National Geographic

Elephants give each other a helping trunk

In Lampang, Thailand, two elephants have a problem. They’ve walked into adjacent paddocks separated by a fence. In front of them is a sliding table with two food bowls, but it’s out of reach and the way is barred by a stiff net. A rope has been looped around the table and one end snakes into each of the paddocks. If either jumbo tugs on the rope individually, the entire length will simply whip round into its paddock, depriving both of them of food. This job requires teamwork.

And the elephants know it. Joshua Plotnik from Emory University has shown that when confronted by this challenge, elephants learn to coordinate with their partners. They eventually pull on the rope ends together to drag the table towards them. They even knew to wait for their partner if they were a little late. It’s yet more evidence that these giant animals have keen intellects that rival those of chimps and other mental heavyweights.

There are, of course, many reasons to think that elephants are highly intelligent. They have large brains and they live in complex social groups. They can recognise themselves in mirrors, manipulate objects with their trunks, and smell the differences between human ethnic groups.  They’re interested in their dying and dead, they help stuck or distressed individuals, and they babysit each others’ calves.

But very few people have ever tested their intelligence in experiments, in the ways that primates, crows and dolphins have. Why? As Plotnik puts it, “This void in knowledge is mainly due to the danger and difficulty of submitting the largest land animal to behavioural experiments.” In short, working with a 4-tonne lump of muscle, tusk and trunk poses challenges that scientists don’t face when they work with a 70-kilogram chimp.

At the Thai Elephant Conservation Center, Plotnik saw a way around these problems. Each of the Center’s Asian elephants forms a bond with a “mahout” – a human who cares for their needs. With their help, Plotnik put six pairs of elephants through the rope-pulling task, a classic experiment that had been devised for chimpanzees in the 1930s.

When the mahouts released the elephant pairs at the same time, they eventually pulled the rope together. Even when one individual was released first, they soon learned that tugging on the rope themselves was futile. By the second day of training, they almost always waited for their partner to arrive before pulling, even if they had to wait for 45 seconds.

In a final experiment, Plotnik coiled one of the rope ends at the base of the table, so only one of the elephants could reach their end. There was no way they could pull the table forwards, and four out of five elephants realised this. When their partner couldn’t grab their end of the rope, they were far less likely to bother. Often, they just turned around and went back to their house.

This experiment shows that the elephants weren’t obeying a simple rule like “pull on the rope when my partner arrives” or “pull when I feel tension”.  They didn’t necessarily understand the mechanics of the rope and table, but they certainly knew that if their partner had walked away, there was no point in doing anything themselves.

Capuchin monkeys, hyenas and rooks (a type of crow) have all managed to learn to pull on the rope with a partner. But it’s not clear whether any of these animals understood their partner’s contribution to the task. “The rooks pulled together, but never waited for their partners,” says Plotnik. “This of course doesn’t mean corvids [crows and their kin – Ed] don’t cooperate. Personally, I think one big difference between corvids and elephants is patience.  Corvids are flighty, and waiting patiently for a partner’s arrival and inhibiting rope pulling until then may be very difficult for them. Elephants, on the other hand, are natural waiters. In the wild, elephants will often wait for other family members to “catch up” before moving on to other areas.”

The case for chimps is far stronger. In previous experiments, chimps have solved the task, even when they had to first let their partner into the room with the ropes. If the rope ends were close enough that a single chimp could pull the table in on its own, it never let its partner in. “It showed quite compelling evidence that the chimpanzees knew they needed a partner and that their partner needed to help,” says Plotnik.

The elephants’ success is equally compelling, even though their task was simpler. They clearly knew enough to wait for their partner and to abandon their end of the rope when their partner couldn’t reach theirs. “These results put elephants, at least in terms of how quickly they learn the critical contingencies of cooperation, on a par with apes,” says Plotnik.

There is a final twist to this tale. The numbers and graphs in Plotnik’s experiment only related to the elephants that solved the task as he intended. But two of them came up with their own solutions. One youngster called NU always stood on her end of the rope so that it didn’t yank away when her partner pulled on the other end. It was an altogether lazier strategy – NU got her food bowl while her partner did all the work!

Reference: Plotnik, Lair, Suphachokasahakun & de Waal. 2011. Elephants know when they need a helping trunk in a cooperative task. PNAS http://dx.doi.org/10.1073/pnas.1101765108

More on elephants:

There are 3 Comments. Add Yours.

  1. Trond Engen
    March 9, 2011

    Never mind the cooperating elephants. I’m interested in the non-people primates, crows and dolphins doing experiments on them.

  2. Messier Tidy Upper
    March 10, 2011

    Great (true) story & write up. :-)

    This reminds me of Stephen Baxter’s Mammoths trilogy – SF novels written from the point-of-view of those extinct animals & incl. intelligent mammoths on Mars! Sounds strange I know but really well written. :-)

    (Hope this isn’t too far off -topic.)

  3. Heather Spoonheim
    March 13, 2011

    There is a common second part to this experiment where only one food bowl is available as a reward. The dominant animal will of course take the food, but after that happens once or twice, subordinate chimpanzees will stop ‘cooperating’. Essentially this indicates the level of altruism to the cooperation. I would be interested in seeing how elephants responded to this, although I would prefer to watch by CCTV rather than be there first hand if they got upset.

Add Your Comments

All fields required.

Related Posts