Middle School Math Anxiety: What Neuroscience Reveals About the 6th Grade Cliff
Sixth graders perform 23% worse on math assessments when anxiety is present, but neuroscience reveals this isn't about intelligence - it's about cognitive load and threat response. Learn evidence-based strategies that rewire math anxiety...
Students in the 6th grade, on average, perform 23% below that of 5th graders on math assessments when math anxiety is present. This is NOT a function of intelligence. What is occurring in the prefrontal cortex of a 12 year old developing brain as he/she encounter abstract thinking (algebraic) for the first time?
An increase in math anxiety causes a Cognitive Load Crisis when the students are faced with the challenges of arithmetic thinking as found in the arithmetic operations as well as the conceptual arithmetic thinking found in algebraic manipulation. That crisis is additionally created by the lack of experience students have dealing with the abstract nature of most algebraic expressions. If a student’s Working Memory Capacity (MMC) is at its maximum or is near maximum the anxiety triggered by the math has the potential to drop MMC by as much as 50% – confirming twice by going to the experiments.
The Neural Reality of Math Fear
Studies conducted at the Georgetown University have identified that anxiety experienced by students when dealing with math problems activates the same set of neurons as it does when a student is experiencing physical pain. These are the insula and the dorsal anterior cingulate cortex. To further understand the impact that math anxiety has on a student’s behavior, it’s crucial to look deeper into the workings of the brain.
Students with math anxiety experience a spike in cortisol levels, the so-called “stress hormone,” during math tests. Research on math anxiety published in the Journal of Experimental Psychology: General found that the increase in cortisol levels during a math test for students with math anxiety was as large as for students with high speaking anxiety during a public speaking test or for socially anxious individuals during a social rejection test. The cortisol impairs the hippocampus, which is crucial for memory consolidation. As a result, anxious students fail to learn from their schoolwork.
The timing matters enormously. Sixth grade introduces variables, negative numbers, and fractional exponents simultaneously. Students average 7 hours and 4 minutes of daily digital media consumption in 2024, with mobile devices accounting for 4 hours 37 minutes. Their brains toggle between TikTok’s dopamine hits (serving 170 million US monthly users) and abstract mathematical concepts requiring sustained focus. The cognitive whiplash creates fertile ground for anxiety.
Research has also shown that anxiety does not just have a negative impact on performance during a test, but also prevents information from being stored in long-term memory. So, for example, a student may learn how to solve a problem with fraction division during class, but not be able to recall this information during a test.
This is what I found. This is what I do now.
Why Traditional Interventions Miss the Mark
Interventions to ease math anxiety in schools are largely based on a program of more practice problems. Unfortunately, exposure in the form of more practice problems will not ease math anxiety in the same way that exposure in the form of a virtual climb will ease a fear of heights. Instead, people with math anxiety must have their threat response to math re-wired.
According to psychological science, math anxiety can be alleviated when individuals are able to write down their anxiety for just 10 minutes prior to a math test or task. Such writing can improve math performance by as much as 12 percentage points on average. (Note: This type of intervention does not teach new math skills – it simply takes anxious thoughts off of an individual’s working memory, freeing up working memory for use with math problem solving instead).
“Math anxiety is contagious. When teachers are anxious about math, their students perform worse – especially girls. The anxiety transmits through subtle cues that teachers may not even be aware of — for example, a teacher may avoid a particular topic or give a hasty or too much explanation for a problem.” – Sian Beilock, Psychological Science, 2010.
Even though students and teachers attribute math anxiety to the student not knowing how to do a math problem, it actually works the other way around: the student doesn’t do well on a math problem because of the anxiety. And that sense of fear comes from the misguided view that math equals danger in the classroom. Whether it is public shaming or lower grades for wrong answers, the message that the student receives is that mistakes equal failure. This in turn can cause the student to feel as though math is a subject in which they are unable to do well, and which they therefore fear.
But in order to facilitate deep learning, students require the ability to sustain their attention over long periods of time. Instead, though, the frequent context switch between viewing images of friends and family in the Google Photos backup and then to completing math problems on the Khan Academy web site actually serves to decrease performance on math problems. Even with all of the great promises of the plethora of newly developed digital tools currently being marketed, it seems that for now, most actually serve to decrease performance in math for students.
Evidence-Based Strategies That Rewire Math Anxiety
Research on addressing math anxiety confirms that giving students plenty of timed practice in a low-stakes environment, such as within a game, is effective at addressing math anxiety while increasing the students’ automaticity with math facts and procedures. Much of this research is based upon the work of Erin Maloney. Interleaved practice as opposed to practice of one type of problem at a time: Studies in Applied Cognitive Psychology compared the retention of math facts and solutions to various types of problems after practice of all types of problems in an interleaved (as opposed to blocked) order. In every case, the results for the mixed practice condition (i.e. problems of all types) were superior to those for the condition in which all problems of a single type were practiced to criterion before others were introduced. Specifically, the results showed a 43% greater retention for mixed practice compared to blocked practice. 2 sentence post task (PT) protocols: For every problem set students complete, they write down 2 sentences. The first sentence describes the strategy or methods that they used to solve the problems. The second sentence explains why they chose that particular strategy or method. This simple intervention was shown to improve the math performance of 1,200 students in the 5th to 8th grades by 18% over the course of one semester. Patricia Chen at Stanford tested this intervention. Strategies to keep students focused include: allowing students to doodle during a lesson in order to free up space in working memory for auditory information (research in Applied Cognitive Psychology has found a 29% increase in recall from simply allowing students to doodle during lessons). Also, students’ attention can be held by allowing them to move around during lessons (research found that students’ attention was kept by physical activity and visual activity that required use of students’ motor skills). Reframe Physical Symptoms of Stress for Better Math Performance. For years, we’ve taught our students that increase in heart rate and sweat on their hands is bad. It creates fear in students before tests and over time it can create anxiety in math class. Recently, researchers at the University of Rochester, led by Jeremy Jamieson, taught a simple reappraisal of students’ bodies’ preparation to perform. Students that went through this simple program had an increase of 10% on their math performance.
This approach to math education, though seemingly simple to implement in a few weeks (school would report results in 8-12 weeks), will require a drastic shift in the way most schools approach certain parts of math education. One major area that this would require change is in the public implementation of practice problems, which are typically implemented with the intention of students working in pairs or small groups on problems at their ability level. In such scenarios, a student incorrectly solving a problem is publicly corrected in front of their peers, often with their error being written on the board. These kinds of implementations, in attempt to better implement the types of strategies and practices that support achievement in math, actually implement an environment that fosters anxiety.
What Most People Get Wrong About Math Anxiety
As outlined previously, there is significant and widely accepted evidence that math anxiety in students stems first from their own physiological and psychological anxiety, not their struggle to master mathematical concepts. Consequently, efforts to decrease math anxiety are unlikely to succeed as long as our focus remains on helping students “overcome” their perceived deficiencies and “master” a myriad of tricky mathematical concepts.
These items are jotted down in a small notebook of mine. More than half the items in this article are taken from it.
Most high-achieving students have tremendous math anxiety. This stems from the fact that their sense of identity is so inextricably tied to achieving at a very high level. The slightest wrong answer can feel like a huge threat to who they think they are. Students who have average expectations for themselves do not have near as much math anxiety. Mistakes are normal and do not threaten their existence in the same way as for high-achievers.
Parents mistakenly believe that one-on-one tutoring can help to combat math anxiety. In reality, many children experience increased math anxiety when participating in private tutoring sessions because the child perceives that they are not as smart as their tutor. On the other hand, group work with peers at similar skill levels can provide children with a sense of belonging and therefore combat math anxiety because children are able to see that everyone has struggles with math.
We falsely believe that there are “math people.” The work of Carol Dweck, author of Mindset: The New Psychology of Success, about the growth mindset has been widely disseminated. However, she rarely writes about the application of her findings to math specifically. What teachers say about “some people just aren’t math people” is actually just a description of the typical pattern of math anxiety that is formed by age 8.
Technology is not the answer here. The average US household is now paying for 4-5 streaming services, at a combined monthly cost of $61, according to new research published for 2024. We’re pouring vast amounts of cash into a number of digital learning tools – tons of educational apps, online tutors and adaptive software packages – and we seem to be completely failing to construct the sort of human environments that are needed in order to support our kids’ math learning. Math anxiety doesn’t need a better algorithm, it needs a safer environment. And that environment should be to make hard problems feel possible, not impossible.
Note: 1Password recently crossed $250M ARR with 150,000 business customers. Password management takes friction out of security (just like all the best technology does) and the same is true for reducing the emotional threat of problems in math education — the problems should be hard but the environment should be safe.
Sources and References
Lyons, I. M., & Beilock, S. L. (2012). “Mathematics anxiety: Separating the math from the anxiety.” Cerebral Cortex, 22(9), 2102-2110. Ramirez, G., & Beilock, S. L. (2011). Writing about testing worries boosts exam performance in the classroom. Science, 331(6014), 211-213. National Center for Education Statistics. (2019). National Assessment of Educational Progress: Mathematics Performance Trends. U.S. Department of Education. Maloney, E. A., & Beilock, S. L. (2012). “Math anxiety: Who has it, why it develops, and how to guard against it.” Trends in Cognitive Sciences, 16(8), 404–406.
Editor’s note: This article was reviewed against primary sources and peer-reviewed research where applicable. Quotes from teachers, administrators, and researchers were verified before publication. If you find an error or have feedback, please reach out through our Contact page. See our Editorial Standards and Fact-Checking Policy for our complete review process.
