assessed between visual conditions.
In Study 2, the same birds were exposed to video
stimuli of both conspecific yawns and control
behavior, and yawning frequency was compared
between conditions. Results from both studies
demonstrate that yawning is contagious. To date,
this is the first experimental evidence of
contagious yawning in a non-mammalian species.
We propose that future research could use
budgerigars to explore questions related to
basic forms of empathic processing.
Yawning is characterized by a powerful
gaping of the jaw with inspiration, a brief
period of peak muscle contraction, and a passive
closure of the jaw with shorter expiration
(Barbizet 1958). Nonsocial yawning, also known
as spontaneous yawning, is believed to be
relatively widespread among vertebrates
(Baenninger 1987) and may function in promoting
cortical arousal (Baenninger 1997) and/or state
change (Provine 1986, 1996, 2005) by decreasing
brain temperature (Eldakar et al. in press;
Gallup and Gallup 2007, 2008; Gallup and Eldakar
2012; Massen et al. 2014; Shoup-Knox et al.
2010). Contagious yawning, which can be elicited
by sensing or thinking about the action in
others (Provine 2005), appears to be a more
recently derived behavior that may function in
group coordination and vigilance in social
species (Gallup and Gallup 2007; Gallup 2011;
Miller et al. 2012a). Introduction
Yawning is characterized by a powerful
gaping of the jaw with inspiration, a brief
period of peak muscle contraction, and a passive
closure of the jaw with shorter expiration
(Barbizet 1958). Nonsocial yawning, also known
as spontaneous yawning, is believed to be
relatively widespread among vertebrates
(Baenninger 1987) and may function in promoting
cortical arousal (Baenninger 1997) and/or state
change (Provine 1986, 1996, 2005) by decreasing
brain temperature (Eldakar et al. in press;
Gallup and Gallup 2007, 2008; Gallup and Eldakar
2012; Massen et al. 2014; Shoup-Knox et al.
2010). Contagious yawning, which can be elicited
by sensing or thinking about the action in
others (Provine 2005), appears to be a more
recently derived behavior that may function in
group coordination and vigilance in social
species (Gallup and Gallup 2007; Gallup 2011;
Miller et al. 2012a).
To date, experimental evidence of contagious
yawning is restricted to humans (Homo sapiens;
e. g., Provine 1986; Platek et al. 2003),
chimpanzees (Pan troglodytes) in response to
conspecifics (Amici et al. 2014; Anderson et al.
2004; Campbell et al. 2009; Campbell and de Waal
2011; Massen et al. 2012) and human yawns
(Campbell and de Waal 2014; Madsen et al. 2013;
but see Amici et al. 2014), domesticated dogs
(Canis familiaris) in response to human yawns
(Joly-Mascheroni et al. 2008; Madsen and Persson
2013; Romero et al. 2013; Silva et al. 2012; but
see Harr et al. 2009; O'Hara and Reeve 2011;
Buttner and Strasser 2014), and, most recently,
a subline of high-frequency yawning
Sprague-Dawley rats (Rattus norvegicus; Moyaho
et al. 2014). Video-induced yawning has also
been reported in stumptail macaques (Macaca
arctoides; Paukner and Anderson 2006), but this
response also cooccurred with heightened
self-directed behaviors and thus appears to be
due to social tension or stress rather than
contagion. Species that have thus far failed to
show contagious yawning in an experimental
design include bonobos (Pan paniscus),
orangutans (Pongo abelii), and gorillas (Gorilla
gorilla) in response to both conspecific and
human yawns (Amici et al. 2014), and
domesticated dogs and redfooted tortoises
(Geochelone carbonaria) in response to
conspecifics (Harr et al. 2009; Wilkinson et al.
2011).
To date, experimental evidence of contagious
yawning is restricted to humans (Homo sapiens;
e. g., Provine 1986; Platek et al. 2003),
chimpanzees (Pan troglodytes) in response to
conspecifics (Amici et al. 2014; Anderson et al.
2004; Campbell et al. 2009; Campbell and de Waal
2011; Massen et al. 2012) and human yawns
(Campbell and de Waal 2014; Madsen et al. 2013;
but see Amici et al. 2014), domesticated dogs
(Canis familiaris) in response to human yawns
(Joly-Mascheroni et al. 2008; Madsen and Persson
2013; Romero et al. 2013; Silva et al. 2012; but
see Harr et al. 2009; O'Hara and Reeve 2011;
Buttner and Strasser 2014), and, most recently,
a subline of high-frequency yawning
Sprague-Dawley rats (Rattus norvegicus; Moyaho
et al. 2014). Video-induced yawning has also
been reported in stumptail macaques (Macaca
arctoides; Paukner and Anderson 2006), but this
response also cooccurred with heightened
self-directed behaviors and thus appears to be
due to social tension or stress rather than
contagion. Species that have thus far failed to
show contagious yawning in an experimental
design include bonobos (Pan paniscus),
orangutans (Pongo abelii), and gorillas (Gorilla
gorilla) in response to both conspecific and
human yawns (Amici et al. 2014), and
domesticated dogs and redfooted tortoises
(Geochelone carbonaria) in response to
conspecifics (Harr et al. 2009; Wilkinson et al.
2011).
Observational evidence for contagious
yawning to conspecifics has been reported in
bonobos (Demuru and Palagi 2012; Palagi et al.
2014), gelada baboons (Theropithecus gelada;
Palagi et al. 2009), budgerigars (Melopsittacus
undulatus; Miller et al. 2012b), and wolves
(Canis lupus lupus; Romero et al. 2014).
Research suggests that contagious yawning is
triggered by mechanisms that differ from those
involved in other involuntary actions (Amici et
al. 2014). In particular, contagious yawning is
thought to represent a basic form of empathy
rooted in a perception-action mechanism known as
state matching or emotional contagion (Preston
and de Waal 2002). Consistent with this view,
previous research has documented a strong
association between contagious yawning and
empathy (for a discussion, see Campbell and de
Waal 2014). For example, in humans, contagious
yawning is more common among participants who
score high on empathy measures (Platek et al.
2003; but see Bartholomew and Cirulli 2014), and
thinking about yawning has been shown to
activate brain areas implicated in empathic
processing (e.g., Platek et al. 2005; Nahab et
al. 2009). Comparative studies investigating the
developmental onset and decline of contagious
yawning also generally support this view
(Anderson and Meno 2003; Bartholomew and Cirulli
2014; Giganti et al. 2012; Madsen et al. 2013;
Madsen and Persson 2013; Massen et al. 2014),
since empathy-related capacities in humans
emerge in early childhood and decline in old age
(Bailey and Henry 2008; Perner and Lang 1999).
Growing comparative research also demonstrates a
positive relationship between contagious yawning
and group affiliation or social
closeness/bonding (Campbell and de Waal 2011;
Demuru and Palagi 2012; Norscia and Palagi 2011;
Palagi et al. 2009; Romero et al. 2013, 2014;
Silva et al. 2012), which supports a connection
with empathic processing. For example, an
initial study on chimpanzees provided evidence
for an ingroup bias for contagious yawning
(Campbell and de Waal 2011). In particular,
captive chimpanzees shown video stimuli of other
chimpanzees yawning demonstrate contagion to
ingroup members but not to unfamiliar
conspecifics. Subsequent studies, however, have
failed to demonstrate a familiarity bias for
chimpanzees viewing human yawns (Campbell and de
Waal 2014; Madsen et al. 2013), and at least one
study provided evidence that relationship
quality among chimpanzees did not predict yawn
contagion (Massen et al. 2012). There has also
been mixed support for domesticated dogs to yawn
more in response to yawns from familiar humans
(Madsen and Persson 2013; O'Hara and Reeve
2011).
Since contagious yawning may represent a
primitive form of empathy, unequivocally
demonstrating the presence of this behavior in a
laboratory animal, with the ability to
manipulate it experimentally, could be important
for exploring basic questions related to this
cognitive capacity. For example, budgerigars (M.
undulatus), a species of social parrot
indigenous to Australia, represent a good
candidate species because these birds form
lasting bonds with breeding pairs and interact
within coordinated flocks throughout the year
(Wyndham 1980). Furthermore, previous research
has demonstrated that these birds show automatic
imitation to video stimuli (Mui et al. 2008),
whereby the sight of another's action tends to
elicit the same action in the observer (Sturmer
et al. 2000). Using a naturalistic approach,
Miller et al. (2012b) initially examined this
species and revealed that yawns were temporally
clustered in an undisturbed, established flock
of captive budgerigars. These findings suggest
yawning is contagious, similar to the other
aforementioned observational studies; however,
such methods cannot completely account for
circadian physiological synchrony and/or
collectively sensed environmental stimuli (both
of which influence the expression of yawning,
see Baenninger 1997).
Here, we describe two separate experimental
studies designed to investigate the presence of
contagious yawning in budgerigars in a
controlled laboratory setting. Study 1 paired
birds in adjacent cages with and without visual
barriers, and the frequency of potentially
contagious yawns (i.e., those occurring within a
restricted temporal proximity) was compared
based on the opportunity for visual information
transfer. Given that in the wild, the group size
and composition of budgerigars can fluctuate
depending upon the season (Wyndham 1983) and
that both sexes within this species show vocal
call convergence to ingroup members when housed
in captivity (Farabaugh et al. 1994; Hile et al.
2000), pairs of familiar and unfamiliar birds
were tested together in this experiment to
assess the potential for a familiarity bias in
this response. Unlike chimpanzees or wolves,
however, in which observing unfamiliar
conspecifics may elicit hostility and
familiarity biases for contagious yawning have
been reported (Campbell and de Waal 2011; Romero
et al. 2014), budgerigars show more fluid
flocking and thus we did not expect a strong
bias for contagious yawning. Since previous
research has demonstrated that budgerigars
respond to video displays of real and virtual
conspecifics (Mui et al. 2008; Moravec et al.
2010; Mottley and Heyes 2003), Study 2 compared
the yawn frequency of the same birds from Study
1 during exposure to video presentations of
yawns versus control behaviors.
Discussion
Yawning in response to sensing or thinking
about the action in others may represent a
primitive form of empathy. Despite the potential
importance of identifying various non-human
species showing this capacity, comparative
investigations of contagious yawning have been
limited. These two experimental studies reveal
the presence of contagious yawning in
budgerigars in a controlled laboratory setting,
corroborating a previous observational report
assessing the temporal distribution of yawns in
an undisturbed flock (Miller et al. 2012b). The
experiment in Study 1 provides an ecologically
valid measure of social contagion utilizing the
signaling of a live demonstrator producing real
yawns (improving upon Wilkinson et al. 2011).
Temporally classified contagious yawns during
these trials (i.e., those occurring within 5 min
of a yawn from the matched pair) occurred more
than three times as often when the birds could
see one another, whereas there was no difference
in the frequency of spontaneous yawning across
visibility conditions. The experiment in Study 2
used a more traditional laboratory-based
methodology for testing contagious yawning
(e.g., Anderson et al. 2004), by presenting
small groups of birds with repeated video clips
of conspecific yawns and control behavior. In
comparison with matched control stimuli,
budgerigars in this experimental condition
yawned more than twice as often and this
response was not linked with indicators of
stress or anxiety (i.e., self-directed
behaviors). It is important to note, however,
that this represents a modest effect given the
magnitude of the video stimulus (over 150 yawns
were displayed during the 10-min experimental
session).
We found no evidence of an ingroup bias for
contagious yawning in budgerigars. That is,
yawns appeared similarly contagious between
birds housed within the same aviary and across
outgroup pairs. Previous experiments on
chimpanzees and domesticated dogs have
demonstrated mixed support for a familiarity
bias in contagious yawning (Campbell and de Waal
2011; Romero et al. 2013; Silva et al. 2012; but
see Madsen et al. 2013; Madsen and Persson
2013), while observational studies on primates
(human and non-human) and wolves have shown a
positive association with social closeness and
affiliation (Demuru and Palagi 2012; Norscia and
Palagi 2011; Palagi et al. 2009, 2014; Romero et
al. 2014). There may be reasons to suspect that
budgerigars would be less sensitive to
familiarity or social closeness for various
forms of state matching since they naturally
live in much larger, more fluid groups of
unrelated individuals than the abovementioned
species and can form flocks of over 1000
coordinated members in the wild (Wyndham
1980).
Conversely, we have recently discovered
evidence of an ingroup bias for stretching in
this species (unpublished data), a more overt
behavior directly relevant to flocking. It
remains possible that limitations in the current
study do not permit a true comparison to the
documented ingroup/outgroup effects of
contagious yawning in other species. For
example, since all of the budgerigars were
housed in the same room, this permitted vocal
communication between the two groups and perhaps
created some degree of familiarity or social
affiliation prior to testing. Previous research
has demonstrated that vocal call convergence
occurs among budgerigars housed in captivity
(Farabaugh et al. 1994; Hile et al. 2000),
though in our study, vocalizations between the
pairs were uncommon to nonexistent. Another
limitation to the current study was that the
degree of social closeness between cage mates
was not recorded prior to the experiment, so
this effect could not be assessed along a
continuum.
In summary, budgerigars represent the
first non-mammalian species, and only the fifth
species to date (including humans, chimpanzees,
domesticated dogs, and a subline of high-yawning
Sprague-Dawley rats), to show contagious yawning
in an experimentally controlled setting. These
findings provide an example of convergent
evolution. Given the association between
contagious yawning and empathy, we suggest
additional comparative research be conducted to
assess the prevalence of contagious yawning in
other social vertebrates. Furthermore, since
empathic responses have already been
demonstrated in avian species (e.g., Edgar et
al. 2011; Wascher et al. 2008), and the current
findings show that contagious yawns can be
experimentally manipulated, we propose that
budgerigars represent a good model for exploring
primitive forms of empathic processing.
