Computer game improves children's math performance

Jinjing "Jenny" Wang is a graduate student in psychological and brain sciences at Johns Hopkins University. She works with children and infants to study and improve human's intuitive sense of quantities.
Jinjing "Jenny" Wang is a graduate student in psychological and brain sciences at Johns Hopkins University. She works with children and infants to study and improve human's intuitive sense of quantities. (Caitlin Faw)

Parents whose children struggle with math may have new reason to be hopeful: A recent study at the Johns Hopkins University suggests that young people can improve their performance by carrying out a few simple computer exercises unrelated to numbers or math symbols.

A team of researchers in the university's department of psychological and brain sciences found that 5-year-olds who played a five-minute computer game — and played it in a particular way — scored significantly higher than their peers on a given set of math problems.


Instead of asking the children to work with numbers, the game required them to work with pictures of blue and yellow dots — targeting the kids' "intuitive number sense" rather than any knowledge of math they might possess.

The findings are noteworthy because they suggest that a simple method actually exists for quickly improving children's math performance — and that it might work because it targets a brain function rarely associated with this area of learning.


"It's not the case that if you're bad at math, you're bad at it the rest of your life. It's not only changeable, it can be changeable in a very short period of time," said Jinjing "Jenny" Wang, a doctoral student in the Krieger School of Arts and Science's department of psychological and brain sciences at Johns Hopkins and the study's lead author.

"Our key claim is that we can change children's math performance by working with their intuitive number sense — a capacity each of us is born with, not something we have to learn," she said.

An article on the findings appears in the July issue of the Journal of Experimental Child Psychology

Neuroscientists have long drawn a distinction between the intellectual capacities that all humans possess at birth and those we acquire through learning and study.


The first set of skills — called primary cognitive skills — include humans' innate sense of quantity, a capacity they share with many animals, including monkeys and rats. Studies have shown, for example, that infants given a choice between a plate that holds a few crackers and a plate with many more crackers will gravitate toward the plate with more.

They have yet to learn numbers, but they possess what cognitive-development researchers call "approximate number sense."

The second set of cognitive skills are those that must be learned and practiced — a grasp of numbers, for example, or the ability to add, subtract, divide and multiply those numbers.

People generally believe that children must practice math problems similar to those they will see on a test in order to get better at math in school.

Wang's team took a different approach, testing whether exercising children's approximate number sense, not their learned abilities, would help them perform better in math.

It did.

"Interestingly, this new study shows that boosting children's reliance on this intuitive understanding at least temporarily improves their performance on a standardized math test," said Melissa Libertus, a psychology professor at the University of Pittsburgh who studies children's cognitive development.

Wang and Lisa Feigenson, the study's senior author and a professor in the psychology and brain sciences department, have long been interested in the relationship between primary cognitive skills and those that are learned.

Earlier research by Feigenson and team member Justin Halberda, also a professor in the department, had shown that children who demonstrate exceptional approximate number sense tend to excel in school-based math. But no one had explored whether manipulating the first affects the second.

The team devised a computer game that exercises the innate number sense of children. For the experiment, they had 40 children play the game and then tested how it affected their math performance.

They showed each child a series of split-screen pictures that included blue dots on one side and yellow dots on the other. The 5-year-olds were asked to judge, without counting, whether there were more blue or yellow dots on each screen.

Working at the Johns Hopkins Laboratory for Child Development, the researchers showed one group of the children the most difficult screens first and worked their way toward the easiest. A second group saw the hardest and easiest screens in random order. A third saw the easiest screens first and were led progressively to the hardest — a sequence that best approximates the way in which we learn.

After proceeding through the game, each student took the same math quiz derived from a standardized math ability assessment test. They were asked to count backward, write down numbers, solve simple word problems and more.

The third group scored the highest, with about 80 percent correct answers — suggesting that proceeding in the proper order through the primary-skill exercises leads to better math performance.

The hardest-to-easiest group performed most poorly, with 60 percent correct.

The children also took a verbal test, but none showed improvement in that area.

Wang, who initiated the study as a first-year doctoral student three years ago, said the team had no idea what to expect and was stunned by the clarity of the results.

"It was really surprising, even to all of us, that this simple method would actually work," she said.

However interesting, the researchers said the findings generated many more questions than the study, supported by the National Institutes of Health, could address.

For instance, how long does such improvement last? Would younger or older children have the same outcomes? Would adults? Would a larger study with more children yield subtler results?

And in the longer run, can computer games like this one be developed to help students of all ages sharpen their computational abilities?

Wang and her team are following up. They've already completed a similar study on a group of 6-month-olds — eliminating any possibility that they've been "tainted" by mathematical knowledge — with the results to be published later this year.

Those findings, she said, only affirm that the human mind is quantitatively very active long before it's exposed to math concepts — and that it makes sense to explore this poorly understood function further.

"There are a million questions to be asked," Wang said. "We're trying to knock them down one at a time."

Recommended on Baltimore Sun