Monkey see, monkey do?

Stare at a chimp and he will earnestly stare back. Scratch your head and he will most likely scratch his. But don't be fooled, he really isn't all that smart

Chimpanzees look and behave very like us - a similarity approximate enough that advertisers have used those resemblances for years in adverts for tea, and film-makers for plots about exploited, human-like animals that children can relate to.

More academically, there are many who claim that chimps also share our ability to understand that others have intentions and beliefs. However, recent studies suggest that chimps comprehend the world very differently, leaving us quite alone in our talent.

The results have astounded other scientists as well as the experimenters themselves; but the result has been to generate new ideas to explain how social intelligence in humans may have evolved.

"If you've ever experienced a chimpanzee following your gaze, you know that it is nearly impossible to resist thinking it's trying to figure out what you're looking at," says Dr Daniel Povinelli of the University of Louisiana at Lafayette.

Like most people, Dr Povinelli began his research assuming that chimps were hairy human counterparts. He knew that they could count, acquire rudiments of language, deceive each other and recognise themselves in mirrors. Using their gaze-following habit, he aimed simply to prove that chimps understand that others have simple perceptual experiences, like seeing.

Fifteen experiments later, the chimps had changed his mind. He and his colleagues had given chimps a simple choice of two people from whom to gesture for food: one who could see them, and one who could not. One person, for example, covered her eyes with her hands, while the other blocked her ears. In another test, one person held a bucket over her head, and the other held it beside her. In an impressive failure, the chimps were just as likely to gesture to the person who couldn't see them as the one who could.

"I didn't believe it!" says Dr Povinelli. "The results just flew in the face of common sense and all our previous experience with these animals." With mannerisms and reactions so similar to ours, how could they fail to even understand the state of attention behind a person's gaze? A misunderstanding of the experimental set-up wasn't the problem; the eye-covering techniques were copied from the chimps' own games, and the gestures were their natural begging movements.

Further investigation revealed that the chimps' choices were based on a simple rule of thumb: "gesture to people facing me". When one person faced them with closed eyes, and the other faced away but looked over her shoulder with open eyes, chimps gestured to the person who couldn't see them.

The researchers were forced to conclude that chimps are unable to equate seeing with knowledge in the way that human children even as young as three can do.

However after missing out on their food reward often enough, the chimps soon learnt to choose the correct person. "Make no mistake," says Dr Povinelli. "Chimps are thinking, intelligent, cognitive organisms." And although they may not reason about the mental states underlying others' behaviour, they do reason about the behaviour itself.

This considerable ability, along with fast learning and a portion of instinct, explains reports of chimps deceiving each other and "making up" after fights. Such anecdotes have led many to assume that chimps reason about the mental processes of others, but they are misleading, believes Dr Povinelli.

Another group of researchers, from the University of Pennsylvania, was similarly persuaded about the ability of baboons. They had been impressed by how baboons seemed to deliberately inform each other of their whereabouts by calling whilst out of sight. Dr Drew Rendall and colleagues now think that their impression was wrong.

They tracked foraging baboons in Botswana, and established that mothers gave contact calls when separated from their babies or the group. They then observed the mothers' responses to contact calls given by their missing infants - both live calls, and pre-recorded calls played back by the researchers.

Like human voices, each baboon's call is distinct and recognisable. As some of the playback calls came from unrelated infants, the researchers could verify that mothers did indeed recognise their own babies' calls. They sometimes responded by waiting for their baby or looking for it, but surprisingly failed to call back in answer.

Why? The researchers concluded that mothers could not understand that their reply might influence the infant's perspective, or help reunite the pair. "They may simply call because they themselves are at risk of getting separated from the group and are concerned about it," says Dr Rendall. "Calling is almost certainly to maintain contact with others, but it's not clear that we should interpret it as a deliberate attempt to inform them."

Even in the absence of understanding others' minds, apparently sophisticated behaviours may still evolve, such as in birds. Plovers pretend to be injured when a predator approaches their nests. They adjust their display according to the predator's species, gait and behaviour, then fly away if it attacks.

Perhaps the bird reasons how a feigned broken wing might dupe a fox into thinking it easy prey and ignoring its eggs? Because birds are so different to humans, it is less tempting to make analogies between them and us. But the outstanding similarity between the natural spontaneous behaviour of chimps and humans needs addressing in the light of Dr Povinelli's experiments.

He proposes that, after our lineage separated from chimps, humans underwent a special type of psychological evolution which enabled us to perceive others' mental lives. This went hand-in-hand with the dramatic enlargement of our brain's pre-frontal cortex. Complex social behaviours were already in place before this evolution, and our new intelligence refined those behaviours without replacing them. As a result, the key difference between us and other primates lies not in our behaviour, but in our interpretation of others' behaviour.

Rather than setting up humans as the enlightened species, though, the theory highlights our more primitive tendencies. "I think this has more implications about how humans understand the world than how chimps do," says Dr Povinelli.

He describes our minds as a complex mosaic of ancient low and intermediate-level mechanisms interspersed with higher-level mechanisms. Consider how we can drive home without being aware of the journey, and the automatic way that we, too, follow someone's gaze. Much of our behaviour is still governed by low-level cognitive mechanisms, and sometimes we even perform better this way. Strangely, we cannot tell when our actions stem from this more primitive part of our minds, and tend to give high-level explanations for all our behaviours.

Both humans and chimps often operate above this automatic level. Chimps are far from being hairy robots, and Dr Povinelli warns against polarising the debate over their abilities. What they may not do, he believes, is go beyond reasoning about others' behaviour to reasoning about the mental state behind the behaviour. But this should not diminish them in anyone's eyes.

The investigations headed by Dr Povinelli and Dr Rendall only touch on small aspects of a vast area of psychological territory. Nevertheless, they provide compelling evidence for the basis of human uniqueness. Five million years ago, humans and chimps diverged from a common ancestor, and today 98.8 per cent of our genome is still identical. But in the light of this newer understanding, it is perhaps timely to acknowledge that we also share 50 per cent of it with bananas.