You pull up a zipper without thinking. The teeth close neatly behind your hand, as if the object understands its own purpose. You have used zippers for years. You trust them with jackets, bags, suitcases, jeans.
Now try to explain exactly how one works.
Not what it does. Not why it is useful. The steps. The small mechanical sequence that turns two separated edges into one locked seam.
That little hesitation — the moment confidence thins into fog — is the doorway into one of the more quietly humbling findings in cognitive psychology.
When Recognition Pretends to Be Understanding
In 2002, Leonid Rozenblit and Frank Keil gave this experience a name: the illusion of explanatory depth. Across twelve studies, they found that people often overestimate how well they understand causal mechanisms, especially the ordinary systems they use all the time.
Their examples were wonderfully mundane: zippers, toilets, speedometers, helicopters, piano keys. These objects feel familiar enough to invite confidence, but mechanically opaque enough to expose the gap.
Participants first rated how well they understood an object. Then they tried to explain how it worked. Then they rated their understanding again. After explanation, confidence dropped.
The object had not changed. The mind had simply tested its own model.
This is the strange thing about everyday life: it usually asks us to use things, not explain them. Turn the handle. Press the button. Flush the toilet. Open the laptop. If the system responds, our brain receives a small dose of evidence that we know what we are doing.
But use is not the same as understanding. A toilet flushes, yes. But where does the suction come from? What starts the siphon? When does the tank refill? At what point does your explanation become a vague gesture toward plumbing?
Rozenblit and Keil found that the illusion was stronger for causal explanations than for facts, stories, or simple procedures. That makes sense. Mechanisms have joints. One step has to connect to the next. A label can float. A causal chain has to hold.
The Mind Stores Handles, Not Blueprints
This illusion is not stupidity. In many ways, it is efficiency.
A mind that stored the full mechanism behind every zipper, router, traffic light, immune response, mortgage rate, and refrigerator would be buried under its own inventory. So we do something more elegant: we store handles.
We remember where to look, whom to ask, what category something belongs to, and roughly what it is for. Rozenblit and Keil argued that we often confuse access to knowledge in the world with knowledge stored inside our own heads.
That distinction matters. You may not know how Wi-Fi turns a signal into a loaded webpage, but you know someone does. You may not understand every step in vaccine development, but you know there are institutions, experts, procedures, and papers behind it. Human intelligence is partly communal. We borrow competence from other minds constantly.
The trouble begins when borrowed understanding feels private.
That is why the question “how exactly?” can feel so different from “do you understand?” The first question asks for a mechanism. The second often asks for a feeling.
And feelings of knowing can be persuasive. They arrive quickly, with a bodily ease. Speed feels fluent. Fluency feels true. Once something feels true, the mind may stop checking whether the steps are really there.
Why Explanation Changes Political Confidence
In 2013, Philip Fernbach and colleagues brought the illusion of explanatory depth into politics. Instead of asking people why they supported a policy, they asked them to explain how the policy would produce its intended effects.
That difference is subtle and powerful.
Reasons can become armor. If someone asks why you believe something, you may gather moral claims, personal experiences, party identity, or objections to the other side. But mechanisms behave more like stress tests. They ask: what happens first, what follows from that, and what has to be true for the promised outcome to occur?
Fernbach’s team found that after people attempted these explanations, many rated their understanding lower. They also reported some moderation in policy attitudes, though later researchers have debated how reliably that attitude shift replicates.
So the lesson is not that explanation magically makes everyone reasonable. Psychology is rarely that neat, and politics is even less cooperative.
A smaller lesson survives: asking for the causal chain can reveal where certainty is outrunning comprehension.
A policy may be morally clear even when its mechanism is complicated. Explanation is not the only form of understanding. But when we argue about what a policy will do, not just what it represents, mechanism matters.
The Conditions That Make the Illusion Stronger
Recent research has added texture to the picture.
In 2024, Körner, Schütz, and Petersen tested power across three preregistered studies with 607 participants. They found that power increased general overconfidence, but had only a small impact on the specific illusion of explanatory depth. That distinction is useful. Feeling broadly capable is not the same as believing you can explain a mechanism.
Think of the manager who speaks smoothly about software architecture, supply chains, or inflation. Ease can sound like depth. Status can make it sound even deeper. But explanation has a colder requirement: connect the parts.
Another 2024 study in Memory and Cognition examined cognitive load. The illusion grew when mental resources were strained. Poorer performance on concurrent memory tasks was associated with a larger explanatory-depth illusion. Higher analytical processing, by contrast, related to a lower illusion.
That fits everyday life a little too well. Some of our strongest opinions form when we are tired, rushed, crowded, or overloaded. The mind reaches for the smoothest available story and mistakes that smoothness for structure.
A useful private test is almost painfully simple: first this happens, then this causes that, then this changes the next thing. If your verbs become foggy — affects, impacts, somehow leads to — you may have found the gap.
A Quieter Kind of Confidence
The illusion of explanatory depth has been studied beyond gadgets: health improvement, climate protection, mental disorders, historical events, and financial systems like inflation. In each case, the same question waits underneath the familiar label: could I rebuild the chain without skipping the invisible parts?
This is not a call to distrust yourself about everything. It is a call to become more precise about what kind of knowing you have.
Try three columns: what I know, what I infer, and what I am borrowing from someone else’s expertise. In teams, ask for mechanism before commitment: how will this produce the promised change? In private, use “how exactly?” as a humility test, not a weapon.
The zipper was never really the point. It was a small object that revealed a larger human habit: confidence borrowed from use, repetition, and social knowledge.
Understanding may not always feel like louder certainty. Sometimes it feels like confidence getting quieter, more specific, and more willing to ask for help.
