An elementary school teacher knows this moment well. The child looks at a row of numbers, thinks about the question, and comes up with the right answer. They got the same answer as the kid next to them. There is something different about the child’s work, speed, and the way they get back up after falling. A brand-new study from Stanford Medicine backs up that feeling with real science, and what it finds is very important to pay close attention to.
The study, which was published in the Journal of Neuroscience, used 87 second and third graders and gave them a seemingly easy task: compare two numbers and choose the bigger one. There were groups of dots in some problems. Some people used Arabic numbers. A math learning disability is defined as scoring at or below the 25th percentile on a standard math fluency test. These kids got the right answer about the same number of times as their peers. Not much of a difference at first glance. Brain scans showed that their thoughts were different.
When math problems got harder, kids who didn’t have a learning disability slowed down. They made a mistake and quietly adjusted, taking a half-beat longer on the next question and changing how they were going about it without being told to. This wasn’t done by kids who had trouble with math, especially when they were working with number symbols instead of dot clusters. They worked through problems at the same speed whether the comparison was simple or very close. They were less likely to stop and change gears after making a mistake. It’s also important to note that none of this was shown by their accuracy scores alone.
There was less activity in two parts of the brain that were scanned: the middle frontal gyrus, which helps with focused attention and executive function, and the anterior cingulate cortex, which helps people see mistakes and make decisions. It’s possible that a lot of teachers have been measuring the wrong thing. A child’s right answer tells you something. What a child’s brain does to keep track of its own performance tells you something much more important.
Vinod Menon, the study’s lead author and a psychiatry professor at Stanford, said that math learning disability is like a chain reaction. You could fall behind, lose confidence, get nervous on tests, or just stop caring about the subject. Many years ago, people in academia thought that math problems were caused by a lack of motivation or effort. But more and more evidence shows that they are actually caused by a specific processing problem. Three to seven percent of people have dyscalculia, a more specific type of math learning disability that makes it hard for them to understand basic numbers. But the larger group of kids who just don’t do well—those who score in the bottom quarter—is much less talked about.
It’s interesting what the Stanford team found about dots vs. numbers. After making mistakes, kids who had trouble with math were more careful with dot-based problems. Most of the time, their non-symbolic number sense seems to be fine. When abstract symbols come into the picture, things start to fall apart. When numbers on a page are separated from any physical quantity, they seem to require a different kind of mental effort—one that needs more executive control, more error monitoring, and more of exactly what their brains are giving them less of.
By changing this, we change what teachers think “helping a struggling math student” really means. It’s natural to want to do more problems, go over the basics again, and repeat the steps. That method isn’t wrong, but it might not cover everything. Metacognitive skills—teaching kids to know when they’ve made a mistake, to pause, and to consciously adjust—are what Menon’s team says should be the main focus of interventions. Not just the answer, but also how to think about how to get there.
As you read this, you can’t help but think about how many kids have been told to work harder in math class for years when what their brains really needed was a different kind of practice. The study does not give a full answer. But it might be more useful because it changes how the problem is mapped out. That’s where real progress usually starts.
