In a high-pressure math test, the best students won't necessarily get the top scores, according to recent research.

### Transcript

**Why smart math students make dumb mistakes. I'm Bob Hirshon and this is Science Update. The best math students are also the most susceptible to choking under pressure. This according to the work of University of Chicago psychologist Sian Beilock. She reported new findings at the AAAS Annual Meeting in San Francisco. Her team found that top students rely on their superior working memory to solve complex problems. But average students often fall back on less accurate shortcuts, like estimation. Sian Beilock:**

Under pressure, however, our higher working-memory individuals didn't continue with the complex algorithm. They actually switched to the shortcut. So their performance looked like the lower working-memory individuals' under pressure, and now we have some idea why this is the case.

**She said that because pressure appears to sabotage working memory, high-stakes exams may not accurately identify the best and the brightest. I'm Bob Hirshon, for AAAS, the science society.**

### Making Sense of the Research

At some point or another, everyone has "choked"—in other words, performed more poorly than expected—whether it's on a math test or on a soccer field. Beilock has researched many aspects of choking. In this case, she wanted to understand how pressure affects math students differently, depending on their level of skill.

Her past research suggests that good math students do well in part because they have a higher working memory capacity. This means that they can juggle more numbers and math operations in their heads than the average person. In her experiments, she studied both high and average working-memory students in two different situations. In one, they were simply told that they were practicing math problems. In another, they were subjected to various kinds of pressure. They were offered cash prizes if they scored well, they were seated next to students who had supposedly scored well already, and they even were told they were being videotaped so that teachers could learn how to teach math.

If the added pressure had hurt everyone's scores equally, then in a sense, the pressure wouldn't matter. On a high-pressure test, the best students would still get the top scores; they just wouldn't score as high as they would in a low-pressure situation. Instead, Beilock found that the pressure significantly hurt the best students' scores, while it didn't really affect the scores of average students.

After asking the volunteers how they solved the math problems, she discovered a reason behind the difference. In the low-pressure situation, the students with more working memory were using that memory to solve many of the problems in their heads. The students with lower working memory, on the other hand, used rules of thumb and educated guesses to pick a multiple-choice answer, without actually going through the math involved. Those techniques were more effective than just picking answers at random, but not as reliable as doing the math. In high-pressure situations, however, the better students didn't use their working memory as well, and switched to the rules of thumb themselves. As a result, their scores dropped in high-pressure situations, while the average students' scores stayed the same.

The most effective solution to this problem would be to eliminate all forms of pressure from math tests. Obviously, that's impossible. However, Beilock's research suggests that high-pressure exams shouldn't be the only tool for evaluating a math student's abilities. She also suggests that the pressure of exams might be lessened by letting students practice the form of the exam, including time limits and physical surroundings, before taking one that really counts. This could make the important exams feel a little more ordinary, and a little less intimidating.

Now try and answer these questions:

- How are better math students affected more by pressure, according to this research?
- What role does working memory play in Beilock's findings?
- Do you agree with these findings, based on personal experience? Why or why not?
- What are the implications of this study for teachers? College admissions boards? Math competitions?
- Do you think these findings might apply to other skills besides math? Why or why not?

You may want to check out the February 23, 2007, Science Update Podcast to hear further information about this Science Update and the other programs for that week. **This podcast's topics include:** how addiction is like hunger, a new therapy that targets a virus's genes, the best math students perform the worst, pollution could contribute to obesity, and new immigrants face color and height biases.