The Illusion of Knowing: Why Do We Think We Understand Everything?

In a discussion, the most confident speaker can seem like the person who knows the most. Their sentences flow, their explanation is ready and so is their answer to an objection. Yet a basic question may remain: can we actually explain how the thing we are discussing works?

The illusion of knowing offers a way to think about the gap between feeling that we understand and demonstrating that understanding. It is not a psychiatric diagnosis. Calling someone an “empty can” does not provide a scientific measure of their mind, either.

The issue concerns our own explanations as much as other people’s certainty. How can we notice where an explanation is well supported and where something is missing?

Recognising something and explaining it are different tasks

Knowing what a bicycle is differs from explaining how its parts work together. Being familiar with a psychological term similarly does not mean that we understand the methods and limits of the research behind it.

In Rozenblit and Keil’s 2002 experiments, participants’ ratings of how well they understood complex systems could fall after they attempted detailed explanations. The researchers examined this as an illusion of explanatory depth.

An everyday exercise is to choose a familiar concept and explain it in three short steps. Why does the first step lead to the next? If we can only repeat the same word when asked about the connection, we may have identified a gap in our account.

This need not become a trap designed to expose ignorance. It can help identify which part of an explanation needs further learning.

Understanding also need not be all or nothing. We might know a useful practical procedure while being unable to explain every mechanism behind it. Naming that limit is more informative than pretending to have a complete account.

Can we mistake another person’s knowledge for our own?

Much of everyday life relies on shared knowledge. Repairing a device, translating a text or researching a health question does not require every person to know every detail independently. Drawing on expertise is an ordinary part of learning.

In Sloman and Rabb’s 2016 experiments, believing that scientists understood a phenomenon and that their explanation was accessible could increase participants’ ratings of their own understanding, even without receiving the explanation itself. The finding illustrates how the boundary between other people’s knowledge and our own can become blurred.

A practical question follows: “Can I construct this explanation myself, or do I know whom I could ask?” Both abilities are valuable, but they describe different things.

Trusting a specialist also differs from assuming that we have become a specialist in the same field. We can make responsible use of expertise while remaining clear about where we depend on it.

Finding an answer online and learning from it

Fisher, Goddu and Keil’s 2015 experiments found that searching online for explanations could increase people’s estimates of their own knowledge. The study concerned the possibility of confusing access to information with personal understanding.

This does not mean that using the internet makes learning impossible. It invites a more precise question: have I located an answer, or can I explain the reasoning behind it?

When preparing an article, for example, you might close the source and summarise its main point in your own words. You can then return to the source and compare what you wrote with what it actually says. This is a working method for checking comprehension, not an intelligence test.

The same source-reading questions can be asked of an answer from a search engine or an AI tool. Where is the original source? Which finding supports which claim? Has uncertainty been preserved? These are practical questions, not a claim that the 2015 experiments tested present-day AI tools.

How should we assess a confident claim?

The volume of someone’s voice, the length of their sentences or their use of technical vocabulary cannot establish their intentions or mental health. Examining a particular claim and its support is more useful.

Questions that can help include:

  • What exactly is being claimed?
  • Which evidence or source supports it?
  • Does that source investigate the same question?
  • Under what conditions would the explanation be inadequate?
  • Could new evidence change the conclusion?

If someone cannot provide a source, we need not turn that observation into a verdict on their personality. We may simply conclude that the explanation offered at that moment lacks sufficient support.

Applying the same standard to an idea we like matters, too. If checking sources becomes a test reserved for people we disagree with, gaps in our own thinking can remain unnoticed.

These questions do not require challenging every sentence in an ordinary conversation. They become particularly useful when an important conclusion depends on a specific claim being true.

A short exercise for testing your own understanding

The following is a note-taking exercise rather than a clinical test.

1. State the claim in one sentence

Instead of writing that a topic is “very deep”, specify the idea you are defending. A clear statement makes it easier to see what information is needed to assess it.

2. Build a chain of explanation

Explain how one step connects to the next. Separate examples and analogies from evidence. An analogy can make an idea easier to picture without proving that the underlying claim is true.

3. Mark the gap

Saying “I don’t know this part” does not make the whole idea worthless. It can identify the question worth researching next. A useful note may be a question with a clear scope, rather than a long list of everything you have yet to learn.

4. Consider a counterexample

Ask whether the explanation applies in every situation. Check whether you have extended one example to all people. Our article on quick learning and impatience also discusses the different stages between grasping an idea and becoming skilled.

If the exercise turns into endlessly repeating the same question, taking a break or choosing a specific source question may be more helpful. Our article on overthinking examines when reflection helps us move forward.

A philosophical perspective on defending an idea

This section offers an interpretation of how we approach knowledge, rather than an experimental finding. Taking a view seriously need not mean treating every objection as an attack on the self. We can ask whether a criticism identifies something that would improve the explanation.

Accepting that we might be mistaken also does not require treating every view as equally well supported. Some explanations have stronger evidence. What matters is not inferring that strength solely from the speaker’s confidence.

Recognising the limits of our knowledge need not diminish the value of what we say. It may help us make a more careful statement: “I understand this part; I need further sources for that part.” Such a statement can keep both thought and conversation open.

Frequently asked questions

Is “empty can syndrome” a scientific diagnosis here?

The expression is not used as a diagnosis. The article examines how particular claims are supported, rather than labelling people.

Does being very confident mean someone must be wrong?

No. Confidence alone does not establish whether a claim is true or false. Its evidence and reasoning need to be considered.

Do we have to become experts in everything?

No. We can draw on expertise while distinguishing our own knowledge from the knowledge we rely on others to provide.

Sources

  • Rozenblit, L., & Keil, F. (2002). The misunderstood limits of folk science: an illusion of explanatory depth. Cognitive Science, 26(5), 521–562. https://doi.org/10.1207/s15516709cog2605_1
  • Sloman, S. A., & Rabb, N. (2016). Your Understanding Is My Understanding: Evidence for a Community of Knowledge. Psychological Science, 27(11), 1451–1460. https://doi.org/10.1177/0956797616662271
  • Fisher, M., Goddu, M. K., & Keil, F. C. (2015). Searching for explanations: How the Internet inflates estimates of internal knowledge. Journal of Experimental Psychology: General, 144(3), 674–687. https://doi.org/10.1037/xge0000070
WhatsApp

Leave a Comment

Your email address will not be published. Required fields are marked *