Being Wrong - Kathryn Schulz [61]
It was Donald Leka’s misfortune to illustrate one of these inductive biases: leaping to conclusions. Actually, as I noted above, leaping to conclusions is what we always do in inductive reasoning, but we generally only call it that when the process fails us—that is, when we leap to wrong conclusions. In those instances, our habit of relying on meager evidence, normally so clever, suddenly looks foolish (and makes us look foolish, too). That’s bad enough, but Don’s story also showcases a more specific and disturbing aspect of this bias. Since the whole point of inductive reasoning is to draw sweeping assumptions based on limited evidence, it is an outstanding machine for generating stereotypes. Think about the magnitude of the extrapolation involved in going from “This swan is white” to “All swans are white.” In context, it seems unproblematic, but now try this: “This Muslim is a terrorist” “All Muslims are terrorists.” Suddenly, induction doesn’t seem so benign.
If the stereotypes we generate based on small amounts of evidence could be overturned by equally small amounts of counterevidence, this particular feature of inductive reasoning wouldn’t be terribly worrisome. A counterexample or two would give the lie to false and pernicious generalizations, and we would amend or reject our beliefs accordingly. But this is the paradox of inductive reasoning: although small amounts of evidence are sufficient to make us draw conclusions, they are seldom sufficient to make us revise them.
Consider, for instance, a second story, this one told to me by a woman named Elizabeth O’Donovan. Once, many years ago, Elizabeth got into an argument with a friend about whether Orion is a winter constellation. (Because of the annual rotation of the earth around the sun, most constellations are visible only at certain times of the year.) Elizabeth emphatically insisted that it was not. “The embarrassing part,” she told me, “is that at the time that I was doing all this insisting, my friend and I were standing in a parking lot in December and I had just pointed to the sky and said, ‘That’s weird. Orion shouldn’t be out now; it’s a summer constellation.’”
You might think that gazing directly at evidence that contradicted her claim would have given Elizabeth pause, but it did not. Instead, the argument escalated until she bet her friend that he was wrong. The loser, they agreed, would take the winner out to dinner once a week for four weeks. “I was so damn determined,” she recalled, “that I figured it was some sort of crazy astronomical phenomenon. My logic was something like, ‘Well, everyone knows that every fifty-two years, Orion appears for eighteen months straight.’” As we’ll see later, this kind of tortured theorizing is typical of the crisis that ensues when new evidence challenges an entrenched theory. It’s also a decent indicator that you are about to lose a bet. (For the record, Orion is visible in the night sky from roughly October through March, in both the northern and southern hemispheres. Elizabeth treated her friend to four weeks of fried chicken.)
Elizabeth’s story illustrates another of our inductive biases. This one is famous enough to have earned its own separate name in the psychological literature: confirmation bias. As you might guess, confirmation bias is the tendency to give more weight to evidence that confirms our beliefs than to evidence that challenges them. On the face of it, this sounds irrational (and sometimes looks it, as Elizabeth unwittingly demonstrated). In fact, though, confirmation bias is often entirely sensible. We hold our beliefs for a reason, after all—specifically, because we’ve already encountered other, earlier evidence that suggests that they are true. And, although this, too, can seem pigheaded, it’s smart to put more stock in that earlier evidence than in whatever counterevidence we come across later. Remember how our beliefs are probabilistic? Probability theory tells us that the more common something is—long-necked giraffes, white swans, subject-verb-object sentences