My Student Beat Every Level on the Math App—Here’s Why It Didn’t Matter
The half-sheet on my desk had three problems solved, one left blank, and two answers that were close but wrong. It belonged to Lily, a second-grader who, according to her mom, “loved the math app” and had finished every single level. At the parent-teacher conference that afternoon, her mom said it with real pride. “She does it every night. Twenty minutes, sometimes more. She practically never gets one wrong.”
I nodded and slid the paper across the table. “Lily is great at tapping out an answer. But on paper, without the options in front of her, she’s guessing.”
The mom looked at me like I was describing a different child. “But on the app, she gets them all right.”
That hour of the day is always the hardest. There were two more conferences waiting in the hallway, and I still had to explain to a very patient parent that her daughter’s nightly math practice wasn’t giving her the math skills we were supposed to be building. I didn’t have a clean explanation. Just a stack of half-finished worksheets and a feeling that something about the app was tricking us both.
I thought about that conversation all weekend. Not because the pattern was new—I’ve seen plenty of kids who can ace a digital game and freeze on the same skill printed in black and white—but because I finally went looking for an explanation that wasn’t based on the marketing notes of a fourteen-dollar app. That search is what led me to an APA expert paper on educational technology, the kind of thing that normally hides behind academic login pages. It didn’t tell me to take the tablet away. It told me something more interesting, and more practical.
The paper walked through something called the transfer problem. Kids who learn a skill inside a digital environment often can’t use it outside that environment. The research wasn’t making some broad point about screen time. It was about the structure of these apps. Many are built on short feedback loops and reward cycles—immediate confetti, level-ups, a cheerful voice saying “Great job!” The problem is what that structure does to a young brain like Lily’s.
On the app, she wasn’t really solving math. She was playing a game where correct answers triggered all the fun stuff and wrong answers triggered a gentle, friendly “try again.” Very quickly, Lily got good at reading the game rather than the numbers. She learned that division problems with friendly numbers usually had an answer hiding between two obvious options. She memorized the feel of the screen, the way the gray shading changed when she hovered over the right tile. She wasn’t internalizing division facts. She was memorizing the interface.
On paper, the red shading didn’t exist. The multiple-choice options didn’t exist. So her brain had no cue to start working. That’s why she’d stare at a blank space and raise her hand. The trigger for her thinking wasn’t the math. It was the app.
That single insight changed how I talked to parents for the rest of conference week.
Instead of telling Lily’s mom, “Cut back on the screen time,” I suggested something smaller. Let her finish her level, sure, but then have her do the last three problems by hand on a piece of paper at the kitchen counter before she gets the tablet again. Not the same numbers—different ones. Just a quick check to see if the thinking stuck.
“If she can do a similar problem on paper, even slowly, the skill is hers. If she can’t, then what she’s learning is how to use the app, not how to do math.”
That became our routine for the next few weeks.
I also changed things on my end. During independent math time, Lily was allowed to keep using the educational apps on her Chromebook, but she had to write a small number sentence in the margin of her notebook before she tapped any answer. Not full work. Just her first guess. Then she could tap and see if she was right. If she was wrong, she had to write one sentence about what went wrong. That was it. One sentence.
That one sentence slowed her down just enough to shift her attention from the game to the math underneath it.
One afternoon, I watched her stare at a multiplication word problem on the screen. She had her hand half-raised, ready to ask for help. Then she lowered it, picked up her pencil, and wrote “5 x 4 = ?” in her notebook. She counted on her fingers. Slowly. Inefficiently. Wonderfully. Then she tapped 20 on the screen and got the confetti.
The right answer wasn’t the important part. The counting was. She had to actually hold the groups of four in her head, picture the apples on the table, work out the total. The app wasn’t smuggling the answer out through its feedback loops.
That’s the other thing the APA material helped me understand. It’s about retrieval. Multiple-choice formats are a form of recognition—you’re scanning for a pattern. But kids remember better when they’re pushing the answer out of their own memory, without cues. Retrieval practice, they call it. Flashcards, for all their old-fashioned clunkiness, beat a well-designed app when the goal is checking whether the learning actually made it into long-term memory.
I also learned about something called the seductive details effect. When a learning app throws in animations, sound effects, and little digital pets that dance around, those details feel engaging. But for many kids, the brain ends up devoting most of its attention to the dancing pet. The math becomes background noise.
Lily’s mom texted me a photo of a worksheet a few weeks after we started the margin rule. She wrote: “Took 12 minutes for 6 problems. But she did it.”
Twelve minutes for second-grade arithmetic. Most kids her age finish that in four.
But the six problems she got right, she got right because she actually did the math. She knew where the answer came from.
I don’t want to pretend the app is useless. Lily still uses her math apps at school and at home. They’re genuinely useful as practice—repetition, exposure to new problem types, a way to keep skills warm. And they’re better than a lot of alternatives because kids are actually willing to sit still for twenty minutes and engage. I’ve just stopped treating the app’s report card as the real measure of progress.
If a kid tells me they got a high score on the computational fluency game, I celebrate for a second, then I ask them to show me. I scribble the same kind of problem on a sticky note, hand it over, and watch. Sometimes they get it. Sometimes they don’t.
The app subscription renewed itself again this month, as it tends to do. I haven’t figured out the perfect system for every kid yet, especially the ones whose families don’t have time to sit at the kitchen counter and watch them count on their fingers for twelve minutes.
The conference season ended, and Lily moved on to third grade. Her mom still texted me once in the fall to say that Lily finished a page of division homework without any help. Just once. That’s enough for now.
What I tell new parents in conferences is still shifting. The question I want them to carry home isn’t “How much screen time is too much?” It’s simpler and harder: “What does success on that screen actually mean?”
Please indicate: Thinking In Educating » My Student Beat Every Level on the Math App—Here’s Why It Didn’t Matter