How working memory and mastery help students learn faster

Most students who fall behind in school are not lacking intelligence. According to Joe Liemandt, principal of Alpha School, they are missing foundational skills that never reached fluency, and no amount of motivation or grit can substitute for that gap. His school's data, discussed with Stanford neurobiologist Andrew Huberman, offers a striking look at how learning science and the brain's own limits explain why some students struggle and others thrive.

Working memory is the bottleneck

The brain has a limited number of working memory slots, roughly tied to IQ, that are used for active, conscious computation. When a skill is memorized to fluency, using it takes no working memory at all. When it is not, every use drains a slot that is needed for higher-level thinking. This explains why students who never memorized multiplication tables struggle in algebra: they are burning working memory on basic arithmetic instead of the multi-step reasoning the problem actually requires.

One student scored a 710 on the math SAT and kept making what looked like careless errors despite being told repeatedly to focus. The real issue was that she had never achieved fluency with her multiplication tables. After going back to memorize them, her score rose to 790, evidence that the bottleneck was never attention but working memory capacity being spent in the wrong place.

Why mastery beats "good enough"

Traditional classrooms often move students forward once they hit 80 percent on a topic, the equivalent of a B grade. Liemandt compares this to a basketball coach tolerating a player who loses the ball 20 percent of the time: no coach would accept that, because errors compound as skills build on each other. In education, an 80 percent grasp of third-grade math becomes a fragile foundation for fourth grade, and the gaps stack up year after year.

National data illustrates the cost. On one 300-point standardized test, the average student gains only a single point between eighth and twelfth grade, essentially no measurable growth over four years. Liemandt argues this happens because students are pushed into grade-level content regardless of whether they have mastered the prerequisites.

Spaced repetition and worked examples

Two techniques from cognitive science make mastery achievable without excessive struggle. Worked examples show a student a fully solved problem, then progressively remove steps until they can solve it independently, a faster path to competence than unguided trial and error. Spaced repetition revisits material just before it would normally be forgotten, moving it from working memory into durable long-term storage.

Alpha School's internal data found that older students, working through multi-step algebra, benefited most from worked examples, while younger students needed more initial repetition before spaced review could take over. Combining both approaches allowed students to close an average gap of 2.2 grade levels using only 20 to 30 hours of targeted instruction per grade level, far less than the 180 school days a full grade level normally takes.

Motivation determines whether learning happens at all

Liemandt estimates that motivation accounts for 90 percent of whether a personalized lesson actually works. Even a perfectly calibrated lesson fails if the student disengages. Alpha School reframes setbacks in a specific way: instead of telling a struggling eighth grader they are three years behind, the message is that they are 22 hours of instruction behind, a distance that feels solvable rather than shameful.

The same principle shows up with attention and focus. Children with ADHD diagnoses who cannot sit through a lecture can often focus for 25 minutes on an engaging, game-like task, since attention itself is not the limiting factor. Alpha School structures lessons in short, engaging blocks for this reason, and pairs academic time with activities students already care about, from sports to entrepreneurship projects, so that finishing lessons efficiently earns real time for what they love.

Removing shame changes the trajectory

Because instruction is individualized, a student working three grade levels below their age can do so without classmates ever knowing, removing the social cost that normally attaches to falling behind. Liemandt notes this matters because self-perception, not just skill, drives future effort: once a student experiences visible progress without public failure, they start attempting harder material voluntarily. A group of middle schoolers who believed they were simply "not math people" scored in the top percentile on the math SAT once that belief was directly challenged and disproven through early wins.

What the results suggest

• Fluency, not raw ability, predicts whether new material will be usable under pressure.
• Reframing a learning gap in hours rather than grade levels makes it feel achievable.
• Worked examples suit complex, multi-step material; younger learners benefit more from repetition and initial mastery.
• Engaging, structured short blocks can sustain focus even in students who struggle with sustained attention elsewhere.
• Closing a grade-level gap can take as little as 20 to 30 hours of correctly targeted instruction.

The takeaway

The underlying message is less about any single school and more about how the brain actually learns: fluency frees up limited cognitive resources, spaced repetition and worked examples build it efficiently, and motivation determines whether any of that gets a chance to happen. Applying these same principles, whether in a classroom or in self-directed learning, can meaningfully speed up how quickly gaps get filled at any age.