How math anxiety affects performance (#PERC2013)

by Stephanie Chasteen on July 24, 2013

My other favorite talk at AAPT/PERC was by Sian Beilock (University of Chicago, Psychology), titled “Academic Performance under stress.”  Who would have guessed that from such an innocuous title would spring an intensely interesting, well-researched, sparklingly-clear exposition.  It is so refreshing to find a speaker who has clearly worked hard to communicate her field, and I’m sure that her oratory skills have increased the impact of her work!

She speaks fast, so this will be a long post!


Math anxiety

Math anxiety is a big thing in our culture.  Many people will brag, “I’m not a math person.”  Dr. Beilock studies math anxiety, as well as how peoples’ awareness of stereotypes affects how they perform (“stereotype threat”), particularly in the area of gender and math.

So what she does is create stressful situations.  “I tell math-anxious people that math teachers will be watching them solve problems,” she joked.  She tell them that their performance on a math test will affect whether other people will be able to keep money that they won.  She then studies how that stress affects their performance.

Math anxiety doesn’t just happen in middle school and high school.  She’s found that students can have math anxiety in the first grade! She found that most children are moderately anxious about math.   Did those with math anxiety have lower math achievement?  Yup.  But what direction is the causality — does the poor performance cause anxiety or vice-versa?  We’ll get to that later (spoiler:  the anxiety can cause poor performance).  She also showed that math anxiety and reading anxiety are decoupled; they appear to be different forms of academic anxiety.

Is your teacher math-anxious?

She also studied math anxiety by major.  Who do you think would have the highest math anxiety?  Yep, elementary education majors.  This, of course, is a problem, since these are the people who will be teaching math to our children.  And, of course, elementary ed majors are predominantly women.

She did a study to assess the math anxiety of elementary teachers, and of their students.  She found that, by the end of the school year, the higher the teacher’s math anxiety, the lower the girls’ math achievement, even accounting for prior performance.  The same effect was not true of the boys.  Yikes!

She also assessed students’ stereotypes of “people good at math” and “people good at reading.”  She told students a story about a kid who was good at math, and a kid who was good at reading, and then asked them to draw pictures of those kids.  The higher the student’s math anxiety, the more likely they were to draw boys that were good at math and girls who were good at reading, fulfilling those gender stereotypes.

So, the higher the math anxiety of the teacher, the more likely her students would endorse gender stereotypes.  And the more that the girls endorsed the stereotypes, the worse they were at math.  Boys’ abilities, however, didn’t relate to their tendency to stereotype.  Girls who didn’t endorse that stereotype performed on-par with the boys, but girls who did endorse the stereotype do rose.  And, of course, they found that the attitudes of their parents matter as well.

Does performance cause anxiety?

But what about causal direction?  People often thought that if you’re bad at math, you’re anxious about it.  While it’s true that if you’re bad at math, you’re likely to stay away from it, which will make you overall perform worse on math tests.  But the anxiety also robs you of the ability to succeed because it takes away cognitive resources.

Dr. Beilock tried a cool experiment where she taught subjects a new mathematical technique (modular arithmetic), so that they could control prior knowledge and training about the particular mathematical technique.  It’s known that if you’re in a stressful situation, you’ll do worse on problems requiring your frontal lobe.  So, she gave subjects easy (low working memory) and hard (high working memory) problems, where they had to hold fewer or more numbers in their head to do the problem.  When putting subjects under stress, she should see an effect on the problems requiring high working memory.

She used a stereotype threat situation, with women.  All she did was to mention, “we’re studying gender effects on math.”  It’s not a big intervention, just one sentence.  She found that this intervention had no effect on the easy problems, but those under the stereotype threat condition had much worse performance.  So, math anxiety can impact performance:  It’s not so much about what we know, but that stress puts a load on our limited cognitive horsepower.

What’s happening in the brain?

She has done some functional MRI (fMRI) studies, which allow you to measure neural activity because when blood is being shunted to a neuron, it has different properties when it’s carrying nutrients to an active brain cell.  (Which is also the nicest short explanation of fMRI I’ve heard yet).

She had subjects do math tasks and word tasks while in the MRI tube.  She told them in advance through a visual cue which type of task was coming next — math or word. What happens when they are cued that a math task is coming up?  I missed whether this result was only for math anxious subjects, but I assume that it was, because she was looking for negative anticipation in these subjects.  She saw that the areas of the brain that lit up were those that are associated with physical pain.  So, wow.  She also found that this pain-center activation resulted in less activation of the areas that are needed for actually solving the problem.

Why would the pain centers be activated?  She hypothesizes that mathematics is a relatively new development.  So what the brain did was to co-opt an area of the brain used for pain to deal with this new type of anxiety.  And the brain doesn’t know the difference between real pain and this non-physical anxiety.

How can we help?

Luckily, not all is lost.  She has found that we can help people get around stereotype threat by asking them to reflect on their successes in their career or major so far.

She has also done several studies on journaling, and found that this has many benefits.  If students journal about their worries before a test, then their performance is less impacted by their worries. Her prompt for students was quite long, but here is a truncated version:  “Write for 10 minutes about your thoughts and feelings about the exam you are about to take.  Please try to be as open as possible, really let yourself go.”

She found that this intervention didn’t affect those with low levels of anxiety, but there was a large effect (10 points on the exam) for those with high levels of anxiety. (Ramirez and Beilock, 2011, Science).

This is a way to help students put their best foot forward, she urges.


Dr. Beilock has written a book, Choke: What the Secrets of the Brain Reveal About Getting It Right When You Have To, which I’ll likely be buying.  Interesting stuff!


Sian L. Beilock is a professor in the Department of Psychology at The University of Chicago.  Her research program sits at the intersection of cognitive science and education.  She explores the cognitive and neural substrates of skill learning as well as the mechanisms by which performance breaks down in high-stress or high-pressure situations.  Dr. Beilock’s research is funded by the National Science Foundation and the U.S. Department of Education (Institute of Education Sciences). 

Image by Marinda Fowler Photography, via Flickr

Comments on this entry are closed.

Previous post:

Next post: