How Vellum teaches, and why.
Vellum is built on four ideas. Each one comes straight from learning research — and each was, until now, too time-consuming for teachers and textbook authors to build by hand.
Four ideas Vellum is built on.
You do something with it.
Every idea comes with a picture you act on: tiles on a balance scale, a cone you slice, a row you drag to clear a column, a valve you click on a beating heart.
You can’t cram it.
Questions come back on a schedule measured in days. Get one right and it waits longer; miss it and it steps back one box, not to the start.
Mastered means it stuck.
A unit counts as mastered when most of its questions have held up over weeks — not when you’ve seen them once. Answer choices reshuffle every time, so you learn the material, not where the answer sat.
It never runs out.
Math and grammar problems are generated fresh, and any equivalent answer counts: ½, 2⁄4 and 0.5 are all right.
The 1648 treaties that ended the Thirty Years’ War and recognized each state’s sovereignty over its own religion and affairs.
The highlighted card is the one you’re studying now.
No streaks to protect, no timer on anything you haven’t mastered, and no points for answering a question before it’s due. Racing your classmates is there when you want it — never required.
Methods the research has backed for decades — finally practical to deliver.
None of Vellum’s methods are new. Spaced practice, retrieval practice, mastery learning and hands-on interactive models are among the best-supported ideas in learning science. What’s new is that software can finally do the work that kept them out of most classrooms.
When students were tutored one-on-one and had to master each unit before moving on, the average student outscored 98% of a conventionally taught class. Benjamin Bloom called it a problem because that kind of attention was “too costly for most societies to bear on a large scale.”
A textbook can’t decide which page you need to see again tomorrow. A teacher with eighty students can’t keep eighty review schedules, write fresh problems on demand, check every answer the moment it’s given, or build a 3D heart. Vellum does that bookkeeping, so teachers can spend their time teaching.
Bloom, B. S. (1984). The 2 sigma problem. Educational Researcher, 13(6), 4–16.Practice that’s spaced and effortful.
Rereading and highlighting feel productive but fade fast. Pulling an answer out of memory, and doing it again days later, is what makes it stick. It’s also the hardest thing to schedule by hand.
experiments reviewed: spreading study out beats cramming it together, and the longer you need to remember something, the wider the gaps should be.
Cepeda et al. (2006), Psychological BulletinRereading looked better five minutes later, but on the delayed tests, students who had practiced recalling a passage remembered substantially more.
Roediger & Karpicke (2006), Psychological ScienceOf ten common study techniques, only practice testing and spaced practice earned a “high utility” rating. Highlighting and rereading rated low.
Dunlosky et al. (2013), Psychological Science in the Public InterestDoing beats watching.
When you turn the skeleton, slice the cone or predict where the projectile lands before it flies, you’re inside the idea instead of reading about it. That immersion is the point: every model on Vellum asks you to act, guess and check.
Across 225 studies, students in lecture-only courses were 1.5 times as likely to fail as students in active-learning courses, and active learning raised exam scores by about 6%.
Freeman et al. (2014), PNASLearning deepens as students move from passively receiving to actively doing, to constructing ideas, to building on them interactively.
Chi & Wylie (2014), Educational PsychologistA decade of science-education research found “robust evidence that computer simulations can enhance traditional instruction,” especially in lab work.
Rutten, van Joolingen & van der Veen (2012), Computers & Education