The Science Behind Numiq
Numiq is grounded in research on handwriting, mathematical problem solving, and the role of written work in learning. But the research doesn't stop at the foundation of the product. By capturing students' mathematical work step by step, Numiq creates new possibilities for understanding how mathematical thinking unfolds, and how technology can better support it.
01 — Established Research
These are the findings that informed how Numiq was built. These are findings from established areas of learning science that informed how Numiq was built. It covers the cognitive science of handwriting, how memory works, why retrieval practice beats re-reading, and what makes feedback effective.
Research status: Established findings are presented separately from emerging research and open questions. Numiq does not claim that the studies above validate Numiq itself.
02 — Adjacent & Emerging
Research directly adjacent to what Numiq is doing, which is work on digital handwriting, pen and stylus interaction, automated analysis of student work, and the challenge of capturing process rather than just answers. The field is moving, and these are areas where Numiq's approach may contribute.
Research status: Established findings are presented separately from emerging research and open questions. Numiq does not claim that the studies above validate Numiq itself.
03 — Open Questions
Most educational systems capture outcomes. Numiq is designed to capture the process that produces them. Here's what that could eventually make possible.
Most educational systems capture outcomes: answers, scores, completion rates, and grades. Numiq is designed to capture the process that produces those outcomes.
When a student solves a problem on paper, a great deal happens between the first stroke and the final answer. They make decisions about strategy, encounter obstacles, revise their approach, and sometimes correct their own mistakes before anyone else sees them. Traditional tools see none of this. They see only whether the answer was right.
Numiq captures mathematical work step by step. Over time, across students, classrooms, and problem types, that could create a kind of longitudinal dataset that is rarely available in mathematics education: a record of how students actually work through mathematics, not just what they produce.
The research implications of that are still being worked out. But some of the questions it might eventually help answer are genuinely compelling.
Misconception detection
Can the structure of a student's written work, not just the final answer, predict the specific misconception driving an error? And can that be identified early enough to intervene before it becomes entrenched?
Strategy development
How do students' problem-solving strategies change over time? Are there identifiable transitions, or, moments where a student shifts from a laborious approach to a more efficient one, and what conditions seem to produce them?
Individual variation
How much does mathematical thinking style vary across individuals, and are those differences educationally meaningful? Do certain approaches to written work predict better long-term retention?
The role of writing itself
We know handwriting supports memory. But does the act of writing out mathematical steps, specifically, not just generally, provide cognitive benefits beyond what typing or selecting answers provides? And if so, through what mechanism?
These are not claims Numiq makes about what it does today. They're questions that the kind of data Numiq collects could eventually help answer, questions that couldn't even be asked when educational systems only captured final answers.
