The Intelligence Trap

by David Robson

  Unlike many psychological studies, which tend to use university students as guinea pigs, Bruine de Bruin’s experiment examined a diverse sample of people, aged eighteen to eighty-eight, with a range of educational backgrounds – allowing her to be sure that any results reflected the population as a whole.

  As Stanovich has found with his tests, the participants’ decision-making skills were only moderately linked to their intelligence; academic success did not necessarily make them more rational decision makers.

  But Bruine de Bruin then decided to see how both measures were related to their behaviours in the real world. To do so, she asked participants to declare how often they had experienced various stressful life events, from the relatively trivial (such as getting sunburnt or missing a flight), to the serious (catching an STD or cheating on your partner) and the downright awful (being put in jail).20 Although the measures of general intelligence did seem to have a small effect on these outcomes, the participants’ rationality scores were about three times more important in determining their behaviour.

  These tests clearly capture a more general tendency to be a careful, considered thinker that was not reflected in more standard measures of cognitive ability; you can be intelligent and irrational – as Stanovich had found – and this has serious consequences for your life.

  Bruine de Bruin’s findings can offer us some insights into other peculiar habits of intelligent people. One study from the London School of Economics, published in 2010, found that people with higher IQs tend to consume more alcohol and may be more likely to smoke or take illegal drugs, for instance – supporting the idea that intelligence does not necessarily help us to weigh up short-term benefits against the long-term consequences.21

  People with high IQs are also just as likely to face financial distress, such as missing mortgage payments, bankruptcy or credit card debt. Around 14 per cent of people with an IQ of 140 had reached their credit limit, compared to 8.3 per cent of people with an average IQ of 100. Nor were they any more likely to put money away in long-term investments or savings; their accumulated wealth each year was just a tiny fraction greater. These facts are particularly surprising, given that more intelligent (and better educated) people do tend to have more stable jobs with higher salaries, which suggests that their financial distress is a consequence of their decision making, rather than, say, a simple lack of earning power.22

  The researchers suggested that more intelligent people veer close to the ‘financial precipice’ in the belief that they will be better able to deal with the consequence afterwards. Whatever the reason, the results suggest that smarter people are not investing their money in the more rational manner that economists might anticipate; it is another sign that intelligence does not necessarily lead to better decision making.

  As one vivid example, consider the story of Paul Frampton. A brilliant physicist at the University of North Carolina, his work ranged from a new theory of dark matter (the mysterious, invisible mass holding our universe together) to the prediction of a subatomic particle called the ‘axigluon’, a theory that is inspiring experiments at the Large Hadron Collider.

  In 2011, however, he began online dating, and soon struck up a friendship with a former bikini model named Denise Milani. In January the next year, she invited him to visit her on a photoshoot in La Paz, Bolivia. When he arrived, however, he found a message – she’d had to leave for Argentina instead. But she’d left her bag. Could he pick it up and bring it to her?

  Alas, he arrived in Argentina but there was still no sign of Milani. Losing patience, he decided to return to the USA, where he checked in her suitcase with his own luggage. A few minutes later, an announcement called him to meet the airport staff at his gate. Unless you suffer from severe dysrationalia yourself, you can probably guess what happened next. He was subsequently charged with transporting two kilograms of cocaine.

  Fraudsters, it turned out, had been posing as Milani – who really is a model, but knew nothing of the scheme and had never been in touch with Frampton. They would have presumably intercepted the bag once he had carried it over the border.

  Frampton had been warned about the relationship. ‘I thought he was out of his mind, and I told him that,’ John Dixon, a fellow physicist and friend of Frampton’s, said in the New York Times. ‘But he really believed that he had a pretty young woman who wanted to marry him.’23

  We can’t really know what was going through Frampton’s mind. Perhaps he suspected that ‘Milani’ was involved in some kind of drug smuggling operation but thought that this was a way of proving himself to her. His love for her seems to have been real, though; he even tried to message her in prison, after the scam had been uncovered. For some reason, however, he just hadn’t been able to weigh up the risks, and had allowed himself to be swayed by impulsive, wishful thinking.

  If we return to that séance in Atlantic City, Arthur Conan Doyle’s behaviour would certainly seem to fit neatly with theories of dysrationalia, with compelling evidence that paranormal and superstitious beliefs are surprisingly common among the highly intelligent.

  According to a survey of more than 1,200 participants, people with college degrees are just as likely to endorse the existence of UFOs, and they were even more credulous of extrasensory perception and ‘psychic healing’ than people with a worse education.24 (The education level here is an imperfect measure of intelligence, but it gives a general idea that the abstract thinking and knowledge required to enter university does not translate into more rational beliefs.)

  Needless to say, all of the phenomena above have been repeatedly disproven by credible scientists – yet it seems that many smart people continue to hold on to them regardless. According to dual-process (fast/slow thinking) theories, this could just be down to cognitive miserliness. People who believe in the paranormal rely on their gut feelings and intuitions to think about the sources of their beliefs, rather than reasoning in an analytical, critical way.25

  This may be true for many people with vaguer, less well-defined beliefs, but there are some particular elements of Conan Doyle’s biography that suggest his behaviour can’t be explained quite so simply. Often, it seemed as if he was using analytical reasoning from system 2 to rationalise his opinions and dismiss the evidence. Rather than thinking too little, he was thinking too much.

  Consider how Conan Doyle was once infamously fooled by two schoolgirls. In 1917 – a few years before he met Houdini – sixteen-year-old Elsie Wright and nine-year-old Frances Griffith claimed to have photographed a population of fairies frolicking around a stream in Cottingley, West Yorkshire. Through a contact at the local Theosophical Society, the pictures eventually landed in Conan Doyle’s hands.

  Many of his acquaintances were highly sceptical, but he fell for the girls’ story hook, line and sinker.26 ‘It is hard for the mind to grasp what the ultimate results may be if we have actually proved the existence upon the surface of this planet of a population which may be as numerous as the human race,’ he wrote in The Coming of Fairies.27 In reality, they were cardboard cut-outs, taken from Princess Mary’s Giftbook28 – a volume that had also included some of Conan Doyle’s own writing.29

  What’s fascinating is not so much the fact that he fell for the fairies in the first place, but the extraordinary lengths that he went to explain away any doubts. If you look at the photographs carefully, you can even see hatpins holding one of the cut-outs together. But where others saw pins, he saw the gnome’s belly button – proof that fairies are linked to their mothers in the womb with an umbilical cord. Conan Doyle even tried to draw on modern scientific discoveries to explain the fairies’ existence, turning to electromagnetic theory to claim that they were ‘constructed in material which threw out shorter or longer vibrations’, rendering them invisible to humans.

  As Ray Hyman, a professor of psychology at the University of Oregon, puts it: ‘Conan Doyle used his intelligence and cleverness to dismiss all counter-arguments . . . [He] was able to use his smartness to out
smart himself.’30

  The use of system 2 ‘slow thinking’ to rationalise our beliefs even when they are wrong leads us to uncover the most important and pervasive form of the intelligence trap, with many disastrous consequences; it can explain not only the foolish ideas of people such as Conan Doyle, but also the huge divides in political opinion about issues such as gun crime and climate change.

  So what’s the scientific evidence?

  The first clues came from a series of classic studies from the 1970s and 1980s, when David Perkins of Harvard University asked students to consider a series of topical questions, such as: ‘Would a nuclear disarmament treaty reduce the likelihood of world war?’ A truly rational thinker should consider both sides of the argument, but Perkins found that more intelligent students were no more likely to consider any alternative points of view. Someone in favour of nuclear disarmament, for instance, might not explore the issue of trust: whether we could be sure that all countries would honour the agreement. Instead, they had simply used their abstract reasoning skills and factual knowledge to offer more elaborate justifications of their own point of view.31

  This tendency is sometimes called the confirmation bias, though several psychologists – including Perkins – prefer to use the more general term ‘myside bias’ to describe the many different kinds of tactics we may use to support our viewpoint and diminish alternative opinions. Even student lawyers, who are explicitly trained to consider the other side of a legal dispute, performed very poorly.

  Perkins later considered this to be one of his most important discoveries.32 ‘Thinking about the other side of the case is a perfect example of a good reasoning practice,’ he said. ‘Why, then, do student lawyers with high IQs and training in reasoning that includes anticipating the arguments of the opposition prove to be as subject to confirmation bias or myside bias, as it has been called, as anyone else? To ask such a question is to raise fundamental issues about conceptions of intelligence.’33

  Later studies only replicated this finding, and this one-sided way of thinking appears to be a particular problem for the issues that speak to our sense of identity. Scientists today use the term ‘motivated reasoning’ to describe this kind of emotionally charged, self-protective use of our minds. Besides the myside/confirmation bias that Perkins examined (where we preferentially seek and remember the information that confirms our view), motivated reasoning may also take the form of a disconfirmation bias – a kind of preferential scepticism that tears down alternative arguments. And, together, they can lead us to become more and more entrenched in our opinions.

  Consider an experiment by Dan Kahan at Yale Law School, which examined attitudes to gun control. He told his participants that a local government was trying to decide whether to ban firearms in public – and it was unsure whether this would increase or decrease crime rates. So they had collected data on cities with and without these bans, and on changes in crime over one year:

  Kahan also gave his participants a standard numeracy test, and questioned them on their political beliefs.

  Try it for yourself. Given this data, do the bans work?

  Kahan had deliberately engineered the numbers to be deceptive at first glance, suggesting a huge decrease in crime in the cities carrying the ban. To get to the correct answer, you need to consider the ratios, showing around 25 per cent of the cities with the ban had witnessed an increase in crime, compared with 16 per cent of those without a ban. The ban did not work, in other words.

  As you might hope, the more numerate participants were more likely to come to that conclusion – but only if they were more conservative, Republican voters who were already more likely to oppose gun control. If they were liberal, Democrat voters, the participants skipped the explicit calculation, and were more likely to go with their (incorrect) initial hunch that the ban had worked, no matter what their intelligence.

  In the name of fairness, Kahan also conducted the same experiment, but with the data reversed, so that the data supported the ban. Now, it was the numerate liberals who came to the right answer – and the numerate conservatives who were more likely to be wrong. Overall, the most numerate participants were around 45 per cent more likely to read the data correctly if it conformed to their expectations.

  The upshot, according to Kahan and other scientists studying motivated reasoning, is that smart people do not apply their superior intelligence fairly, but instead use it ‘opportunistically’ to promote their own interests and protect the beliefs that are most important to their identities. Intelligence can be a tool for propaganda rather than truth-seeking.34

  It’s a powerful finding, capable of explaining the enormous polarisation on issues such as climate change (see graph below).35 The scientific consensus is that carbon emissions from human sources are leading to global warming, and people with liberal politics are more likely to accept this message if they have better numeracy skills and basic scientific knowledge.36 That makes sense, since these people should also be more likely to understand the evidence. But among free-market capitalists, the opposite is true: the more scientifically literate and numerate they are, the more likely they are to reject the scientific consensus and to believe that claims of climate change have been exaggerated.

  The same polarisation can be seen for people’s views on vaccination,37 fracking38 and evolution.39 In each case, greater education and intelligence simply helps people to justify the beliefs that match their political, social or religious identity. (To be absolutely clear, overwhelming evidence shows that vaccines are safe and effective, carbon emissions are changing the climate, and evolution is true.)

  There is even some evidence that, thanks to motivated reasoning, exposure to the opposite point of view may actually backfire; not only do people reject the counter-arguments, but their own views become even more deeply entrenched as a result. In other words, an intelligent person with an inaccurate belief system may become more ignorant after having heard the actual facts. We could see this with Republicans’ opinions about Obamacare in 2009 and 2010: people with greater intelligence were more likely to believe claims that the new system would bring about Orwellian ‘death panels’ to decide who lived and died, and their views were only reinforced when they were presented with evidence that was meant to debunk the myths.40

  Kahan’s research has primarily examined the role of motivated reasoning in political decision making – where there may be no right or wrong answer – but he says it may stretch to other forms of belief. He points to a study by Jonathan Koehler, then at the University of Texas at Austin, who presented parapsychologists and sceptical scientists with data on two (fictional) experiments into extrasensory perception.

  The participants should have objectively measured the quality of the papers and the experimental design. But Koehler found that they often came to very different conclusions, depending on whether the results of the studies agreed or disagreed with their own beliefs in the paranormal.41

  When we consider the power of motivated reasoning, Conan Doyle’s belief in fraudulent mediums seems less paradoxical. His very identity had come to rest on his experiments with the paranormal. Spiritualism was the foundation of his relationship with his wife, and many of his friendships; he had invested substantial sums of money in a spiritualist church42 and written more than twenty books and pamphlets on the subject. Approaching old age, his beliefs also provided him with the comforting certainty of the afterlife. ‘It absolutely removes all fear of death,’ he said, and the belief connected him with those he had already lost43 ? surely two of the strongest motivations imaginable.

  All of this would seem to chime with research showing that beliefs may first arise from emotional needs – and it is only afterwards that the intellect kicks in to rationalise the feelings, however bizarre they may be.

  Conan Doyle certainly claimed to be objective. ‘In these 41 years, I never lost any opportunity of reading and studying and experimenting on this matter,’44 he boasted towards the end of his life. But he
was only looking for the evidence that supported his point of view, while dismissing everything else.45

  It did not matter that this was the mind that created Sherlock Holmes – the ‘perfect reasoning and observing machine’. Thanks to motivated reasoning, Conan Doyle could simply draw on that same creativity to explain away Houdini’s scepticism. And when he saw the photos of the Cottingley Fairies, he felt he had found the proof that would convince the world of other psychic phenomena. In his excitement, his mind engineered elaborate scientific explanations – without seriously questioning whether it was just a schoolgirl joke.

  When they confessed decades after Conan Doyle’s death, the girls revealed that they simply hadn’t bargained for grown-ups’ desire to be fooled. ‘I never even thought of it as being a fraud,’ one of the girls, Frances Griffiths, revealed in a 1985 interview. ‘It was just Elsie and I having a bit of fun and I can’t understand to this day why they were taken in – they wanted to be taken in.’46

  Following their increasingly public disagreement, Houdini lost all respect for Conan Doyle; he had started the friendship believing that the writer was an ‘intellectual giant’ and ended it by writing that ‘one must be half-witted to believe some of these things’. But given what we know about motivated reasoning, the very opposite may be true: only an intellectual giant could be capable of believing such things.*

  * In his book The Rationality Quotient, Keith Stanovich points out that George Orwell famously came to much the same conclusion when describing various forms of nationalism, Orwell writing that: ‘There is no limit to the follies that can be swallowed if one is under the influence of feelings of this kind . . . One has to belong to the intelligentsia to believe things like that: No ordinary man could be such a fool.’