The forgetting curve is older than the internet
In 1885, a German psychologist named Hermann Ebbinghaus sat down and memorized lists of nonsense syllables. Then he tested himself. Then he waited, tested again, and charted how fast his recall decayed.
What he found was uncomfortable. Without any reinforcement, he forgot roughly 40% of new material within 20 minutes. After a day, he retained less than a third. After a week, around 25%. The decay wasn't gradual — it was steep, fast, and predictable.
A hundred and forty years later, every modern study on memory confirms the same basic shape. Your brain treats unused information as clutter and clears it out. Fast.
Online learning makes this worse, not better. When you read a textbook, you physically interact with the pages. When you attend a class, social accountability keeps you engaged. When you watch a Udemy lecture, you're lying on a couch, half-distracted, hitting play on something that feels a lot like Netflix.
Passive consumption is comfortable. It's also nearly useless for long-term retention.
Why most "learning" strategies don't actually work
Most people try to solve the retention problem with one of three strategies — and most of those strategies are less effective than people believe.
| Strategy | What people think | What research says | Effectiveness |
|---|---|---|---|
| Re-watching lectures | More exposure = more memory | Familiarity feels like memory but isn't | Low |
| Highlighting notes | Important points are marked | Passive; doesn't require recall | Low |
| Re-reading notes | Reviewing reinforces memory | Recognition vs. recall — different things | Medium |
| Practice testing | Testing is for exams, not learning | Retrieval practice is highly effective | High |
| Spaced repetition | Complicated flashcard system | Most effective method for long-term recall | Very high |
Re-watching a lecture feels productive. Your brain recognizes the content and that recognition creates a warm feeling of familiarity. But recognition is not recall. Recognition is "I've seen this before." Recall is "I can reconstruct this from scratch." Only recall actually reflects learning.
The testing effect: Studies consistently show that attempting to recall information — even if you get it wrong — strengthens memory significantly more than re-reading the same material. Getting an answer wrong and then seeing the correct answer is particularly powerful.
What spaced repetition actually does
Spaced repetition solves the forgetting curve by working with it instead of against it. The idea is simple: review a concept just before you'd naturally forget it. Each successful recall pushes the next review date further out. Eventually, the concept is in long-term memory and needs only occasional reinforcement to stay there.
The most widely used algorithm for this is SM-2, developed by Piotr Wozniak in the late 1980s. Anki, the popular flashcard app, runs on a version of SM-2. So does Ravenote.
Here's what the review schedule looks like for a typical concept:
- First review: 1 day after learning
- Second review: 3 days later
- Third review: 8 days later
- Fourth review: 21 days later
- And so on — intervals grow with each successful recall
The catch with Anki and most spaced repetition tools is that someone has to make the cards. You have to stop, write out the question and answer, add it to the deck. For a 10-lecture course, that's an hour of setup before you even start reviewing. Most people skip it.
Why most learners never fix their retention problem
The tools exist. Anki has been around since 2006. The science has been clear for decades. Yet most online learners still watch, feel like they learned something, and forget most of it within a week.
The gap is friction. Every extra step between "I just watched a lecture" and "this concept is being reinforced at the right interval" is a place where the system breaks down.
Creating Anki cards from scratch. Writing your own quiz questions. Remembering to actually open the review app. It's not that these things are hard. It's that they require consistent execution on top of an already-demanding learning habit. Life gets in the way.
The only version of spaced repetition that works reliably for online learners is one that requires nothing extra from them.
What automatic spaced repetition looks like in practice
Ravenote is built around this exact problem. When you watch a Udemy lecture, it captures the transcript and generates structured notes automatically. Those notes become the raw material for a quiz when you pause the video — questions generated from the actual content, not generic flashcards.
Your response to each question feeds into an SM-2 scoring system. Concepts you know well get scheduled further out. Concepts you're shaky on come back sooner. The whole thing runs in the background — you don't manage a deck, you don't decide what to review, and you don't have to remember to open a separate app.
When you return to a course after a few days, Ravenote primes you with a short review of what the algorithm says needs attention. Before you've even hit play on the next lecture, you've already reinforced three concepts from two lectures ago.
That's the difference between a tool and a system. The tool gives you notes. The system makes sure those notes actually produce knowledge you can use.
Spaced repetition that runs itself
Ravenote captures your Udemy lectures, quizzes you when you pause, and tracks every concept with SM-2. No decks to build. No app to remember. It just works.
Get Ravenote — from $9What you can do right now, tool or no tool
If you want to improve your course retention starting today, three habits will move the needle:
Test yourself, don't re-read. After every lecture, close your notes and write down everything you remember. Then check what you missed. The act of trying to recall — even imperfectly — strengthens the memory far more than another pass through the content.
Space your sessions. Watching three hours of a course in one sitting feels efficient but produces poor retention. Two one-hour sessions with a day between them will leave you with measurably more. Distributed practice consistently outperforms massed practice in the research.
Explain it to someone. The Feynman Technique works because teaching something forces you to locate every gap in your understanding. Even explaining a concept out loud to yourself — or to an imaginary listener — is significantly more effective than passive review.
The bottom line is straightforward. Forgetting is not a personal failure. It's what your brain does by default. Retention requires deliberate effort, timed correctly. The sooner that effort becomes automatic rather than manual, the more of what you learn will actually stay with you.
Related: How to take better notes on Udemy without typing · Best Udemy Chrome extensions for 2026