A&O READING – Anthropomorhism: “In Our Own Image”




(Douglas Fox 2010)



IT WAS a classic open-and-shut case. As Jehan Martin slept in his bed an intruder crept in, killed him and mutilated his body. Witnesses had seen a female enter the house on the day of the attack. She was subsequently taken into custody and tried in court. The trouble was the perpetrator happened to be a pig.

Historical records document at least 200 trials in which an animal was the principal defendant. So this case in 1457 is by no means an isolated incident. The accused were often assigned lawyers and confined to jail during trials. Occasionally they were acquitted – a donkey on trial for lewd sexual acts, for example, was freed after loyal supporters testified that she was “in all her habits of life a most honest creature”. When these animals were found guilty, however, they were usually hanged like a human criminal.

As rational, educated people, it’s easy to smirk at attempts to try animals in a court of law – but one should not be too hasty. After all, the people involved were falling prey to an irrational trait that afflicts us all from time to time: they were anthropomorphising.

Maybe you talk to your plants, name your car, or shout at your computer from time to time. Or perhaps you believe in a personified God. “We are hard-wired to see human-like beings everywhere,” says Stewart Guthrie, an anthropologist at Fordham University in New York City who has documented the rampant anthropomorphism in the world’s religions.

Despite its prevalence in human life, this bizarre trait had largely been ignored by credible scientists until very recently. Now the study of anthropomorphism is booming, with new work revealing how and why our brains are compelled to do it. It turns out the phenomenon has far-reaching effects on human behaviour, explaining why gamblers or traders on the stock market are inclined to push their luck too far, for example. And the work might just help engineers design a variation of Microsoft’s office assistant “Clippy” that you don’t want to strangle.

There’s no doubt that anthropomorphism is ingrained in human nature. Some of the oldest known pieces of cave art show figures who are half-human, half-animal, suggesting the trait may have been present in our ancestors at least 30,000 years ago. Since then, anthropomorphic figures have been ubiquitous in folk-lore and religion, and many of them are still going strong. Think Jack Frost, Mother Nature and, of course, God.

Ancient civilisations were well aware of this strange quirk of human psychology. Xenophanes, a philosopher in ancient Greece, coined the term anthropomorphism 2600 years ago. He observed that people worshipped gods that resemble themselves: Greeks kowtowed to white-skinned gods, while the Ethiopians preferred theirs a bit darker. From this observation, he predicted that if horses and donkeys believed in gods, theirs would trot on four legs. He may have had a point. Primatologists have documented a curious behaviour in chimpanzees, called the “rain dance”: when a thunderstorm blows in, they sometimes climb trees, tear off branches, and brandish them while screaming at the clouds – as if confronting a rival male. Those primates may well be “chimpomorphising” the storm – shaking their sticks at the hairy-knuckled Zeus who is hurling lightning bolts from the great treetop in the sky.

“Apes ‘chimpomorphise’ a storm when they shake their sticks at a hairy-knuckled Zeus in the sky”

Anthropomorphism had been somewhat overlooked by modern psychologists, until about 10 or so years ago, when a couple of trends in psychological research finally put the trait under the spotlight. For one thing, the psychology of religion has become increasingly fashionable, leading some researchers to question why most faiths feature human-like gods. And the growing popularity of avatars in virtual environments like Second Life has prompted greater interest in the ways in which we interact with non-human entities.

First of all, psychologists studying the phenomena wanted to establish exactly what is going on in our brain when we anthropomorphise. Nicholas Epley, a social psychologist at the University of Chicago’s Booth School of Business, had some ideas. He had studied our tendency to think egocentrically, meaning that we use our own preferences to predict how someone else will react to a dirty joke or a present. “It’s a well-known contributor to buying bad gifts,” says Epley. The next logical step was to question whether we also use our mind as a starting point for divining the “thoughts” of non-humans too. So in 2004, Epley and John Cacioppo, also at the University of Chicago, decided to test the idea.

The duo asked a group of volunteers to think about their own beliefs, other peoples’ beliefs, and God’s beliefs on issues like capital punishment, while the researchers viewed their brain activity with a functional MRI scanner. As you might expect, given Epley’s earlier work, the brain activity was pretty similar when the subjects considered their own or another human’s views. But the closest resemblance came when the subjects thought about God’s views – here, the brain activity was virtually identical to the scans taken when the subjects thought about their own viewpoint. This was reflected in their own reports, when they told the researchers that they considered their own beliefs to be much closer to God’s than to those of Bill Gates or George W. Bush, for example.

The results, published last December in Proceedings of the National Academy of Sciences (vol 106, p 21533), might simply confirm that some people use the idea of God to elevate their own beliefs, titillating themselves with the knowledge that those who disagree with them will spend eternity burning in hell. To Epley it signified something more profound: the less evidence we have of another’s beliefs – and for God we have very little indeed – the more likely we are to project our own beliefs into the voids. The same probably applies to anything else to which we attribute a personality or mind of its own, such as our car, our plant or our pet.

Further evidence that we use the same neural processes to understand the behaviour and minds of both humans and anthropomorphic entities came with brain scans by Christian Keysers, a social neuroscientist at the University Medical Center Groningen in the Netherlands. Under an fMRI scanner, his volunteers viewed two movies – one in which two people pushed or chased each other around, and another in which the people were replaced with a circle and square. Both movies activated the “mirror neuron system” in the viewers’ premotor and somatosensory cortices – brain areas that respond to action by other humans and help us interact socially. “The brain areas activated were virtually indistinguishable,” says Keysers. “People found that these geometric shapes were pushing each other around very intentionally.”

Bolstered by this convincing evidence that our brains really do consider inanimate objects in the same ways as they consider other human beings, the researchers decided to investigate the reasons why we might have evolved the trait in the first place.

One of the most obvious explanations is that it’s an attempt to make sense of a largely meaningless world. Humans have a habit of looking for useful cues in nature, even when they are not there, since the pay-off is huge in the few cases when there is cause for concern. We may be on guard whenever we hear rustling in the bushes, for example, even though the sound may be insignificant 99 times out of 100, simply because it might save our life in the one instance that it really does signal a predator intent on eating us. Many superstitions appear to be an extension of this behaviour (New Scientist, 4 February 2009, p 30) and anthropomorphism may be no different. “I wouldn’t say that the anthropomorphism itself is adaptive,” says Guthrie. “It’s mistaken by definition. But the strategy that leads to it is totally adaptive.”

Predicting the unpredictable

Consistent with this hypothesis, Cacioppo, Epley, and their former PhD student Adam Waytz, now at Harvard University, recently found that we are more likely to anthropomorphise when faced with unpredictable situations or entities. For example, their subjects were more likely to assign intentions, consciousness and emotions to unpredictable devices like Clocky, an alarm clock that wheels itself onto the floor and runs away as the alarm goes off, making it difficult to catch and press the snooze button. fMRI scans supported this finding, showing that thinking about unpredictable gadgets like Clocky leads to greater activation of the ventromedial prefrontal cortex – a brain area known to be involved in thinking about other peoples’ thoughts (Journal of Personality and Social Psychology, vol 99, p 410). This tendency even runs to desktop computers, with the team finding that we are more likely to consider that our PC has a mind of its own if it freezes up unexpectedly.

It’s also evident in the way we react to natural disasters. “When there’s human suffering or an earthquake,” says Waytz, “it can provide people with a sense of meaning to attribute them to the intentions of God or Mother Nature.” And in the case of those townspeople in 15th-century France, putting a pig on trial for murdering a child may have provided them with a sense of control by imposing human standards of behaviour on an amoral animal kingdom.

Our attempts to explain the unexplainable are unlikely to be the whole story, though. Anecdotal reports suggest that lonely people anthropomorphise more than those with a buzzing social life, leading Cacioppo and Epley to wonder whether it might also be a coping mechanism to deal with social isolation. “If you can’t connect to people, [maybe] you connect inferentially to dogs or gadgets or gods,” says Cacioppo. The idea is certainly fixed in pop culture. Think of the movie Cast Away, in which Tom Hanks’s character, stranded on a desert island, draws a human face on a volleyball and names it “Wilson”. People have even been known, in rare cases, to fall in love with and “marry” landmarks such as the Eiffel Tower or the Berlin Wall; one could speculate that they’re looking for intimate connections, albeit in a maladapted fashion.

Cacioppo, Epley, and Waytz set out to test the loneliness hypothesis. They presented subjects with descriptions of consumer gadgets such as Pillow Mate (a pillow that hugs you back) and Clocky, and asked people to rate them on human qualities such as “has a mind of its own”. As expected, those who saw the most human-like traits in Clocky and its ilk were the ones who showed greater signs of loneliness in personality surveys.

In a further experiment, they showed volunteers several movie clips, including a scene from Cast Away depicting loneliness, and then asked them to describe God or their pet. People who had just watched the lonely scene rated their pets and God as higher on supportive, person-like traits such as “thoughtful”, “considerate”, or “sympathetic” compared with people who watched other scenes.

“These are not people whom I would clinically diagnose as being upset,” says Waytz, who published the results with Epley and Cacioppo in Psychological Science in 2008 (vol 19, p 114). “What it says is that anthropomorphism is a very natural response to being just temporarily isolated.”

At first glance, this could seem to be somewhat damaging, since finding kinship in your car might keep you from seeking out real friends when you need them. “It’s a little bit like eating celery when you’re hungry,” says Cacioppo. “There’s nothing nourishing in it.” But when you consider the negative biological effects of loneliness, there could also be a very real upside to blunting these bad feelings. Social isolation shortens lifespan in humans and fruit flies alike, and studies of prisoners in solitary confinement demonstrate that our social brain can unravel over weeks and months, leading to long-term difficulties in building relationships afterwards. Maybe anthropomorphism is one of the brain’s efforts to minimise these effects. “It allows people to sort of tread water while searching for that real sense of connection,” says Waytz. “In that sense it’s certainly adaptive to reduce stress.”

All of which shows that anthropomorphism is an innate characteristic of human psychology, suggesting it could have far-reaching and unexpected effects on our behaviour. Beyond the obvious superstitions and quirks, however, few of these effects had been documented, so the researchers set about examining the possible situations in which we might anthropomorphise, and its consequences.

Our interactions with technology turned out to be one of the most promising lines of investigation. Engineers have long assumed that we prefer devices if they resemble humans, and there is some evidence to support this view. Combining human-like looks with unpredictability, for instance, turns out to be a potent mix for anthropomorphism that makes robots particularly engaging. In one study, toddlers went wild over a walking and dancing robot when it was difficult to tell what the device would do next. Significantly, they treated it like a human playmate – watching it, touching it frequently, and putting a blanket over it when it laid down as though it were sleeping. When the robot was reprogrammed to repeat its acts at regular intervals, however, it lost its anthropomorphic allure, and the children largely ignored it (Proceedings of the National Academy of Science, vol 104, p 17954).

Yet there is now a growing realisation that anthropomorphised devices also have their pitfalls. People often feel disproportionately angry, for example, when their computer freezes, hitting or shouting at the device. This irrational anger arises because people feel that they’ve established a partnership with the machine, says Clifford Nass, a psychologist of human-computer interaction at Stanford University in California. “We feel like it should be on our side,” he says. “It’s a betrayal of our trust.” Similar factors might also explain the demise of Clippy, the infamous animated assistant in early versions of Microsoft Office (see “R.I.P. Clippy”).

The best approach, says Nass, is to create an anthropomorphic interface with a bit of tact and diplomacy. He recently studied how customers interacted with Amazon.com‘s telephone voice-recognition system. These systems can be intensely irritating when they don’t recognise what we say, but people responded best when the computer blamed a crackly phone line rather than itself or the caller. The customers were more willing to repeat their answer, stay on the phone and buy books. It worked, says Nass, because the response implied the computer was trying hard, without making it seem too stupid.

Lucky streaks

Away from technology, anthropomorphic figures can make us behave ourselves when no one else is looking (see “Someone to watch over me”). Anthropomorphism can also determine, to a certain extent, our views on environmental matters, abortion or animal rights – all depending on the kinds of characteristics we bestow on the beings in question.

The phenomenon is especially rampant in high-risk, unpredictable situations like gambling or trading on the stock market, precisely the time when we would like to keep our wits about us. Michael Morris, a social psychologist at Columbia University’s business school in New York, recently examined the way that television commentators use volitional terms to describe the stock market – they might say the prices “climbed higher” or “flirted with the 2000 mark”. Reviews of TV transcripts showed that commentators were more likely to use volitional terms if the market was on a streak-that is, if it had just fallen or risen substantially (Organizational Behavior and Human Decision Processes, vol 102, p 174). This kind of subtle anthropomorphism can be dangerous for investors. Waytz, Epley, and their Booth colleague Eugene Caruso found that people who see volition behind a streak (say, a steadily rising stock or a string of red numbers in roulette) are more likely to believe that the streak will continue (Cognition, vol 116, p 149). By causing people to believe that a stock’s price will keep rising, this dynamic can perversely push investors to buy when the price is high rather than when it’s low.

“Anthropomophism is especially rampant in high-risk situations like gambling”

To make matters worse, these effects could have their biggest impact when we are already feeling particularly cocky, as Ann McGill, also at the Booth School of Business recently discovered when she presented some volunteers with two groups of slot machines – half of which looked somewhat human-like.

Those who felt more socially powerful were more likely to play with the anthropomorphised slot machines than those who felt shyer. “Nobody would look at a slot machine and say ‘I can talk it into letting me win’,” says McGill. “But if they’re feeling kind of powerful… they may start leaning toward the direction of thinking ‘I can probably get my way’.” This dynamic could embolden a trader or gambler to feel that they have more control than they really do – leading them to take excessive risks.

The biggest irony, however, emerges when you compare the trait with the way that humans often treat each other. Since lonely people seem more likely to see human qualities in inanimate objects, for instance, Waytz and his colleagues wondered if the opposite would be true: do socially connected people fail to see the humanity in real people? His early experiments have found exactly that.

People who had just recalled a family holiday, making them feel more socially secure, were more prone to endorse harsh interrogation techniques such as waterboarding and electric shock than other people. Since uncertainty also seems to trigger anthropomorphism, he predicts that feelings of power and security might also make us see other people as objects rather than human beings. “Both of those psychological factors put people in a mindset where they’re licensed to dehumanise [others],” says Waytz.

Maybe we’re no more rational than those pig-arresting townspeople in 15th-century France after all.



R.I.P. Clippy

Attempts to anthropomorphise technology backfired spectacularly in Clippy, the puppy-eyed paper clip that used to pop up sporadically in Microsoft Office to say things like: “It looks like you’re writing a letter – can I help you?” Radio skits portrayed Clippy being led into the woods for a gang-style execution as he cheerfully asked, “It looks like you’re digging a grave – is this a personal grave or a business grave?”

We disliked Clippy because it seemed to be a busybody, and made us feel less in control. Worse still, it popped up uninvited. This unpredictability may have made Clippy more human-like but it’s a bad trait when all you want to do is type up a report so you can leave the office and catch a train home. “Results showed again and again that the distraction was huge,” says Ben Shneiderman, a computer scientist at the University of Maryland in College Park. “The users’ performance was worse in that it took them longer to complete the task.”

Even on the rare occasions that Clippy did provide useful advice, it backfired in another way: users tended to credit the computer for their success, rather than themselves, reducing their personal satisfaction (Interacting with Computers, vol 19, p 293).


Someone to watch over me

We know that humans are more likely to behave themselves under supervision – but what if the supervisor has no mind of its own? Researchers at the University of California in Los Angeles studied people’s behaviour in an anonymous economic game under two situations, one with a normal computer screen and another with two cartoon eyes staring out of the screen. When the eyes were present, people showed more generosity to other players – even though they were still completely anonymous (Evolution and Human Behavior, vol 26, p 245). Most people care so deeply about what others think of their actions that the mere appearance of a face can evoke that feeling and influence behaviour, says Waytz, as if they really were being watched.

This effect has been put to powerful effect throughout history, with statues of Jesus, Buddha and the Virgin Mary appearing in countless public places, and the faces of rulers like Chairman Mao, Kim Jong Il, or Saddam Hussein plastered on the sides of buildings. “Good propaganda really takes advantage of the fact that humans are sensitive to what others are thinking,” says Waytz.