Why we're building lingi, and how it's different

lingi is an app for Polish speakers learning Spanish from Spain. It's built on the learning methods with the strongest research behind them: spaced repetition, retrieval practice, interleaving, listening you can understand, and speaking with feedback.

Many language apps measure success by whether you come back every day. Hence the points, leagues, hearts and a reward for every answer. That works for engagement. We measure something else: how many words you still know a month later, and what you can say.

Here's the research lingi is built on, and how each piece shows up in the app.

Feeling productive isn't the same as learning

The best-studied learning methods have one thing in common. While you're using them, they feel harder and less effective than they are. Robert Bjork called them desirable difficulties. Spacing reviews out, recalling instead of rereading, and mixing material all slow you down in the short run and improve what you keep in the long run [1].

Learners can't feel this. In Nate Kornell's flashcard study, spacing worked better for 90% of participants, yet after the first session 72% of them believed cramming had worked better [2].

So an app designed to feel good in the moment can easily drift away from what teaches. Deci, Koestner and Ryan's meta-analysis of 128 experiments found that rewards for simply completing a task reduced people's later wish to do it for its own sake, while positive feedback increased it [3]. In lingi you don't collect points for tapping. You see what you know.

Spaced repetition

In 1885 Hermann Ebbinghaus memorised lists of nonsense syllables and measured how much effort he saved when he relearned them. After a day about a third of the saving was left, after a month about a fifth [4]. Murre and Dros repeated the experiment in 2015 and got a very similar curve [5].

The more useful finding is what happens when you spread reviews out. Cepeda and colleagues pooled 839 assessments from 317 experiments. Spaced practice beat massed practice, and the best gap between sessions grew with how long you need to remember [6]. In a study with more than 1,350 people, the best gap produced 64% better recall than reviewing on the same day [7].

Dunlosky and colleagues reviewed ten common study techniques. Only two got their top rating: spaced practice and practice testing [8].

In lingi. The FSRS algorithm schedules your reviews. For each card it estimates how well you remember the word from your ratings, and brings it back around the point where the chance you'd recall it drops to 90%. Prefer fewer reviews? Set it to 85%. FSRS is open source, and a public benchmark tests it on hundreds of millions of real reviews. There it predicts recall better than older models such as HLR [9].

Recall beats rereading

In Roediger and Karpicke's study, students read a passage and then either read it again or tried to recall it. Five minutes later, rereading came out ahead. A week later it flipped: 56% against 42% [10].

Their 2008 study with 40 Swahili-English word pairs is even starker. When students kept recalling words they'd already got right once, they remembered about 80% a week later. When those words dropped out of testing, they kept 36% and 33%. More study after the first successful recall had no measurable effect. And everyone, whatever the method, predicted they'd remember about half [11].

In lingi. Every card asks you for the answer before it shows it. On Say cards you say the word out loud before you reveal it. If you'd rather type, turn on typed answers and the app tells you when an accent or an article is missing. Then you rate how it went, and that rating decides when the card returns.

Mix it up instead of working in blocks

Practise one kind of problem at a time and it goes smoothly, because you know what's coming. Mix them and you have to work out what you're looking at every time. Rohrer and Taylor showed this with maths problems. Students who practised with the problems shuffled did more than three times better on a test a week later than students who practised in blocks [12]. Brunmair and Richter's meta-analysis found that interleaving helps most when you have to tell similar things apart, such as paintings by different artists. For memorising plain word lists, blocks did better [13]. That's why lingi introduces new words one at a time and mixes the reviews and exercise types.

In lingi. A session doesn't stick to one lesson. Reviews from every lesson come up when they're due, new words slot in every few cards, and Listen, Say and gap cards are mixed together. The same material also comes back in a different form: a word from your flashcards turns up in an episode, and then you say it to the tutor.

Common words first, always in sentences

Words aren't equally useful. Paul Nation worked out that to follow a text without help you need to know about 98% of its words. For English that means 8,000 to 9,000 word families for reading and 6,000 to 7,000 for listening [14]. The lesson carries over to any language: order matters, because common words pay off straight away.

A one-word translation is only the start of knowing a word. Stuart Webb found that the more often you meet a word in context, the more you know about it, and that even ten encounters may not be enough for full knowledge [15]. Much of language is ready-made. Erman and Warren estimated that around 55% of the English texts they studied was prefabricated [16].

In lingi. The course runs 95 lessons from A1 to C1, about 7,900 words and phrases in total. Inside each lesson the most common words come first, and we check levels against the Instituto Cervantes inventories (PCIC). Every word has an example sentence with audio. Gap cards show the base form as a hint, as in Mi amigo ____ (ser) polaco, and the line under the sentence names the form: es = ser, 3rd person singular. From B1 on, more and more cards are whole phrases like darse cuenta de or a pesar de.

If you already know some Spanish, a placement test of at most 30 words, about 2 minutes, shows you where to start. Some of its words are made up. That's Meara and Buxton's method for catching people who tick "I know it" too generously [17].

Listening you can follow

Stephen Krashen argued that we acquire a language mainly by understanding messages slightly above our level [18]. Nation's numbers show where that line sits. When you know nearly every word, you can pick up the rest from context [14].

In lingi. You listen to short episodes from everyday life in Spain. The series Un piso en Madrid has eight episodes at A1 and A2: a new flat, the bar downstairs, the market, the chemist's. Before an episode you see what share of its words you already know, and the app suggests it once you know at least 95%. You listen first, then read the text with translations and answer questions about what was said. Turn captions off or slow it down. Save a line with one tap and it comes back in your reviews.

Speaking, feedback and pronunciation

Understanding isn't enough on its own. Merrill Swain studied students in Canadian French immersion schools. After years of lessons they understood French almost as well as native-speaking peers, but their speaking and writing lagged well behind. Her conclusion was that learners need to speak and write too, because that's when they notice what they're missing [19]. In a study by Boers and colleagues, learners who used more set phrases got higher ratings for their speaking [20].

Feedback works too. Lyster and Saito's meta-analysis found that oral corrective feedback clearly improves results, and the effect lasts [21]. Pronunciation can be taught. Lee, Jang and Plonsky pooled 86 studies and found a clear effect of pronunciation instruction [22].

In lingi. You speak from the first lesson. On Say cards you answer out loud. In the sentence bank you hear a sentence from real life, repeat it out loud along with the audio, then recall it from memory. In conversations with the AI tutor you talk in Spanish by voice: a warm-up, words from your lesson and a role-play at your level. The tutor corrects you as you go, and afterwards you get a recap with your corrections and the new words, which you can save as flashcards. We don't store the audio.

Explanations in Polish

Translating into your first language is often treated as a shortcut to avoid. The research disagrees. In Laufer and Shmueli's study, words glossed in the learners' first language were always remembered better than words glossed in the target language [23]. A meta-analysis by Yanagisawa, Webb and Uchihara, covering 42 studies and 3,802 learners, found the same [24].

In lingi. Translations, hints and grammar notes are in Polish. If you'd rather have English, switch whenever you like and keep your progress.

A daily plan that mixes it all

Paul Nation suggests splitting study time roughly equally between four strands: listening and reading for meaning, speaking and writing, deliberate study of words and grammar, and fluency practice with what you already know [25]. Rhythm matters too. In Baddeley and Longman's classic study, postal workers learning to type made the fastest progress with one hour a day and the slowest with two two-hour sessions [26].

In lingi. The Today screen suggests what to do next: reviews, new words, an episode, the sentence bank or a talk with the tutor. At first most of your time goes on flashcards, because without words there's nothing to talk about. The higher your level, the more of the day goes on listening and speaking. You pick a daily goal of 5, 10, 15 or 20 minutes. Your streak counts study days, and you get one freeze a week for a bad day. Miss a few days and the backlog is spread over the next few instead of landing all at once.

Numbers instead of points

There are no points, badges, leagues or hearts. After a session you see numbers that mean something: words known, minutes studied, the share of answers you remembered, and how many reviews are coming up. Apps built around streaks and points reward you for coming back. lingi shows you what stuck.

If you'd like to try it, join the waitlist. We'll email you once, when lingi launches.