I’ve gotten a lot of emails from people about my recent WSJ article, which looked at the superstar effect, choking and performance anxiety. Most of the letters ask a similar question:
Can anything be done to prevent choking? Or are we destined to sabotage ourselves when it matters the most?
Let’s begin with the bad news: choking, which I’ll define as “performing below skill level due to performance related anxieties,” is far more prevalent than most people assume. In recent years, choking has been shown to underlie everything from the achievement gap to the poor performance of girls on math tests. It afflicts professional golfers and public speakers, opera stars and creative problem solvers.
However, there are reliable ways to reduce the possibility of choking. In 2008, Daniel Gucciardi and James Dimmock, psychologists at the University of Western Australia, performed a study of 20 experienced golfers with handicaps ranging from zero to 12. The scientists had the golfers play under three separate conditions. In the first, they were told to fixate on specific components of their swing, such as “hips” or “straight wrist”. The second condition consisted of the golfers focusing on irrelevant words, such as “blue” or “white”. In the third, the golfers were told to focus on general aspects of their intended movement, or what the psychologists refer to as a “holistic cue word”. For instance, rather than contemplating the precise position of their wrist, they contemplated descriptive adjectives such as “smooth” or “balanced”. To make the experiment a bit more realistic, and to induce some anxiety, the scientists awarded a modest cash prize to the best golfer.
Gucciardi and Dimmock got two interesting results: the first was that anxiety only interfered with performance when it was coupled with self-consciousness. Nervous golfers who thought about the details of their swing, such as how to position their hips, hit consistently worse shots. That makes sense, since one of the main causes of choking is “thinking too much,” as we start analyzing actions (like a golf swing) that are best performed on autopilot.
The second interesting result was that there was a way to ward off choking. When the expert golfers contemplated a holistic cue word, their performance was no longer affected by anxiety. Because the positive adjectives were vague and generic, they didn’t cause the athletes to lose the flow of expert performance or overrule their automatic brain.
I’ve always mocked those silly motivational posters, which feature lofty adjectives like “courage” and “strength” and some inspiring photograph of a soaring eagle or snow capped mountain. (It didn’t help that my freshman roommate filled our dorm room with these posters.) But this research suggests that the pictures might actually work, at least to the extent they allow us to fixate on the cliche in capital letters. Thinking about “determination” won’t make us more determined, but it just might keep us from choking.
Sian Beilock, one of the researchers I wrote about in the WSJ article, has also led a really clever study that illuminates another way to prevent choking, at least when it comes to solving math problems. It turns out that that inner monologue of anxieties causing us to choke – this includes the golfer fretting about his swing and the student worrying about her algebra abilities – depends largely on the left prefrontal cortex. This makes sense, as the left prefrontal cortex has long been associated with working memory tasks that consist of verbal information. (Think, for instance, of a sample word problem on a math test.) Since our worries take the form of words, it makes perfect sense that they’d get stuck in the left PFC.
Interestingly, it turns out that simple differences in presentation can influence whether or not a math problem gets processed as a “verbal” problem or as a “spatial” problem. (More spatial problems, such as rotating an object in your mind, are more likely to be processed in the right PFC.) For instance, when an arithmetic problem is presented horizontally (84 – 18 = 66) it’s seen as more verbal, since we’re “reading” the numbers from left to right, just like a sentence. In contrast, when a problem is presented vertically it’s seen as more spatial:
-18
66
In other words, merely changing the presentation of the problem can dramatically alter how the brain processes the information. This allowed Beilock and colleagues to explore the effects of vertical and horizontal math problems on student performance. Her hypothesis was that horizontal problems, since they depend on the same brain area as verbal anxieties, might be more vulnerable to choking and pressure. After all, the same bit of brain can only contemplate so many things at once: if we’re worrying about our performance, there’s going to be less processing power left over to deal with the actual numbers.
The experiment went like this: a group of female undergraduates was given a series of simple math problems. Some students were given the problems in a horizontal presentation, while others were given them in a vertical fashion. Then, half of the female students were exposed to a stereotype about women and math ability. They were told, for instance, that “men outnumber women in math departments” and that “there is a good deal of evidence that men score higher on standardized tests of quantitative ability”. This was the stereotype threat condition, and it built on decades of research demonstrating that reminding people of a stereotype can decrease performance by increasing anxiety.
Sure enough, the activation of the stereotype led to decreased performance, but only on the horizontal problems. The reason has to do with the local processing differences of the PFC, and the fact that horizontal problems depended on the same cortical areas that were also worried about our math performance. In contrast, performance on vertical problems was unaffected, since our right prefrontal cortex wasn’t distracted by our anxieties or threatened by those untrue stereotypes.
The moral is that even minor tweaks can help reduce the performance problems caused by choking. Given the newfound importance of standardized tests – and this includes everything from high-school graduation exams to the SAT – we really need to better understand how the act of being testing warps the results. This is especially true for groups (women, minorities, etc.) that have been subject to various stereotypes. Sometimes, all it takes is a math problem presented in a different direction.