Short answer: upload your own lecture notes to NotebookLM and select Mind Map, or paste them into ChatGPT or Gemini and ask for a mind map outline using only your notes. Use the result as an overview of how the topic fits together, then close it and redraw the map from memory, check it against the AI version, and quiz yourself on the branches you missed. Research comparing concept mapping with retrieval practice found that pulling information from memory produced more learning than mapping with the material in view.
Which AI tools can make a mind map?
NotebookLM has a built-in Mind Map. Google announced it on its Workspace Updates blog on 19 March 2025 as "a new interactive Mind Map" that lets users navigate complex topics in their notebooks and explore new connections, available to NotebookLM and NotebookLM Plus users. The New School's help article on the feature explains how it works: open a notebook with your sources, select the Mind Map chip in the chat, and the map appears as a note in the Studio panel. You can expand or collapse branches, zoom and scroll, select a node to ask questions about that topic in the chat, and download the map as a file.
ChatGPT, Claude and Gemini do not need a special feature. Paste in your notes and ask for a mind map as a nested outline: a central topic, main branches, then sub-branches. You can copy the outline onto paper or into a mind-mapping app, or ask for it as Markdown or Mermaid code if your app imports either.
Whichever you use, start from your own lecture material, not "the topic" in general. Your exam covers what your course taught, and a map built from a general prompt can wander well beyond it.
What does the research say about mind maps and memory?
Maps are not useless. Nesbit and Adesope (2006) pooled experiments in which students learned by constructing, modifying or viewing node-link diagrams: 67 effect sizes from 55 studies with 5,818 participants, from Grade 4 to university. Across many conditions, using concept maps was associated with better knowledge retention, though the size of the benefit ranged from small to large depending on how the maps were used and what they were compared with.
The catch comes when maps are compared with retrieval practice. Karpicke and Blunt (2011), in Science, found that practising retrieval produced greater gains in meaningful learning than elaborative studying with concept mapping. The advantage showed up on questions that tested comprehension and asked students to make inferences, and it held even when the final test itself involved creating a concept map.
Notice what NotebookLM is doing when it makes a map for you: it does the organising. Google's own study guide for NotebookLM describes the feature as generating interactive visual diagrams "to help you grasp the overall structure" of your material. That is a fair description of what it is good for. Grasping the structure on screen is not the same as being able to rebuild it in an exam. For the full research story, including the debate since 2011, see do concept maps help you study.
How do you make a mind map with AI, step by step?
- Gather your own material. Lecture slides, your notes, the set reading for one topic. One lecture per map keeps it readable.
- Generate the map. In NotebookLM, add the sources and select Mind Map. In ChatGPT or Gemini, paste the notes and use a prompt from the next section.
- Check it against the source. Click through the branches. Delete anything your course did not cover, and fix any link that does not match your notes. AI tools can merge, split or mislabel ideas.
- Use it once as an overview. Spend a few minutes tracing how the main branches connect. This is the step where an AI map genuinely saves time.
- Close it. Then go to the retrieval steps below. The map you just looked at is now your answer key.
Prompts for a mind map from your notes
- "Using only my notes below, make a mind map as a nested outline: one central topic, four to seven main branches, and no more than four sub-branches each. Use short phrases, not sentences."
- "For each link between branches, add a short label saying how they are related, such as 'causes', 'is a type of' or 'is regulated by'."
- "List any idea in my notes that does not fit on the map, and any branch you added that my notes do not mention."
- "Using the map, ask me one question at a time about each branch. Wait for my answer before giving the next one, and quote the line from my notes that backs each answer."
Labelled links make the map far more useful for revision, because "photosynthesis, glucose" says little while "produces, using light energy" says what you know. The third prompt is your accuracy check. The fourth turns the map into a quiz; our guide to getting ChatGPT to quiz you has more like it, and NotebookLM users can follow how to make a quiz in NotebookLM.
How do you turn an AI mind map into retrieval practice?
- Redraw it from a blank page. Central topic first, then every branch and link you can remember. Do not peek.
- Mark it against the AI map in a different colour. Add what you missed and correct wrong links. The second colour is a map of your gaps.
- Quiz yourself on the coloured branches. Turn each gap into a direct question and answer it from memory, or use the quiz prompt above.
- Redraw again a few days later. Another blank page, then compare. Each redraw is spaced retrieval, not another look at a finished diagram.
This is the same idea as the blurting method, drawn as a diagram rather than written as a list. The AI map saves you the time it takes to build a tidy reference version; the learning happens when you try to rebuild it.
Once you know which branches are weak, direct questions on them help. Ace (acenow.app) is a study app that turns your own lecture PDFs, slides or YouTube lectures into multiple-choice questions checked against the source. Depending on the material, you may also get put-the-steps-in-order questions for processes, and you pick which chapters to study, so you can aim practice at the branches you keep missing. It is free to try on the web and iPhone.
Frequently asked questions
Can NotebookLM make a mind map?
Yes. Add your sources to a notebook and select the Mind Map option in the chat. The map appears in the Studio panel, where you can expand branches, select a node to ask about it, and download it.
Can ChatGPT make a mind map?
It can produce a mind map as a nested outline, or as Markdown or Mermaid code for a mind-mapping app. Ask it to use only your notes, and check every branch against them.
Are AI mind maps good for revision?
They are a quick overview and a useful answer key. On their own, looking at one is close to rereading. Redrawing it from memory and checking it turns it into retrieval practice, which research on concept maps favours.
Is it better to make a mind map yourself or with AI?
Use both. Let the AI build an accurate reference map from your notes, then make your own version from memory. Your version is where the learning happens; the AI version is how you check it.
Sources
- Karpicke, J. D., & Blunt, J. R. (2011). Retrieval practice produces more learning than elaborative studying with concept mapping. Science, 331(6018), 772–775.
- Nesbit, J. C., & Adesope, O. O. (2006). Learning with concept and knowledge maps: A meta-analysis. Review of Educational Research, 76(3), 413–448.
- Google Workspace Updates (19 March 2025). New features available in NotebookLM and NotebookLM Plus.
- Hu, C.-W. (3 April 2025). 3 ways to level up your studying with NotebookLM. Google blog.
- The New School IT Knowledge Base. Mind Maps in NotebookLM.



