The lush riverside vegetation sways as a herd of elephant wends its way between the broken pools. Standing at the top of an embankment, a half-grown male is watching a larger elephant trudge up the slope toward it.

Without warning, the youngster squats down on his haunches (just like a dog) and launches himself down the slope. Slithering at a good speed, he collides (with an audible thump) into the elephant below, sweeping them both, in a flurry of waving limbs and trunks, to the foot of the hill. There, lying on their stomachs, the pair jousts, twisting and parrying with trunk and tusk.

Meanwhile up above, an onlooker waits, scuffing his feet impatiently and swinging his trunk from side to side. He seems to be waiting for them to clear the trail, but when the two finally begin to traipse up the slope, he squats and whooshes down to create a three-elephant pile up.

What these elephants are up to is a mystery.

In fact, it’s one of the greatest enigmas in the field of animal behaviour.

A pachyderm rite of passage? An evolutionary precursor to bob-sledding? Itchy-rump syndrome?

No, these elephants are playing, and science has no idea why.

Now I know what you’re thinking:

‘Come off it, everyone knows that young animals play to prepare for adulthood.’

The lion cub pinning his sister in a headlock is learning how to make dinner; the capering impala fawns are practicing their escape routine. And the tobogganing elephants are... er... um...  Oh yeah, play also promotes camaraderie, so the elephants are slip-sliding their way into life-long alliances.

I hate to tell you, but knowing this is like knowing that chocolate gives you pimples, or you’ll catch a cold from walking in the rain.

OK, I realise it’s not your fault; I’ve seen the myriad wildlife documentaries too.

But as far as science is concerned, there’s not one iota of evidence to support these myths.

And if anyone should know, it’s me (but I’ll get to that).

The sorry state of play...

Now I don’t want you to imagine that our ignorance about play is for want of trying.


Back in the 1970s and 80s dozens of zealous play researchers rushed out to quantify play, eagerly counting the pinches and pokes of brawling baboons, the horn-clashes of sparring of antelopes or the writhings of wrestling rodents.

Scientists figured that once they understood the mechanics of play, in a whole array of beasties, its function would become obvious. Heck, this approach has worked with almost every other type of animal behaviour.

But not with play.

In quiet desperation, play researchers have come up with more than two dozen possible benefits of play but, in spite of four decades of effort, they’ve found conclusive evidence for none.

Pretty grim, huh?

So maybe play simply doesn’t have a function.

I mean it’s obviously fun. Isn’t that reason enough to do it?

There’s a problem: having fun is costly. Those mud-sliding otters and stick-tossing cormorants could have channelled their time and energy into out-growing rivals or rearing more kids.

What’s worse, play is perilous. Frolicking bighorn sheep lambs impale themselves on cacti, elephant calves get stuck in mud, ibex lambs plummet from cliffs, fur seal pups are snapped up by sealions or washed out to sea, and my front teeth suffered irrevocably in the great roller-skating catastrophe of 1972.

You see natural selection quickly weeds out any activity that reduces an individual’s chances of survival or breeding (and let me tell you, missing teeth severely impair one’s breeding potential) unless the behaviour also helps in some way. So play must have benefits, we just can’t find them!

Now given the history of play research, you’d think I’d have been a bit cautious.

You’d think I’d have said, ‘Function of Play?  Um, I dunno...’

But could YOU have resisted the lure of working with playing meerkats?


Meerkat pups start to play as soon as they can totter about the burrow. On stubby legs they crawl over one another, gnawing on any protruding ear, limb or nose. Within a week, the little guys are play-fighting with gusto, standing up on their hind legs to clasp each other around the shoulders.

Like sumo wrestlers, they totter back and forth, struggling to throw their opponent and nipping at one another’s cheeks. Eventually one pup topples backwards on to the sand, and the victor leaps on its tummy, pinning it down while snapping at its paws and ears.

My plan was to follow 45 wild meerkat pups throughout their lives (at the Kalahari Meerkat Research Project – aka 'Meerkat Manor') to see whether play affected their destinies.

I wanted to know if play-fighting taught pups crucial combat skills (the ‘practice theory’) or whether meerkats that played together stayed together (the ‘social bonding theory’). You see no-one had ever tested the idea that play cements friendships or reduces aggression. And as far as the practice theory went, researchers had only ever looked at whether predator spawn played to learn their trade.

Of course, the results of these studies were less than encouraging.

Games of cat and mouse: practice makes purrfect?

Take the northern grasshopper mouse for example. This fierce little rodent stalks the grasslands of North America at night, massacring small animals to satisfy its carnivorous tastes.

But being mouse-sized puts a bit of a dampener on the whole predatory thing and young grasshopper mice must learn how to subdue their dinner. Researchers looked to see whether little mouselets that played a lot (pouncing, boxing, chasing and wrestling with their siblings) were more efficient at snagging feisty crickets (1). They weren’t.

The same was true for coyote pups, although they were tested with bigger munchies of course (2). And domestic cats? Surely all that ultra-cute pouncing, batting and chasing gives a kitten the edge when procuring lunch? 

Well, one study fitted kittens with vision-distorting goggles so they couldn’t see to play (3) and another compared cats reared without toys to those who’d enjoyed a kittenhood chockfull of faux-fur mice and balls of yarn (4).

And guess what: there was no difference at all in anyone’s ability to catch mice.

But maybe play-fighting helps youngsters learn battle skills. Mock combat (also called ‘rough and tumble’) is central to the games of many species and in polygynous beasts (where big macho males must brawl to secure a harem), males play-fight much more than their sisters (5). Doesn’t this imply practice?

Meerkats seemed the ideal species to test out the idea. You see these desert-dwelling carnivores live in close-knit groups in which everyone lends a paw rearing the pups. The problem is, the group’s reigning pair hog the breeding, parenting around 80% of the group’s sprogs (6). So if you’re a meerkat of either sex and you want to breed, you’re going to have to win supremacy in a group. And to do that, ‘Ya gotta learn ta fight!’

I figured that if play-fighting was able to help youngsters perfect their combat skills, meerkats should be giving it their all.

Grappling with meerkats

But measuring play proved mighty tricky. Even in very playful critters like meerkats, youngsters romp for only about 6% of their waking day, and the mood can hit at any time. So I spent 94% of my waking day trudging around the Kalahari Desert, grinding my (grit encrusted) teeth in frustration. During the sweltering summer months, I had to wade through endless swathes of thigh-high sour grass (which secretes skin-blistering acid), enshrouded in a miasma of tiny bees (that settle in the crook of elbow and knee to sip sweat, and sting at the slightest movement).

And when my little darlings did finally get down to a bit of rough and tumble, it was chaos! Everyone leapt and bit and rolled at such a furious pace I couldn’t identify who was doing what. Often the whole group joined in (up to 30 animals), creating a seething ball of fur and waving limbs.

The only meerkat bits visible in the melee were wildly threshing tails. So - not to be defeated - each morning, while the meerkats were soaking up the early sun, I’d crawl around the group on my stomach, clutching a fistful of marker pens and surreptitiously scribbling coloured rings on everyone’s tails.

So now I’ve got your sympathy (well, at least you can pretend), I’ll tell you what I found out.

I figured that if meerkats play to hone their fighting manoeuvres, you’d expect youngsters who play-wrestled lots (i.e. practised more) to be more triumphant (or at least to show greater improvement) than sluggards who rarely indulged in a game.

But this wasn’t the case. More importantly, when my target meerkats eventually left home and set out to rule the world (well, at least a group), their chances of winning a real, bloody, no-holds-barred stoush were utterly unrelated to how much they’d played, or how good they were at besting their play partners (7). A bit of a bummer really.

So what about the warm, fuzzy, social effects of play? We know that growing spotted hyena cubs pick fewer fights when they start to romp more (8) and that captive chimps play most often just before mealtime, possibly to defray group tension (9).

But play did NOT reduce aggression in my meerkats (10).

Meerkats that spent lots of time rolling about in play were just as antsy as those that didn’t. Regardless of whether it was pups squabbling over a juicy scorpion, teenagers bickering for status, or barneys initiated by the dominant female (who grows peevish when pregnant and takes it out on her daughters), favoured playmates were just as likely to come to blows (or bites) as individuals that never shared a game.

So how about social bonding?  Were frequent players more committed to their group? Did they put off leaving home, or sneak off alone less often than non-players? Did they help out more, or show more affection by grooming others a lot?

No, no, no! Play had no affect on any of these this things (11).(I know you can see a pattern emerging here...) Similarly, dispersing meerkats, who almost always left home in the company of a same-sex sibling, were no more likely to team up with a favoured playmate than with an animal they rarely played with (12).

So there you are. Five years of my life and no answers. I simply cannot tell you why meerkats play.

But before we all get too depressed (alright, it’s only me, but play along for moral support), there is some good news. Other play researchers have been making headway. Finally they’ve managed to uncover evidence that play does help animals to survive and breed. This is thrilling stuff (OK, I admit I don’t have much of a life).

Play DOES help!

One study looked at the games of Belding’s ground squirrels, stumpy little rodents that frolic in the alpine meadows of western USA. Baby squirrels indulge in playful fisticuffs for about three weeks before getting down to more serious pursuits (like eating), but those that play more seem to be better coordinated (at least they’re better at tight-rope-walking when plunked onto a horizontal stick). But what’s really important is that these extra playful individuals go on to make better mums, rearing more little ankle-biters in their first breeding season (13).

Play also seems to help feral horses. Youngsters that kick up their heels a lot are more likely to reach their first birthday than foals that are disinclined to horse around (14). Of course, it’s hard to separate the effects of play from the benefits of good nutrition (since well-fed critters play more than lean ones), but one study has managed to do it.


Long-term research on American brown bears has revealed that cubs who romp a lot are more likely to survive to independence, even after taking into account the cub’s (and its mum’s) condition, and the availability of food (15).

But why, oh why?

Of course, this still doesn’t tell us how play helps critters.

And you’re probably wondering about all the other (20 or so) potential benefits of play which I’ve steadfastly failed to mention. Well, clearly I can’t describe all of them here (I think you’re doing amazingly well sticking with me this far!), so I’ll just tell you about the evidence that I find most compelling. This evidence, by the way, comes to us courtesy of the play researcher’s BBF: the good ol’ laboratory rat.

Surprisingly, the domestic rat is one of the most playful creatures on earth (house mice, by comparison, virtually never play). Plunk two little rats together and it’s almost impossible to stop them whooping it up. But thwart a young rat’s zeal for play (by rearing it alone or with drugged companions that won’t play) and you create an adult that loses its cool in social situations.

When things start getting edgy, play-deprived rats either succumb to rat-rage or scarper, quaking, to a corner. And the lack of play is responsible, because if you let an isolated rat fool around for just one hour daily, it turns into a normal chilled dude (16). And there’s also evidence that primates (including humans) behave in the same way (17, 18).

So does merrymaking teach young critters how to read the intentions of others (19)? Or perhaps it boosts their confidence by letting them experience winning and losing in a non-threatening way (20)? I have to admit that if I had to put money on it, I’d go for the theory about stress.

You see when a baby animal experiences stress, its brain changes so that it’s subsequently less sensitive to stress hormones. This means that, as an adult, the critter recovers more rapidly after a hair-raising experience (21).  And we know that play (which normally consists of exciting ‘flight or fight’ behaviors) activates the same neurochemical pathways as stress (22). So maybe young animals are using play to prime or fine-tune their own stress response.

The other very important thing we’ve learnt from the humble rat is that when they’re reared with lots of companions and interesting objects, they develop larger brains than rats that grow up in austere surroundings. These enriched rats not only have heavier cerebral cortexes, with more neural connections, they learn more quickly too.

Researchers teased apart the factors that promoted this brain growth and found that sensory stimulation and arousal (even together) couldn’t increase cortical growth unless they were coupled with interactive behavior (i.e. play or training). And it was play that had the biggest impact; in fact, the more a young rat played, the more rapidly its brain grew (23). 

Of course, we’ve got a long way to go before we know whether other creatures benefit in the same way as lab rats (although there’s evidence that monkeys do).

Nevertheless – personally - I can’t imagine any better reason for indulging in a bit of fun.

So next time you’re lounging on the sofa watching the telly and those cute little lion cubs come on, just ignore the silly commentary and know that you’re witnessing one of Mother Nature’s mightiest of mysteries.


1. Davies, V. A. & Kemble, E. D. 1983. Social play behaviors and insect predation in northern grasshopper mice. (Onychomys leucogaster). Behavioural Processes, 8, 197-204.

2. Vincent, L. E. & Bekoff, M. 1978. Quantitative analyses of the ontogeny of predatory behaviour in coyotes, Canis latrans. Animal Behaviour, 26, 225-231.

3. Thomas, E. & Schaller, F. 1954. Das Spiel der optisch isolierten Kasper-Hauser-Katze. Naturwissenschaften, 41, 557-558. Reprinted and translated in: Evolution of play behaviour. 1978. (Ed. by D. Muller-Schwarze.) Stroudsburg, PA: Dowden, Hutchinson & Ross.

4. Caro, T.M. 1980. Effects of the mother, object play, and adult experience on predation in cats. Behavioral and Neural Biology, 29, 29-51.

5. Fagen, R. M. 1981. Animal Play Behavior. New York: Oxford University Press.

6. Griffin, A. S., Pemberton, J. M., Brotherton, P. N. M., McIlrath, G. M., Gaynor, D., Kansky, R. & Clutton-Brock, T. H. 2003. A genetic analysis of breeding success in the cooperative meerkat (Suricata suricatta). Behavioral Ecology, 14, 472-480.

7. Sharpe, L.L. 2005. Play fighting does not affect subsequent fighting success in wild meerkats.  Animal Behaviour, 69, 1023-1029.

8. Palagi, E., Cordoni, G., Borgonini Tarli, S.M. 2004. Immediate and delayed benefits of play behaviour: New evidence from chimpanzees (Pan troglodytes). Ethology, 110, 949-962.

9. Drea, C. M., Hawk, J. E. & Glickman, S. E. 1996. Aggression decreases as play emerges in infant spotted hyaenas: preparation for joining the clan. Animal Behaviour, 51, 1323-1336.

10. Sharpe, L.L. & Cherry, M.I. 2003. Social play does not reduce aggression in wild meerkats. Animal Behaviour, 66, 989-997

11. Sharpe, L.L. 2005. Play does not enhance social cohesion in a cooperative mammal. Animal Behaviour, 70, 551-558.

12. Sharpe, L.L. 2005. Frequency of social play does not affect dispersal partnerships in wild meerkats. Animal Behaviour, 70, 559-569.

13. Nunes, S., Muecke, E-M., Lancaster, L.T., Miller, N.A., Mueller, N.A., Muelhaus, J. & Castro, L. 2004. Functions and consequences of play behaviour in juvenile Belding's ground squirrels. Animal Behaviour, 68, 27-37.

14. Cameron, E.Z., Linklater, W.L., Stafford, K.J. & Minot, E.O. 2008. Maternal investment results in better foal condition through increased play behaviour in horses. Animal Behaviour, 76, 1511-1518.

15. Fagen, R.M. & Fagen, J. 2009. Play behaviour and multi-year juvenile survival in free-ranging brown bears, Ursus arctos. Evolutionary Ecology Research, 11, 1–15.

16. Potegal, M. & Einon, D. 1989. Aggressive behaviors in adult rats deprived of playfighting experience as juveniles. Developmental Psychobiology, 22, 159-172.

17. Pellis S., Pellis V. 2009. The Playful Brain: Ventures to the Limits of Neuroscience. Oxford, UK: Oneworld Press.

18. Saunders, I., Sayer, M. & Goodale, A. 1999. The relationship between playfulness and coping in preschool children: a pilot study. American Journal of Occupational Therapy, 53, 221-226.

19. Bekoff, M. & Allen, C. 1998. Intentional communication and social play: how and why animals negotiate and agree to play. In: Animal Play: Evolutionary, Comparative and Ecological Perspectives. (Ed. by M. Bekoff & J. A. Byers), pp. 161-182. Cambridge: Cambridge University Press.

20. Biben, M. 1998. Squirrel monkey playfighting: making the case for cognitive training function for play. In: Animal Play: Evolutionary, Comparative and Ecological Perspectives. (Ed. by M. Bekoff & J. A. Byers), pp. 161-182. Cambridge: Cambridge University Press.

21. Meaney, M.J., Mitchell, J.B., Aitken, D.H. & Bhatnagar, S. 1991. The effects of neonatal handling on the development of the adrenocortical response to stress: implications for neuropathology and cognitive deficits in later life. Psychoneuroendocrinology, 16, 85-103.

22. Siviy, S. M. 1998. Neurobiological substrates of play behavior: glimpses into the structure and function of mammalian playfulness. In: Animal Play: Evolutionary, Comparative and Ecological Perspectives. (Ed. by M. Bekoff & J. A. Byers), pp. 221-242. Cambridge: Cambridge University Press.

23. Ferchmin, P. A. & Eterovic, V. A. 1982. Play stimulated by environmental complexity alters the brain and improves learning abilities in rodents, primates and possibly humans. Behavioral and Brain Sciences, 5, 164-165.

Photo credits:

(1) Elephants, posted on Flickr by Andrew Pescod. (2) Dog, posted on Flickr by Arno Meintjes. (3) Meerkats, Copyright: Andrew Young. (4) Kittens, posted on Flickr by Aquila. (5) Meerkats, Copyright: Andrew Young. (6) Bear, posted on Flickr by Beverly and pack. (7) Foxes, posted on Flickr by Pat Gaines.

About the author: Lynda Sharpe is a postdoctoral research fellow at the University of Stellenbosch, South Africa. Originally from Australia, she has spent the last 14 years pottering around in remote places in Africa studying the social behaviour of mongooses. Her struggles to discover why meerkats play earned her a PhD in 2005, and she is currently obtaining lots of interesting, negative results from an habituated population of dwarf mongooses. You can read about her daily adventures at her blog Mainly Mongoose.

The views expressed are those of the author and are not necessarily those of Scientific American.