Content uploaded by Gordon Burghardt
Author content
All content in this area was uploaded by Gordon Burghardt on Mar 02, 2019
Content may be subject to copyright.
Special Issue
R9
Play in fishes, frogs
and reptiles
Gordon M. Burghardt
What animals engage in play?
Not too many years ago play was
considered by most scholars and
scientists as something we see in
rather intelligent warm-blooded
animals, such as monkeys and apes,
dogs, cats, elephants, otters, bears,
and some birds, such as crows and
parrots. Of course, horses play,
especially young ones: the origins of
the phrase ‘horsing around’ are not
hard to fathom. In fact, many other
mammals play, including marsupials
such as wombats and kangaroos.
Indeed, the play behavior of no
animal has been studied as much as
that of the laboratory rat: Serge and
Vivien Pellis wrote a book largely
devoted to reviewing some of the
literature on rat play. But what about
animals other than warm-blooded
vertebrates? Do they play and, if so,
how do we know? To answer this
we need to clear up some neglected
business.
What is play and how do we
recognize it? While many definitions
of play have been proposed, none
has worked well; behavior has
therefore been labeled as play
largely on anthropomorphic grounds,
so that people applied to animals
a largely human-derived view of
play as non-serious fun. Termed
anthropocentrism, this may work well
enough with puppies and kittens,
animals we know well, where we
recognize signs of pleasure and lack
of seriousness that often are integral
to definitions of play. A definition
should actually help to identify play
in species or contexts in which we
do not already accept that play is
being performed. We also need a
definition or set of criteria that apply
to all the major types of animal play:
social play, play with objects, and
solitary locomotor play — gamboling,
twisting, leaping, and swinging. All
these types of play can occur on land,
in the air, and in the water.
A set of five minimal criteria that
helps identify play of any type in any
animal has been generally adopted
(Graham and Burghardt, 2010). The
five criteria are that the behavior
should: (1) be incompletely functional
in the context in which it appears;
(2) be spontaneous, pleasurable,
rewarding, or voluntary; (3) be different
from other more serious behaviors in
form (for example, be exaggerated)
or timing (for example, occur early in
life, before the more serious version
is needed); (4) be repeated, but not in
abnormal and unvarying stereotypic
form (for example, rocking or pacing);
and (5) be initiated in the absence of
severe stress. In a single sentence:
play is repeated, seemingly non-
functional behavior differing from
more adaptive versions structurally,
contextually, or developmentally,
and initiated when the animal is in a
relaxed, unstimulating, or low stress
setting (Burghardt, 2014).
Applying these criteria allows us
to determine if a possible example
of animal behavior satisfies all the
criteria or just some of them, pointing
to others that we need to investigate
and apply. But even when the five
criteria are met, this only sets the
stage for further analysis. Just labeling
a behavior as play does not identify
the brain and behavior mechanisms
underlying it, the adaptive functions,
if any, served by the behavior, its
evolutionary history, how it develops
in individual animals, or how it is
experienced by the animals. To
answer these questions is often
difficult, even in well studied playful
species such as rats and monkeys.
For now, we first need to identify play
wherever it may occur in the animal
world and retain an open mind. Here
are a few contenders for play in the
so-called lower vertebrates, largely,
though not universally, denied the
playful moniker.
What are some examples of play
in fish? Whether fish play is a
controversy going back to the 19th
century. Perhaps the earliest detailed
claims were by the naturalist Charles
Holder, who described needlefishes
leaping over floating sticks and even
a turtle. But the bias against fish
being let into the pantheon of players
prevented locomotor, social, and
object play being seriously considered
as occurring in them. An influential
paper on concepts of play published
in 1945 by Frank Beach, perhaps then
the leading comparative psychologist
in America, debunked such early
work so effectively that his seriously
flawed arguments, along with the
desire of fish behavior researchers to
avoid anthropomorphic and anecdotal
observations, led those in the field
to dismiss the possibility that fish
could play. Today the play criteria and
readily available video footage has
reopened the question.
These early observations and many
others satisfy all or most of the play
criteria. Thus, leapfrogging in fish
fits, as does the balancing of twigs
and batting of balls in mormyrid fish
species, stingrays batting around balls
and competing for the opportunity to
do so, and cichlid fish that repeatedly
strike a self-righting thermometer.
While the incidence and complexity of
play is low as compared to mammals,
fishes provide good tests of play
criteria.
Are there any examples of play
in frogs? Unlike in fish, play in
amphibians has been very rarely
considered in discussions of their
behavior. To date there are still no
claims for, or observations of, play in
salamanders or caecilians, two major
groups of amphibians. In frogs, there
are some tantalizing examples. For
instance, dendrobatid (dart poison)
frogs are diurnal and active species
whose toxicity seems to encourage
more conspicuous and active
behavior. They are also quite social.
Adults often engage in brief wrestling
Figure 1. A young Nile soft-shelled turtle
interacting with colored rings. (Photo: Gordon
M. Burghardt.)
Current Biology Vol 25 No 1
R10
bouts irrespective of sex. This
behavior seems to fit the play criteria
and certainly warrants closer study.
I have studied Vietnamese mossy
frog tadpoles repeatedly ‘riding’
bubbles from an airstone at the
bottom of a tall tank to the top. I have
observed similar behavior in marine
fish in a large very tall communal
aquarium with such an air column.
Here it is actually possible to be a
bit anthropomorphic as the behavior
does look as if it would be fun for us!
What about play in reptiles?
Convincing examples of play have
been found in lizards, turtles, and
crocodilians. Komodo dragons, the
world’s largest lizards, engage in
complex interactions with objects
such as buckets, boxes, old shoes,
and balls. In fact, sped up a little bit
on video, their behavior is similar to
that of dogs. They even play tug of
war with their keepers over objects
such as cans and handkerchiefs.
Aquatic Nile soft-shelled turtles will
bounce basketballs and floating
bottles back and forth, manipulate
hoops (Figure 1) and play tug of war
with their keepers using hoses. North
American Emydid pond turtles often
engage in foreclaw titillation displays
in social interactions with each other
as hatchlings, behavior that otherwise
is only found in sexual and sometimes
agonistic encounters as adults.
Crocodilians also engage in object
play. A giant saltwater crocodile
played with a basketball on a tether
as part of enrichment. Although only a
few papers have been published and
cited in the references below, behavior
patterns meeting the play criteria have
been met.
If play is so widespread in the
animal kingdom, how and why did
this happen? Play is often found in
the most intelligent and adaptable
species, but we now know that it is
not restricted to them. The presence
of play facilitates novel and creative
behavior, but this does not tell us
about its origins. Indeed, play is
so diverse and heterogeneous that
no single factor can explain when
and where it appears in the lives of
animals. We also know little about
the function of play in these animals,
but as we are just beginning to get
a handle on the function of play in
mammals, our relative ignorance
about fish, frogs, and reptiles is not
surprising. But invertebrates play
also — in fact, some of the best
evidence for the function of play
comes from work on spiders, where
play was never observed until recently.
So, play, while very prominent in
mammals and many birds, is relatively
rare in other species. One proposal,
termed Surplus Resource Theory,
is that the origins of play are found
in animals with sufficient metabolic
resources for sustained activity and
complex behavior that needs to be
deployed in varying ways. They also
need the time and safety to engage in
behavior that may not be immediately
advantageous, but through which
animals learn or perfect behavioral
skills, social acumen, physiological or
perceptual abilities, and other means
that enhance survival compared to
non-playing conspecifics. On the
other hand, in its ancient and more
primitive incarnations, playing may
not have had any specific advantage
over non-playing, but eventually the
benefits outweighed the often serious
costs of play in energy and risks
of injury and predation. A door has
been opened, and exploring what lies
beyond may be both fascinating and
important.
Where can I find out more about
play?
Bateson, P., and Martin, P. (2013). Play, playfulness,
creativity, and innovation. (Cambridge, UK:
Cambridge University Press.)
Beach, F.A. (1945). Current concepts of play in
animals. Am. Nat. 79, 523–541.
Bekoff, M., and Byers, J.A. (Eds.). (1998).
Animal Play: Evolutionary, Comparative, and
Ecological Perspectives. (Cambridge, UK:
Cambridge University Press.)
Burghardt, G.M. (2005). The Genesis of Animal
Play: Testing the Limits. (Cambridge, MA: MIT
Press.)
Burghardt, G.M. (2011). Defining and recognizing
play. In The Oxford Handbook of the
Development of Play, A.D. Pellegrini (Ed.) (New
York: Oxford University Press.) pp. 9–18.
Burghardt, G.M. (2014). A brief glimpse at the long
evolutionary history of play. Anim. Behav.
Cogn. 1, 90–98.
Burghardt, G.M., Dinets, V., and Murphy, J.B.
(2014). Highly repetitive object play in a cichlid
fish (Tropheus duboisi). Ethology 120, http://
dx.doi.org/10.1111/eth.12312.
Graham, K.L., and Burghardt, G.M. (2010). Current
perspectives on the biological study of play:
Signs of progress. Q. Rev. Biol. 85, 393–418.
Pellis, S.M., and Pellis, V.C. (2009). The Playful
Brain, Venturing to the Limits of Neuroscience.
(Oxford, UK: Oneworld Press.)
Pruitt, J.N., Burghardt, G.M., and Riechert, S.E.
(2012). Non-conceptive sexual behavior
in spiders: a form of play associated with
body condition, personality type, and male
intrasexual selection. Ethology 118, 33–40.
Departments of Psychology and Ecology
& Evolutionary Biology, University of
Tennessee, Knoxville, TN 37996-0900, USA.
E-mail: gburghar@utk.edu
Fun and play in
invertebrates
Sarah Zylinski
Where should we look for playful
invertebrates? The notion that
invertebrates might indulge in play,
and especially that they might have
fun doing it, is generally met with
scepticism. But given that the same
was true of play in ‘lower’ vertebrates
such as reptiles and fish until relatively
recently, perhaps we shouldn’t
discount the possibility outright. So
where should we look? Given that
play is most frequently observed in
large-brained vertebrate lineages,
perhaps our first port of call should be
the cephalopods. These large-brained
molluscs are heralded as uniquely
intelligent amongst the invertebrates,
and their deep evolutionary split from
the vertebrates provides us with a
unique independent data point against
which to investigate general trends
in intelligence, cognition and, in this
case, play. Shallow water coleoid
cephalopods — octopuses, cuttlefish
and squid — are well known for their
capacity for complex learning and
their flexible, complex behaviours.
Their brains are comparable to
vertebrates in relative size, with
dedicated learning and memory
centres analogous in many ways to
the vertebrate cortex. On the flip side,
cephalopods don’t afford parental
care to their offspring, are typically
short lived (often one or two years),
and are often semelparous (that
is, they die after their first attempt
at reproduction). Furthermore, the
species considered to have the
highest cognitive intelligence are
solitary and show little or no social
behaviour. What evidence is there
then that cephalopods play, and more
importantly, do they have fun doing it?
So do cephalopods play? When
introducing my behavioural
experiments with cephalopods in
seminars, I often joke that there is
nothing more demoralizing than being
outsmarted by your experimental
animal. Indeed, there are some
individuals that seem to delight in
being mischievous. For example, some
cuttlefish use their siphons to squirt
water at their keeper when impatient
to be fed. However, there is currently
