Smarkerapp — How readers capture, tag and retrieve | smarkerapp.com
What the research says about highlighting versus retrieval practice, and when a highlight is worth converting into a question instead.
Mostly no — not by itself. In Karpicke and Blunt's 2011 experiment, students who read a passage once and then re-highlighted it recalled about 36% on a delayed test, while students who practised retrieval recalled about 80%. Highlighting is exposure; memory is built by retrieval.
The gap exists because re-highlighting feels like work while demanding nothing from memory. The pen moves, the page looks familiar, and fluency is mistaken for learning — which is exactly why the delayed test, not the same-day one, shows the difference.
On narrow screens, swipe or scroll the plate sideways.
This does not make highlights useless; it reassigns their job. A highlight's job is to be findable — that is what tags and exports are for. The remembering job belongs to the question note: one line, in your own words, phrased as a question the highlight answers.
Convert sparingly. A long reading session can produce 300 or more highlights; turning all of them into questions would take longer than the reading. Pick the few you would actually want to recall in a month — five to ten per session is a workable budget — and let search handle the rest.
The monthly loop ties it together: audit tags with the routine on the main page, confirm the export with the storage page, and re-answer last month's questions before writing new ones. Retrieval practice only works if the retrieval actually happens.
Further reading