The science, honestly

Asilak is built on a handful of findings about how memory actually works — that spacing study out beats cramming, that the best moment to review is right before you'd forget, and that making yourself remember builds stronger memory than just recognizing the answer. Here's the research behind each of those, in plain terms, with links so you can check it yourself.

1. Spacing your reviews out beats cramming

   Studying something a few times spread over days or weeks leaves a far stronger memory than the same study packed into one sitting. It's one of the most reliable findings in learning research, and it holds up specifically for language learning too.

   Sources:Cepeda et al. (2006), a meta-analysis of distributed practice · Kim & Webb (2022), spacing in second-language vocabulary learning

2. The best moment to review is just before you'd forget

   There's a sweet spot for each review: wait too long and it's gone; review too soon and you waste the effort. Researchers have measured this gap, and how far apart your reviews should be depends on how long you want to remember. Asilak aims for that moment for every card.

   Sources:Cepeda et al. (2008), the optimal spacing gap

3. Our scheduling uses FSRS — an open, independently benchmarked method

   To decide when to show each card, Asilak uses FSRS: a free, open method that models how memory fades and learns the timing that fits you. It's developed in the open and independently benchmarked against other approaches, so we don't have to take its accuracy on faith — and neither do you.

   Sources:How FSRS works (the fundamentals) · The open FSRS benchmark against other schedulers

4. Making yourself remember — not just recognize — builds stronger memory

   Picking the right answer from a list is easier than producing it from memory, but the harder kind of practice is what makes a memory last. That's why Asilak starts by letting you recognize the answer, then gradually asks you to recall it and finally to produce it yourself.

   Sources:Nakata (2016), retrieval format and vocabulary learning · Vos et al. (2018), productive vs. receptive practice

5. Streaks and points don't build real skill — and can backfire

   Badges, points and streaks can make an app feel rewarding without making you any better at the thing you came to learn — and chasing them can even crowd out genuine motivation. So Asilak leaves them out and shows you honest progress instead.

   Sources:Mekler et al. (2013), do game elements improve performance? · Mogavi, Hassan et al. (2022), learners' experience of gamified apps

6. Memory tricks and real context help things stick

   Linking a new fact to a vivid cue or a real example gives your memory more to hold on to. Where it makes sense, Asilak supports examples and context alongside the plain answer.

   Sources:Wang et al. (2024), mnemonics and contextual learning