JavaScript disabled. Please enable JavaScript to use My News, My Clippings, My Comments and user settings.

If you have trouble accessing our login form below, you can go to our login page.

If you have trouble accessing our login form below, you can go to our login page.

So random: Why we persist in seeing streaks

Date

Carl Zimmer

From time to time, athletes get on a streak. Suddenly, the basketball goes through the net every time, or a batter gets a hit in every game. This blissful condition is often known as the hot hand, and players have come to believe it is real – so much so that they have made it a part of their strategy for winning games.

''On offence, if someone else has a hot hand, I constantly lay the ball on him,'' wrote NBA player Walt Frazier in his 1974 memoir Rockin’ Steady: A Guide to Basketball and Cool.

In the 1980s Thomas Gilovich, a psychologist at Cornell University, and his colleagues did a study of the hot hand. They confirmed that the vast majority of basketball players believed in it. The audiences at basketball games were also convinced. But then Dr Gilovich and his colleagues analysed the hot hand statistically, and it fell apart.

The hot hand was, they concluded, an illusion caused ''by a general misconception of chance''.

Today, there still isn’t much evidence for a hot hand in basketball or beyond. But our belief in it is unquestionably real. Roulette players will bet on more numbers after they win than after they lose, psychologists have found. A store that issues a winning lottery ticket will tend to sell more lottery tickets afterward, economists have observed. Investors often assume that a rising stock’s price will keep rising.

Time and again, we don’t want to believe that streaks can be the result of pure chance – probably because the bias appears to be deeply ingrained in our minds, researchers say. Indeed, a new study in The Journal of Experimental Psychology: Animal Learning and Cognition suggests that the hot hand phenomenon is so ancient that monkeys display it, too.

''What it suggests is that there’s something going back at least 25 million years,'' said Benjamin Hayden, a neuroscientist at the University of Rochester who wrote the study with his graduate student, Tommy Blanchard, and a psychologist at Clarkson University, Andreas Wilke.

The new study builds on earlier tests that Dr Wilke and his colleagues carried out on people. In one such study, the scientists had volunteers play a computer game that showed a picture of either a pear or a bunch of cherries. The volunteers had to guess which fruit would appear next.

The order was random, and yet the volunteers tended to guess that the next fruit would be the same as the current one. In other words, they expected the fruit to arrive in streaks.

In another trial, Dr Wilke and his colleagues let volunteers choose among different versions of the game so that they could increase their winnings. One version of the game was more likely to switch the fruit each time. As a result, the game had fewer streaks.

It should have been an easy game to win. All Dr Wilke’s volunteers needed to do was guess that the next fruit would be different each time. And yet the volunteers tended to avoid the alternating game in favor of the random game, ''where they see a pattern that doesn’t exist,'', Dr Wilke said.

Dr Wilke and his colleagues argue that this mental quirk is a side effect of how our brains have evolved.

''Our idea is that the driving force of the hot hand phenomenon was our history of foraging,'' Dr Wilke said.

Our ancestors were constantly searching for food, either gathering plants or hunting animals. As they searched, they had to continually decide where to look next. The wrong choice could mean starvation.

Dr Wilke argues that this threat led our ancestors to evolve some rules of thumb based on the fact that animals and plants aren’t scattered randomly across a landscape. Instead, they can be found in clumps.

That meant if our ancestors picked up a fruit from the ground, they were likely to find more by looking nearby, rather than going somewhere else. As a result, they became very sensitive to these streaks. They were an indication that good fortune would keep coming.

On the other hand, if our ancestors kept looking in a place for food and found nothing, they could predict that another look wouldn’t yield anything to eat.

In the modern world, Dr Wilke argues, we can’t get rid of this instinct to think that streaks will continue, even when we’re dealing with random patterns.

To vet this hypothesis, Dr Wilke has collaborated with H. Clark Barrett of the University of California, Los Angeles, to give his hot hand tests to a group of people who live deep in the Amazon rainforest and depend in part on hunting and foraging for their food. They had the same bias found in American volunteers, suggesting that the hot hand phenomenon went beyond Western societies obsessed with basketball and slot machines.

Dr Wilke’s latest experiment sought to test whether the hot hand bias was even more universal.

''The strongest test to see if it’s evolutionary is to find it in another species,'' said Dr Hayden, who studies how monkeys make decisions.

So he and Dr Wilke developed a game for monkeys to play. In each round, the monkeys saw either a purple rectangle on the left side of a computer screen or a blue rectangle on the right. In order to get a reward, the monkeys had to guess which rectangle would appear next, directing their gaze to the left or right side of the screen.

The monkeys played thousands of rounds, developing a strategy to get the biggest reward they could. And their performance revealed that they have a hot hand bias in their decision-making, researchers said.

When streaks were common, the scientists found, the monkeys learned to get a high score. In other versions of the game, with fewer streaks, they did worse. They couldn’t help guessing that a new rectangle would be the same as the previous one.

Dr Barrett, who was not involved in the new study, cautioned that the results need to be replicated. Nevertheless, he agreed that it raised the possibility that foraging gave rise to the hot hand phenomenon millions of years ago.

''This may be a deep evolutionary history indeed, stretching back to before we were human,'' Dr Barrett said.

Similarly, Laurie Santos, a psychologist at Yale, said, ''They’re on to something.'' But she questioned whether the new study actually showed that monkeys experienced the same feeling as a basketball player on a streak.

Just because the monkeys expected the rectangles to come in streaks didn’t mean that they believed their own actions had anything to do with it. Some psychologists distinguish between these two effects, calling them ''hot hands'' and ''hot outcomes''.

Dr Santos was confident that Dr Hayden and his colleagues could devise another experiment to test the two alternatives. ''That’s harder, but Ben does all kinds of crazy things,'' Dr Santos said. ''I’m sure he could find a way to do it.''

Dr Wilke said that hot hands and hot outcomes both could have evolved from the same underlying rules of thumb for searching for food. By understanding their origins, we may be able to better understand the particular ways they influence our thinking today.

In the July issue of Evolution and Human Behavior, Dr Wilke and his colleagues report that habitual gamblers have a stronger hot hand bias than non-gamblers. It might eventually be possible to predict who will be at risk of problem gambling by measuring hot hand bias in advance.

''That’s a first step into some sort of application of this research,'' Dr Wilke said.

New York Times

Featured advertisers