Do concept maps help you learn?

What a concept map is

A concept map is a diagram of ideas (nodes) joined by labelled links: aldosterone increases sodium reabsorption; monopoly causes deadweight loss. Each link reads as a small statement, which is what separates a concept map from a mind map, where branches radiate from a centre and often carry no label at all. Practitioner guidance from the people who developed the method treats the labelled statement as the unit of meaning, and suggests that around 15 to 25 concepts are usually enough for one map [1].

What the meta-analyses found

Two large reviews agree that maps help, with an important difference between building and studying one.

• Nesbit and Adesope (2006) pooled 55 studies with 5,818 learners from primary school to university [2]. Overall, learning with maps had a moderate effect (g = 0.60). Building a map helped more (0.82) than studying one someone else made (0.37). Maps helped recall of main ideas much more than details (0.60 against 0.20).
• Schroeder and colleagues (2018) pooled 142 effect sizes from 11,814 learners [3]. Building a map again helped more (0.72) than studying one (0.43), and for university students studying a map the effect was 0.32. The kind of map, the subject (STEM or not) and working alone or in groups made no clear difference.

Two caveats from the reviews themselves. Effects look much larger against weak comparisons, such as a lecture, than against an equally active alternative such as writing a summary, where they drop to around 0.2 to 0.5 [2, 3]. And the benefit may come partly from the engagement of working on the map rather than from the map itself [2].

When retrieval beats mapping

The best-known challenge to maps is Karpicke and Blunt’s 2011 study in Science [4]. Students either built a concept map with the text open or practiced recalling the text from memory. A week later, the recall group scored 67% against 45%, including on questions that required inference, and even when the final test was to draw a map.

Follow-ups refined the picture:

• Building a map from memory worked as well as writing out what you remember, and both beat restudying [5]. The benefit came from the retrieval, not the map format.
• Mapping and then retrieving did not beat retrieving alone, despite taking longer [6].
• A larger 2023 study that gave both groups equal time found no difference on verbatim questions and a small advantage for mapping on inference questions (d = 0.36) [7].

So “retrieval beats mapping” is contested, but the practical lesson is clear: a map you copy with the source open is not much better than rereading. A map you have to reconstruct, complete or explain from memory is retrieval practice in a useful shape.

What makes a map worth having

• Statements, not just words. Each link should read as a specific, explanatory sentence. A map asked to answer “how does a car work?” produced far more dynamic, explanatory links (55%) than one asked “what is a car?” (21%) [8].
• A hierarchy for beginners. Learners new to a topic do better with a clearly organized structure; experienced learners can cope with, and benefit from, a looser network [9].
• Main ideas, not everything. Maps help most with the main ideas and relationships [2]. Cutting a map from 77 to 50 nodes cost learners nothing in one study [10], and very large diagrams become hard to read at all.
• Cross-links between parts. The links that join different sections are where the map shows understanding, rather than a list drawn as boxes.

How to use a concept map with a lecture

1. Read for understanding first. A map helps organize ideas you already follow.
2. Fill gaps from memory. Cover a link or a concept and say what belongs there before looking.
3. Explain the links aloud. For each arrow, say why it holds. If you can’t, that is what to study.
4. Connect across lectures. The most valuable links are often between topics taught weeks apart, which no single lecture puts together.
5. Come back to it. Revisit the map after a gap, from memory, the way you would review cards.

Where the evidence stops

Most studies are short, use researcher-made tests and measure recall within days or weeks. There is little evidence on transfer to quite different problems, and no controlled studies yet of maps generated by AI. A map is a strong scaffold for understanding; the strongest evidence is still for what you do with it: recalling, explaining and coming back.

References

1. Novak, J. D., & Cañas, A. J. (2008). The theory underlying concept maps and how to construct and use them (Technical Report IHMC CmapTools 2006-01 Rev 01-2008). Florida Institute for Human and Machine Cognition. cmap.ihmc.us/docs/theory-of-concept-maps
2. Nesbit, J. C., & Adesope, O. O. (2006). Learning with concept and knowledge maps: A meta-analysis. Review of Educational Research, 76(3), 413–448. www.sfu.ca/~jcnesbit/research/NesbitAdesope2006.pdf
3. Schroeder, N. L., Nesbit, J. C., Anguiano, C. J., & Adesope, O. O. (2018). Studying and constructing concept maps: A meta-analysis. Educational Psychology Review, 30(2), 431–455. link.springer.com/article/10.1007/s10648-017-9403-9
4. Karpicke, J. D., & Blunt, J. R. (2011). Retrieval practice produces more learning than elaborative studying with concept mapping. Science, 331(6018), 772–775. doi.org/10.1126/science.1199327
5. Blunt, J. R., & Karpicke, J. D. (2014). Learning with retrieval-based concept mapping. Journal of Educational Psychology, 106(3), 849–858. andymatuschak.org/prompts/Blunt2014.pdf
6. O’Day, G. M., & Karpicke, J. D. (2021). Comparing and combining retrieval practice and concept mapping. Journal of Educational Psychology, 113(5), 986–997. learninglab.psych.purdue.edu/downloads/2021/2021_ODay_Karpicke_JEDP.pdf
7. Mayrhofer, R., Kuhbandner, C., & Frischholz, K. (2023). Re-examining the testing effect as a learning strategy: The advantage of retrieval practice over concept mapping as a methodological artifact. Frontiers in Psychology, 14, 1258359. pmc.ncbi.nlm.nih.gov/articles/PMC10783554/
8. Derbentseva, N., Safayeni, F., & Cañas, A. J. (2006). Two strategies for encouraging functional relationships in concept maps. In A. J. Cañas & J. D. Novak (Eds.), Concept maps: Theory, methodology, technology. Proceedings of the Second International Conference on Concept Mapping. San José, Costa Rica. cmc.ihmc.us/cmc2006Papers/cmc2006-p164.pdf
9. Shapiro, A. M. (2008). Hypermedia design as learner scaffolding. Educational Technology Research and Development, 56(1), 29–44. link.springer.com/article/10.1007/s11423-007-9063-4
10. Krieglstein, F., Schneider, S., Beege, M., & Rey, G. D. (2022). How the design and complexity of concept maps influence cognitive learning processes. Educational Technology Research and Development, 70(1), 99–118. pmc.ncbi.nlm.nih.gov/articles/PMC8788906/