Learning and memory
What is microlearning? Definition, examples and evidence
Short answer
Microlearning is learning in small, focused units, usually a few minutes or less; one review of the research put the ceiling at 15 minutes. Studies of it are small and mostly positive for knowledge. The strongest evidence is for two things a short unit makes easy: answering a question about it, and coming back to it days later.
Most definitions of microlearning are written for corporate training teams. There it means short modules that staff can finish between meetings. For a person learning on their own, the idea works the same way at a smaller scale: one idea per unit, finished in one sitting, often on a phone in a queue or on the bus. We make Scroll, an app built on short lessons, so we’ve kept each claim below to what the cited studies measured.
What microlearning means
Researchers haven’t settled on one definition. When Monib and colleagues reviewed 40 studies published from 2020 to 2024, they noted that the field has no agreed definition, and the units in those studies ran from 1 to 3 minutes up to 10 to 15 minutes.
A definition that fits most of that research: microlearning means learning in small, focused units, usually a few minutes or less. One review of the research set the ceiling at 15 minutes. In De Gagne and colleagues’ 2019 scoping review of 17 studies in health professions education, every unit took less than 15 minutes, and some took a few seconds. They described microlearning as “the acquisition of knowledge or skills in the form of small units.”
The size limit applies to each unit. Ten units in a day still count as microlearning, though together they take longer than one 15-minute module. What the definition leaves open is what happens inside the unit: a short video you watch once and a short lesson that ends with a question both qualify, though only the second asks you to recall anything.
Scroll sits at the short end. Across all 1,753 lessons in the app, each runs 21 to 115 words (median 69), the app shows a reading time of 32 to 50 seconds (median 44), and every lesson ends with one question with four options.
Examples of microlearning outside a training department
For someone learning on their own, microlearning usually looks like one of these:
- A flashcard session. Apps such as Anki show you a prompt, wait while you try to recall the answer, then bring the card back on a later day.
- A language app’s daily lesson. Duolingo, which teaches languages in short lessons, reported 58.7 million daily active users in the second quarter of 2026.
- A daily fact with a question. This site’s fact of the day shows one lesson from the app each day, with its source and a quiz, free on the web.
- A short lesson in an app followed by one question, like the Scroll lesson below.
- Hospital training. In the studies pooled by Hunter and colleagues in 2026, hospital staff took short units on caring for older patients, including how to recognize delirium.
Here’s a real one: Scroll’s lesson on the generation effect, exactly as it appears in the app. Try the question before you read on.
The Generation Effect
You remember information better when you generate it yourself rather than passively receive it. Completing the word "psych_l_gy" is more memorable than reading "psychology." Teaching a concept to someone else cements it more than re-reading notes. Active generation forces deeper processing.
Source Slamecka, N.J., & Graf, P. (1978). The generation effect.
Quick quiz
Which study method BEST exploits the generation effect?
That’s the whole unit, about 36 seconds of reading and one question answered from memory. The lesson’s source is Slamecka and Graf’s 1978 paper, and the effect has held up since: a 2007 meta-analysis of 86 studies by Bertsch and colleagues put the average benefit of producing an answer yourself, instead of reading it, at about 0.4 standard deviations, mostly in word-list experiments.
What the evidence supports
Studies of microlearning itself
Microlearning has mostly been studied as staff training and university teaching, and mostly with knowledge tests. These are the reviews we read and checked on October 17, 2026.
| Review | Who learned what | Studies | What it found | Main limit |
|---|---|---|---|---|
| De Gagne and colleagues (2019) | Health professions education | 17 | Positive effects on knowledge and confidence | Health professions only |
| Wang and colleagues (2020) | People learning to look after their own health | 26 | Knowledge improved in 91% of studies that measured it; behavior in 36% | 25 of the 26 papers at moderate to high risk of bias |
| Senadheera and colleagues (2024) | University students | 5 of 12 pooled (654 students) | Higher theory exam scores than students taught in longer sessions | Five studies, scored on different scales |
| Monib and colleagues (2025) | Mixed settings, papers from 2020 to 2024 | 40 | A generally positive effect on learning outcomes | No pooled effect size |
| Hunter and colleagues (2026) | Hospital staff learning geriatric care | 15 | Staff knowledge improved, for example about recognizing delirium | Most studies had no control group |
Read together, the reviews say short units tend to raise scores on knowledge tests, on a small and uneven base of studies. None of them tested Scroll.
Most of the reason to expect short units to help comes from two older and larger lines of research, on habits that a short unit makes cheap: recalling what you read, and coming back to it after a gap.
Answering a question beats reading again
In a 2006 experiment by Henry Roediger and Jeffrey Karpicke, college students read short prose passages, then either read them again or wrote down what they could remember. A week later, the students who had practiced recall remembered more. Tested five minutes after studying, the rereaders were ahead (81% to 75%), and the students who read the passage most often were the most confident they’d remember it.
Rowland’s 2014 meta-analysis pooled 159 comparisons of testing against restudying and found an average effect of g = 0.50, which means the average person who practiced recall scored higher than about 69% of the people who restudied. When people saw the right answer afterward, the effect rose to g = 0.73. A 2017 meta-analysis by Adesope and colleagues reached the same conclusion about practice tests. Most of these studies were lab tasks with students, and the results varied a lot from one study to the next.
Coming back after a gap
Spacing means leaving time between study sessions. In a 2006 review of 317 experiments, Cepeda and colleagues found that the best gap grows with how long you need to remember the material. In their 2008 study of more than 1,350 people learning trivia-style facts, the best gap between two sessions was about 20% to 40% of the time until the test when the test was a week away, falling to about 5% to 10% when it was a year away. For a test in a week, that works out to a review after about 1.4 to 2.8 days.
When Dunlosky and colleagues rated 10 popular study techniques in 2013, only two got their top rating of high utility: practice testing and spacing out study. Rereading and highlighting, which students report relying on heavily, were rated low utility.
A short unit makes both of the high-rated techniques cheap. Answering a question after a 40-second lesson takes a few seconds, and going back to one lesson three days later is quicker than going back to a chapter. The unit’s length doesn’t do either for you, though. If a short explainer has no question and you never see it again, you’ve read it once, and none of this research is about reading once.
How fast unreviewed material fades is the subject of our guide to the forgetting curve.
What the evidence doesn’t support
Pages about microlearning often quote exact gains with no study attached. One training-industry article says splitting lessons into smaller pieces gives a 17% gain in “learning efficiency” and credits a 2017 study in the Journal of Applied Psychology, without a title, authors or a link. When we looked for the source on October 17, 2026, we found the figure credited to that journal on some pages and to a German university on others, and no paper behind either. Other round figures for how much more people remember after microlearning lead back to vendor blogs too. If a microlearning statistic comes without a named study you can open, treat it as unknown.
Two older claims travel with it. One is a chart saying people remember 10% of what they read and 90% of what they teach. Letrud and Hernes (2018) found versions of that chart going back more than 160 years and concluded the percentages didn’t come from research. The other is that people can now concentrate for about 8 seconds, often given as the reason lessons have to be short. The figure appeared in a 2015 Microsoft Canada marketing report, which credited a statistics website. When the Wall Street Journal and the BBC looked into it in 2017, the sources that website listed had no record of the research, and attention researchers told them attention depends on the task.
Short lessons also do more for knowing than for doing. In Wang and colleagues’ review, knowledge improved in 91% of the studies that measured it, while behavior changed in 36% of the studies that measured that. A lesson on the Rule of 72 can teach the arithmetic; using it on your own savings is a separate step.
Microlearning compared with traditional courses
“Microlearning versus traditional learning” usually means short, separate units against a course or class with longer sessions in a set order. The few direct comparisons favor short units on knowledge tests. Senadheera and colleagues pooled five university studies in 2024 (344 students taught with microlearning, 310 taught in longer sessions) and found the microlearning groups scored higher on theory exams. That’s a small base, and the exams were scored on different scales.
For someone choosing how to learn, the formats do different jobs. A course gives you an order to follow, longer stretches of practice and, if there’s a teacher, feedback on your mistakes. Short units fit into spare minutes and are quick to repeat days later, which is where recall and spacing come in. The two also combine well: a course or a book for a new subject, plus a few minutes a day of questions on what it covered.
When microlearning is the wrong tool
Short units are a poor way to learn a complex skill from scratch, such as programming, statistics, a musical instrument or holding a conversation in a new language. In each of those, later steps depend on earlier ones, and progress comes from long stretches of practice with feedback on what you got wrong.
Say Tomás wants to learn to code. A 40-second lesson can tell him what a loop is. Writing a program that works means writing a lot of programs that don’t, finding out why and fixing them, and that doesn’t fit between two quiz questions. A course with exercises, or a book and a project of his own, is the better first tool, and short lessons can come later to review the terms and ideas it taught.
The research doesn’t settle this either way. The reviews above mostly measured knowledge tests, so they say little about building a complex skill, and the one that looked at behavior found it changed far less often than knowledge did.
Short units suit ideas that stand on their own (a cognitive bias, a date in history, how GPS works out where you are), reviewing what you learned somewhere else, and the few minutes you’d otherwise spend on a feed. If that’s what you’re after, a five-minute daily routine puts recall and spacing to work, and our comparison of microlearning apps covers the apps built on short lessons, Scroll included.
Drafted with AI assistance and checked against the sources listed on October 17, 2026. Lesson counts and lengths come from the app's lesson catalog.