On this page
In 2011, Jeffrey D. Karpicke and Janell R. Blunt of Purdue University published a short paper in Science that compared two "active" ways of studying the same science text: building a concept map with the text open, and closing the text and writing down everything you can recall.
What did Karpicke and Blunt test, and on whom?
The first experiment had 80 undergraduate students at Purdue University in the United States, according to the paper. Each read a science text and was randomly assigned to: read once, read four times, build a concept map with the text open, or practise free recall (write what you remember, reread, recall again).
A U.S. Department of Education What Works Clearinghouse review adds the details: the text was a 276-word passage on sea otters, and the concept-mapping and retrieval groups each got the same total time. A week later, everyone took a short-answer test with verbatim questions (ideas stated directly in the text) and inference questions (several ideas joined together).
The second experiment, in the same paper, tested 120 students, each mapping one text and recalling another, so every person was compared with themselves. Half took a short-answer final test; the other half took a final test in which they had to draw a concept map from memory.
What did the study find?
Retrieval practice is the act of pulling information out of memory, for example by writing down everything you can recall after closing the book.
One week later, students averaged 67% after retrieval practice, 45% after concept mapping, 49% after repeated reading and 27% after reading once, as summarised by the What Works Clearinghouse.
In Experiment 1, retrieval practice beat concept mapping by about one and a half standard deviations (d = 1.50), a large effect (Karpicke and Blunt, 2011).
It held for inference questions, not only facts.
In Experiment 2, retrieval practice still won when the final test itself was drawing a concept map (d = 1.01), and 101 of 120 students (84%) did better on the text they had recalled than on the text they had mapped (paper, Table 1).
Asked after studying, 90 of 120 students (75%) predicted concept mapping would work as well or better.
Does this mean concept maps are useless?
No, and the authors say so. The paper notes that concept mapping is not inherently an elaborative task: if you draw the map without the material in front of you, it becomes retrieval practice. A follow-up by the same lab, Blunt and Karpicke (2014) in the Journal of Educational Psychology, found that recalling in paragraph form and drawing a concept map from memory both beat extra studying a week later, and did not differ from each other.
Other limits:
Who was studied. U.S. college students, short passages, one lab session; not Indian school students or a full syllabus.
Time. The WWC review cautions that the mapping and retrieval groups had more study time than the read-once and reread groups, so those two comparisons are not like-for-like.
Delay. The final test was one week later. Mintzes and colleagues questioned the study in Science; as summarised in the authors' reply, they raised the one-week delay and minimal concept-map training; the authors answered that concept mapping is usually described as quick to learn.
Has anyone replicated it?
Mostly, yes. A Spanish team, Lechuga, Ortega-Tudela and Gómez-Ariza (2015), ran a similar study. We have not read that paper itself; in a 2024 paper, the same researchers describe it as essentially replicating the advantage for retrieval, though for students already experienced with concept maps the advantage showed up only on verbatim questions, not inference ones. In O'Day and Karpicke (2021), recall again beat concept mapping a week later, and mapping first and then recalling added nothing over recall alone, despite taking more time.
The 2013 Dunlosky review of ten study techniques rated practice testing among the most useful, and a 2018 meta-analysis on practice tests and new questions asks whether the benefit reaches unfamiliar problems.
How can a JEE or NEET student use this?
This is our inference, not something tested on Indian students: keep your maps, but build them from memory.
Read a section once, carefully (say, digestion or Newton's laws).
Close the book. On a blank page, draw the concept map from memory: key terms as boxes, arrows labelled with how they connect.
Open the book and mark what you missed in a different colour.
Two or three days later, redraw the map on a fresh page without looking at the first one.
In physics, chemistry and maths, recall the method before looking at a worked solution.
For structure, see this guide to concept maps for science; just draw with the book shut.
After you have attempted a problem from memory, the Fermi Problem Library has worked problems with a step-by-step solver that explains the reasoning rather than printing an answer, so you can see where your attempt went off track.



