The study that should have changed everything
In 2014, Scott Freeman and colleagues published a meta-analysis in PNAS (Proceedings of the National Academy of Sciences) that should have been an earthquake in education. They analyzed 225 studies comparing active learning to traditional lecturing in STEM courses at the university level.
The results were not subtle. Students in traditional lecture courses were 1.5 times more likely to fail compared to students in courses with active learning. Active learning raised average exam scores by about half a letter grade. Failure rates dropped from 33.8% under traditional lecturing to 21.8% under active learning — a 55% reduction.
The authors wrote something extraordinary for an academic paper: if this were a clinical trial, the trial would be stopped for ethical reasons. Continuing to use passive lectures, given this evidence, harms students.
What counts as 'active learning'
Active learning is not a vague feel-good concept. It has specific, measurable characteristics. At its core, it means the learner is doing something that forces them to process and apply the material — not just receive it.
Self-testing (retrieval practice): Trying to recall information from memory. This is the single most effective learning technique identified by Dunlosky et al.'s comprehensive 2013 review. It works because recall strengthens memory traces in a way that re-exposure simply doesn't.
Elaborative interrogation: Asking 'why' and 'how' about facts you've just learned, then answering those questions. This forces you to connect new information to your existing knowledge, which is how durable memories form.
Generation: Producing answers, explanations, or solutions before being shown them. Even generating wrong answers before seeing the correct one improves later retention (Potts & Shanks, 2014).
What doesn't count (but feels like it does)
Dunlosky et al. (2013) evaluated 10 common study techniques. The results are uncomfortable for anyone who spent college with a highlighter.
Highlighting and underlining: Rated 'low utility.' It creates an illusion of engagement while requiring almost no cognitive processing. You can highlight an entire page while thinking about dinner.
Re-reading: Also 'low utility.' The second pass feels easier, which your brain interprets as learning. It is actually just familiarity. Recall after re-reading is barely better than after a single read.
Summarizing: Rated 'low utility' as typically practiced. Most students just compress the text rather than engaging with the ideas. The exception is when summarizing from memory (which is really retrieval practice in disguise).
Notice the pattern: passive techniques that feel productive score poorly. Active techniques that feel effortful score well. This is the fluency paradox at work.
The 55% number in context
A 55% reduction in failure rates is enormous. To put it in perspective: if a university course typically fails 100 students per year under traditional lecturing, switching to active learning methods would save roughly 55 of them. Scale that across every course in every university, and the human impact is staggering.
Freeman's team also found that the benefits were especially pronounced for disadvantaged students. Active learning didn't just raise the average — it disproportionately helped those who were struggling most. This makes passive lecturing not just ineffective but actively inequitable.
The finding was so robust that the authors called for an end to the debate. The question is no longer whether active learning works. It's why so many educational tools still ignore it.
Why most ed-tech still ignores this
Passive content is cheap to produce. Record a lecture, upload it, collect subscription fees. Active learning requires building comprehension checks, adaptive pathways, feedback loops. It's architecturally harder and more expensive.
There's also a demand-side problem. People want learning to feel easy. An app that makes you struggle feels broken. An app that plays soothing videos and fills progress bars feels effective. The market rewards fluency, not efficacy.
This is why Oivalla takes a fundamentally different approach. Every concept in the learning tree requires you to demonstrate understanding through quiz questions before you advance. The app doesn't let you slide past material you haven't grasped — because that's what the research says actually works.
Applying this to your own learning
Whatever you're studying, build active recall into the process. After reading a section, close the book and try to write down the key points. After watching a lecture, pause and explain the concept out loud. After learning a new procedure, try to execute it before watching the demonstration again.
The discomfort you feel when trying to recall something is not a sign that the method isn't working. It's the method working. That effortful retrieval is literally the mechanism by which your brain consolidates information into long-term memory.
225 studies. 1.5x failure rate reduction. Half a letter grade improvement. The evidence is not ambiguous. Stop consuming content passively. Start testing yourself actively. That's the entire secret.
Also available in: Deutsch, Español, Italiano, 日本語, Português, 中文