How TrainRun Works

1. Choose Your Content

Start by pasting a YouTube URL, entering a topic, or providing your own transcript. TrainRun works with any educational content - university lectures, language tutorials, science documentaries, podcast episodes, audiobooks, or personal study notes. The platform is content-agnostic: if it has words, TrainRun can turn it into a game.

2. AI Analyzes and Builds Your Level

When you submit content, our AI pipeline processes it in several stages. First, it extracts or generates a transcript with precise word-level timing. Then it performs semantic analysis to identify key vocabulary, group related concepts into thematic chains, and determine each word's educational significance. Finally, it compiles everything into a game map - a structured level where words are placed along a scrolling track, synchronized to the original media's timeline.

The same pipeline also extracts a concept graph from your content and generates targeted retrieval questions - multiple-choice quizzes designed to test understanding, not memorization. Each question is tied to specific concepts, calibrated for difficulty, and validated for quality before it reaches you.

3. Play: Collect Words Through Active Recall

In the game, you control a runner moving through a scrolling environment. Words from your content appear as collectible pickups along the track. As the media plays alongside the game, you navigate to collect each word at the right moment. Every successful collection is a micro-retrieval - a brief act of recognition and recall that strengthens the memory trace for that concept.

This approach is based on active recall, the most effective learning technique supported by cognitive science research. Unlike passive methods like re-reading or highlighting, active recall forces your brain to reconstruct information from memory, which builds stronger and more durable neural pathways.

4. Knowledge Check: Prove What You Know

After each section, TrainRun presents a short knowledge check - 3 to 5 multiple-choice questions designed to reinforce the concepts you just encountered. These aren't trivia questions. They test mental models, causal reasoning, and practical application, targeting the kind of understanding that transfers to real situations.

Every question includes immediate, explanatory feedback. When you answer correctly, you see why the answer works and a key takeaway to remember. When you miss one, you see the correct answer, why each distractor is tempting, and the common misconception it represents. This turns every question into a learning moment, not just a measurement.

At the end of a full run, a final test brings it all together with integration questions that connect concepts across sections - testing whether you can apply ideas in combination, not just in isolation.

5. Confidence Calibration

Before each answer is revealed, TrainRun asks how confident you are. This simple prompt unlocks powerful learning signals: answering correctly with low confidence reveals fragile knowledge that needs reinforcement, while answering incorrectly with high confidence exposes misconceptions - the most valuable things to correct early.

Why Active Recall Works

Decades of research in cognitive psychology have established that retrieval practice - the act of actively pulling information from memory - produces significantly better long-term retention than passive review. This finding has been replicated across hundreds of studies and is sometimes called the “testing effect.”

Traditional study methods like reading, highlighting, and re-watching create a “fluency illusion” - the material feels familiar, so you believe you know it. But familiarity is not the same as recall. When it matters (during an exam, a presentation, or a real-world application), you need to retrieve the information from scratch. Active recall trains exactly this skill.

TrainRun embeds active recall at two levels: first through the word-collection game, where you engage with vocabulary as it appears, and then through retrieval quizzes that test whether you actually understood the concepts - not just recognized the words.

Spaced Repetition Through Replay

Replaying levels introduces natural spaced repetition - another evidence-based learning technique. Each time you encounter a concept across different sessions, the spacing between exposures optimizes long-term retention. Combined with the variety of content you can turn into levels (different videos on the same topic, your own notes, generated content), TrainRun creates a comprehensive learning loop without the tedium of traditional flashcard systems.

The Learning Loop

TrainRun's core loop is designed around how memory actually works:

1. Watch - Consume the concept through the original media.
2. Run - Engage with the vocabulary through the word-collection game.
3. Retrieve - Pull the concept from memory during a retrieval quiz.
4. Feedback - Receive immediate, explanatory feedback on every answer.
5. Continue - Move to the next section with reinforced understanding.

The runner gets your attention. The knowledge check creates durable learning. Together, they build a retention engine that works with content you were already going to consume.

Content Modes

Frequently Asked Questions

Is TrainRun free?

Yes. Every user gets four free runs and ten minutes of play time per day. You can watch a short rewarded video to earn additional runs or time, or upgrade to Pro for unlimited access.

Do I need to download anything?

No. TrainRun is a web application that runs entirely in your browser. It works on desktop, tablet, and mobile - just open the site and start playing.

What kind of content works best?

Any content with educational value - lectures, tutorials, language lessons, podcasts, audiobooks, documentaries, study notes, or textbook passages. The platform works best with content that has clear vocabulary and concepts to learn.

How is this different from flashcards?

Flashcard systems require you to manually create cards and review them in isolation. TrainRun automatically extracts vocabulary from your content, embeds it in a game that plays alongside the original media, and generates targeted retrieval quizzes - all without any manual setup. You learn in context, not in isolation.

How do the knowledge checks work?

TrainRun's AI generates multiple-choice retrieval questions from your content, focused on understanding and application rather than trivia. Short section knowledge checks appear after each section, and a final check brings concepts together at the end. Every answer includes immediate explanatory feedback so every question is a learning opportunity.

Does it work with any YouTube video?

TrainRun works with any YouTube video that has captions (auto-generated or manual). Most educational content on YouTube has captions available. If a video doesn't have captions, you can use Topic or Transcript mode instead.

Can AI-generated questions be wrong?

Yes. AI-generated content may contain inaccuracies or omissions. Quizzes are designed as learning aids, not authoritative assessments. We validate questions through an automated pipeline, but mistakes can occur. If you spot an issue, use the Report button on any question.

Why are questions different each time I refresh a module?

Questions are generated once when a module is created and stored for consistency. If you refresh a module's map, new questions may be generated from the updated content structure. This keeps quizzes aligned with the latest version of your study material.