How you can learn in a world of information overload | Tania Lombrozo | TEDxNewEngland

Main ideas, concepts, and lessons

• Rubber ducking as a learning mechanism

  • Software engineers “rubber duck” by explaining a problem to a rubber duck step-by-step.
  • The duck doesn’t respond, so the benefit comes from explaining to oneself, not from receiving new external information.
  • This is presented as an instance of “learning by thinking.”

• Learning by thinking (core concept)

  • Learning by thinking is about starting with knowledge already inside your head.
  • Your mind contains “vast repositories” of knowledge in memories and skills.
  • By thinking through a problem, you can extract, transform, and combine what you already know to produce something you couldn’t produce before.

• Why it helps during information overload

  • With more internet tools, social media, and AI, people risk information overwhelm.
  • A common cause is approaching learning by immediately looking outside yourself for more information.
  • The talk argues you should often start from within to avoid overwhelm and to build a clearer target for what external information is actually needed.

• Examples to show learning by thinking

  • Quick questions (e.g., windows on a house, 5th letter of the alphabet, left/right position of “Z” on a keyboard) illustrate:
    • You likely didn’t know instantly, but you could reason from stored knowledge to answer.
    • This reasoning reflects the process of using existing knowledge to solve new questions.

• The “transformation and combination” model

  • Knowledge you have may not match what you need directly (like baking with missing ingredients).
  • In cooking, you transform/combine what you have (e.g., convert chocolate and adjust butter with salt).
  • In learning, you similarly re-represent information:
    • Example: instead of storing “windows” as a single fact, you might have visual memories; thinking can transform them into the single numeric answer.

• A key limitation: thinking alone sometimes isn’t enough

  • If the information truly isn’t present in your mind (e.g., knowing the exact number of windows in someone else’s house), you need new information.
  • However, asking about it can still set you up to learn efficiently once you get there—so learning by thinking is often the best starting point.

• Seeking explanations as the strongest near-general strategy

  • Research on effective students found that what differentiates top performers isn’t just access to information, but how they use it.
  • Successful students explain to themselves, ask questions, and attempt good answers.
  • This is called the self-explanation effect: explaining to yourself improves learning and broader transfer.

• Experimental illustration (ducks/frogs task)

  • Adults tend to match “same with same” representations (e.g., two frogs with two ducks).
  • Young children often do something else due to how they represent the reference card (e.g., matching based on partial features like “a duck”).
  • When children were prompted to explain why the cards went together (with no feedback), they began matching like adults.
  • Main takeaway: explanation changes how information is represented, enabling correct reasoning.

• Explaining sets you up to learn (illusion of explanatory depth)

  • When someone tries to explain something, they often realize they don’t understand as well as they thought.
  • This is the illusion of explanatory depth (people feel explanatory mastery that collapses under questioning).
  • The talk uses Roxana’s example:
    • She knew general concepts about tornadoes (pressure differences) but couldn’t explain details of how low pressure creates the funnel.
    • Explaining revealed the gap, which then indicates exactly what missing information to seek.

• Why this matters with AI and external tools

  • AI/chatbots and other tools can provide impressive explanations.
  • The talk warns they can replace learning by thinking if you start by outsourcing understanding.
  • If you start internally, you better police boundaries between:
    • what you actually know vs.
    • what you merely have access to externally.
  • Otherwise, you may fall prey to illusions of understanding, thinking you know when you don’t.

Method / instruction list (as presented)

• Start with what’s inside your head first

  • Don’t begin by immediately searching for external answers.
  • Begin by using existing memory/skills to attempt the reasoning.

• Ask yourself questions and explain to yourself

  • Take time to explain the problem step-by-step in your own words.
  • Be your own “rubber duck” (no external “answer” needed).
  • During explanation, pause to ask:
    • “What do I think I know?”
    • “Why does that follow?”
    • “What detail am I missing?”

• Use explanation to reveal gaps

  • If you get stuck, treat it as diagnostic:
    • the point where explanation breaks often marks missing information.
  • Translate the gap into a targeted question about what to look up next.

• Only then seek external information (when needed)

  • Shift from internal thinking to external research after you have:
    • identified what you don’t know, and
    • clarified what information would resolve the gap.
  • Use external tools/AI to fill the specifically identified missing pieces—not to replace the thinking process.

• Aim for insight rather than easy learning

  • Recognize that learning by thinking may be effortful.
  • The payoff is a structured path through information overwhelm toward genuine understanding.

Speakers or sources featured (all identified)

• Tania Lombrozo (speaker; cognitive scientist)
• Roxana (friend mentioned; her young son’s tornado question is discussed)
• “Scientists” / “research studies” (referenced as unnamed groups; no specific study authors named)