SciSpace Academy
TutorialLiterature review › Find research gaps and new ideas

AI and Cognitive Decentering: Thinking Differently About Research

Prof. Erik CambriaEnglishLiterature reviewFind research gaps and new ideas

TL;DR

1-min read

This talk is for scientists, academics and doctoral researchers interested in how AI can widen the way they frame research problems and generate ideas. It is given by Prof. Erik Cambria, Professor of Artificial Intelligence at the College of Computing & Data Science, Nanyang Technological University (NTU), Singapore, Provost Chair in Computer Science and Engineering, Visiting Professor at the MIT Media Lab and an IEEE Fellow. He draws on his open-access paper "Artificial Intelligence as the Fourth Decentering Revolution", published in Cognitive Computation by Springer.

Key takeaways

  • The tutorial presents AI not only as a way to finish tasks faster but as a thinking partner that can move researchers beyond entrenched habits of thought.
  • Decentering describes shifting away from a single, habitual human viewpoint so that new questions and connections become visible.
  • AI can support this shift by offering alternative framings, links between fields or interpretations a researcher might not reach alone.
  • Many good ideas come from reframing a problem rather than collecting more data, which makes this lens especially useful in the early, exploratory stages of research.
  • Suggestions from AI are starting points that must be checked against the existing literature for novelty and feasibility and refined through the researcher's own expertise.
  • The researcher still decides which angles are sound, so AI widens the options rather than making the decision.
Apply it

To apply this in SciSpace, explore related work around your question to surface alternative framings, then test any promising angle against the literature before committing to it.

How to do it

  1. Understand AI as decentering

    Learn the idea of AI as a fourth decentering revolution and why it matters for science. This frame shifts AI from a speed tool to a thinking partner.

  2. Notice your habits of thought

    Identify the familiar mental habits that shape how you frame your research problems. Seeing them clearly is the first step to moving beyond them.

  3. Use AI as a cognitive collaborator

    Ask AI to look at your research problem from unfamiliar angles. Using AI to generate new research ideas works best when you treat it as a collaborator.

  4. Reframe your research question

    Try reframing your question before collecting more data. A new framing can reveal questions that more data alone would not.

  5. Read the underlying paper

    Read the open-access paper "Artificial Intelligence as the Fourth Decentering Revolution" for the full argument. It deepens the conceptual lens for early, exploratory research.

Watch the tutorial51 min

AI and Cognitive Decentering: New Ways to Frame Research ProblemsPlay video: AI and Cognitive Decentering: New Ways to Frame Research Problems

Chapters

  1. Welcome and talk overview
  2. Will AI replace white-collar jobs and human creativity?
  3. What is the plagiarism singularity? LLMs and the infinite monkey theorem
  4. What is AI overload? Why students and workers feel overwhelmed
  5. Careless AI use: leftover ChatGPT prompts in emails and reviews
  6. Is bigger AI better? Efficiency vs brute-force scaling
  7. AI as humanity's fourth narcissistic wound
  8. How is human thinking different from ChatGPT? Human vs AI cognition
  9. Is artificial general intelligence (AGI) possible?
  10. Should AI have emotions? Bias, the uncanny valley and AI limits
  11. Is AI dangerous? Paperclip scenarios and unintended consequences
  12. Can AI generate research hypotheses? Seven pillars and the Moos AI scientist
  13. Symbol grounding, world models and trust in AI
  14. Will AI replace jobs? Why human-in-the-loop still matters
  15. Interpretability vs explainability in AI
  16. How should researchers start using AI? AI is just a tool
  17. Affective computing and emotion prediction research
  18. Q&A: Can AI develop physical empathy without a body?
  19. Do LLMs hold hidden knowledge? Is hallucination useful for creativity?
  20. Who is responsible when AI causes harm? AI accountability
  21. Can AI make you smarter? AI as a tutor vs over-reliance
  22. Closing and how to get in touch

Try this on SciSpace

Frequently asked questions

What is cognitive decentering in the context of AI?

In this tutorial, decentering describes moving away from a single, habitual human perspective. AI can support that shift by offering alternative framings, connections or interpretations a researcher might not reach alone. The researcher still evaluates which of those angles are sound and worth pursuing, so AI widens the options rather than making the decision.

Can AI help me come up with new research ideas?

AI can suggest connections between fields, surface unexplored angles and challenge assumptions in how a question is posed. These suggestions are starting points, not finished ideas. Each one needs to be checked against the existing literature for novelty and feasibility, and refined through the researcher's own expertise before it becomes a real project.

Which paper is this tutorial based on?

The talk draws on the open-access article "Artificial Intelligence as the Fourth Decentering Revolution" by Prof. Erik Cambria, published in the journal Cognitive Computation by Springer. Reading it alongside the tutorial gives a fuller picture of the argument that AI can expand how scientists reason about research problems.

Related tutorials