Why does not knowing itch? Not every gap gets under your skin.

Image: Library of Congress Main Reading Room. Carol M. Highsmith / Library of Congress via Wikimedia Commons · public domain.

You are telling someone about a film. The actor’s face is perfectly visible in your mind: the eyebrows, the coat, the scene with the lift. His name has vanished.

Now the missing name feels less like ignorance than an obstruction. You try the alphabet. You reject three innocent actors. You open your phone, because apparently the conversation cannot continue until this one extremely non-emergency emergency has been resolved.

But you also do not know the name of the 4,000th-largest lake in Kazakhstan, the chemical composition of the dust beneath your sofa or what a stranger in Dundee ate for breakfast. Those gaps cause no distress at all.

So “not knowing is irritating” cannot be the whole answer. We do not feel the weight of everything we do not know. Something has to turn one invisible absence into the missing thing.

Ignorance is not enough. You have to see the gap

In 1994, economist George Loewenstein proposed an influential “information-gap” account of curiosity. The basic idea is almost irritatingly neat: curiosity appears when attention lands on a discrepancy between what you know and what you want to know. That discrepancy feels like a deprivation, motivating you to obtain the missing information.

The important word is perceived. An unknown fact cannot nag you until it has some shape. A question, contradiction, half-heard story or failed retrieval draws a border around the absence. Now you know enough to know exactly what you do not know.

This produces a prediction that initially sounds backwards. The people who know least about a subject may not be the most curious about it. With no foothold, there is no precise gap to inspect and no obvious progress to make. The people who already know the answer have no gap either. Curiosity should often be strongest in between.

That curve appears repeatedly in trivia studies. Curiosity tends to peak at moderate confidence: the answer is neither securely known nor completely alien. A 2024 study replicated the pattern using both established trivia questions and newer questions about COVID-19. It also found an important qualification. When participants considered the information important, curiosity remained high from very low through moderate confidence. Importance can make a large gap worth approaching.

A 2025 review called the broader pattern the “Curious U”. It found robust evidence for the curve across different tasks, but no settled agreement about its cause. Some theories emphasise uncertainty, familiarity or the expectation of success. Others propose that curiosity directs effort towards places where learning progress is likely: not the things you have mastered, and not the things so remote that your next step is unclear.

That is already a better answer. Curiosity does not simply measure the size of the hole. It responds to what the next piece might do.

The tip of your tongue is a gap with its lights on

The missing actor’s name is a particularly clean case because it comes with a metacognitive feeling: you do not merely lack the answer; you feel that you almost have it.

In a 2017 experiment, Janet Metcalfe and colleagues gave 46 students 82 general-knowledge questions. Participants had to report quickly whether an answer was on the tip of their tongue and whether they wanted to see it later. Tip-of-the-tongue states overwhelmingly predicted wanting the answer. The effect appeared when participants had produced no answer, a wrong answer and even a correct answer accompanied by the feeling of incomplete retrieval.

That last result matters. Objective ignorance was not doing all the work. A subjective signal of closeness helped turn the question into unfinished business.

The researchers described that feeling as a “goad to epistemic action”: keep searching, ask someone, look it up. That proposed function is plausible rather than proven by the experiment. But the behaviour fits. Almost knowing makes the path to relief look short enough to pursue.

Is curiosity unpleasant or enjoyable? Yes

Information-gap theory is often paraphrased as “uncertainty feels bad”. Sometimes it does. But people voluntarily read mysteries, travel without knowing exactly what they will find and conduct experiments whose outcome is the entire point. A theory that makes all curiosity miserable has misplaced the fun.

A useful distinction separates a deprivation motive from an interest or discovery motive. Deprivation says: close this gap. Discovery says: show me what is there. Both can occupy the same question. Their relative strength may help explain why curiosity sometimes resembles a splinter and sometimes resembles anticipation.

One preregistered 2021 study exposed this mixed state with lotteries. In one experiment, 34 participants rated how happy they felt about uncertain gains and losses. In another, 34 different participants rated their curiosity. Greater uncertainty increased curiosity but decreased happiness. Curiosity and pleasure moved in opposite directions.

That supports the possibility that reducing an unpleasant state contributes to some curiosity. It does not prove that curiosity is fundamentally aversive—the two judgements came from separate groups, and the authors explicitly presented the aversive-drive explanation as a hypothesis to test. Other situations can make the anticipated discovery itself enjoyable.

The answer is not that curiosity secretly feels one way. It is that the word covers motivational states with different mixtures.

An answer can be valuable even when it is useless

If curiosity were only practical, we would seek information when it improved a decision and ignore it when nothing could be changed. People do value useful information. They also break that rule constantly.

In a 2009 series of trivia experiments, higher curiosity was associated with activity in caudate regions involved in anticipated value and with a greater willingness to spend scarce tokens or wait for the answer. Participants later remembered more of the surprising answers that had made them curious.

Macaques have shown a starker preference. When offered water rewards of variable size, they preferred cues that revealed the amount in advance even though the cue could not change how much water arrived. Midbrain dopamine neurons signalled both expected water and expected information.

That is strong evidence that advance information can carry value beyond changing the outcome. It is not evidence that curiosity is literally a “dopamine hit”. Dopamine systems participate in learning, motivation, prediction and action; activity in a reward-related circuit is not a chemical translation of delight. The same caution applies to the familiar claim that ADHD is simply too little dopamine: naming a neurotransmitter rarely completes the explanation.

Even non-instrumental information may change the experience before the outcome. It can organise anticipation, reduce uncertainty, make a story coherent or prepare an emotional response. “Useless” to the final result does not mean valueless to the person waiting.

The brain does not price every answer the same

A series of five studies involving 543 participants asked why people wanted or avoided information about personal traits, finances and health. Three kinds of expected value predicted their choices:

• Instrumental utility: will knowing help me do something?
• Hedonic utility: do I expect knowing to feel better than ignorance?
• Cognitive utility: does this connect to things I think about often?

People weighted those motives differently, and an individual’s pattern showed some stability over time. The model is not a final inventory of curiosity; the study measured information choice, which can have motives other than curiosity. But it explains something a simple uncertainty account cannot: the same available fact can look irresistible to one person and mercifully avoidable to another.

Expected emotional impact is particularly disruptive to the comforting idea that humans just want the truth. In another experiment, people were willing to pay for information about favourable outcomes—and willing to pay to remain ignorant about unfavourable ones. Brain activity in mesolimbic and orbitofrontal regions tracked partly different aspects of those choices.

So uncertainty does not issue an automatic command to look. The likely answer, the cost of knowing and the value of not knowing enter the calculation.

Then why do we look at things we expect to hate?

Because emotional cost does not always win.

In four experiments published as “The Pandora Effect”, participants exposed themselves to aversive sounds, images and even mild electric shocks to resolve uncertainty without an apparent practical benefit. Other work has found deliberate choices to view images involving death, violence or harm. A 2023 series of four studies, with 407 younger and older adults, found that higher uncertainty selectively increased exploration of expected negative information; the same effect did not appear for neutral or positive information.

These are boundary results, not proof that humans prefer bad news. People frequently avoid threatening health or financial information. What the experiments show is that expected negativity and uncertainty are separable pressures. Under some conditions, not knowing what the bad thing is can be more motivating than knowing that the answer will be unpleasant.

Curiosity is therefore neither a dependable guardian of wellbeing nor a pure truth detector. It can improve learning, lead to useful warnings, waste an afternoon or persuade someone to press the button labelled “probably unpleasant”.

The itch is a selective valuation, not a universal alarm

Why can not knowing feel so irritating?

Because occasionally ignorance stops being background darkness and becomes a represented, valued gap. You have enough knowledge to see its edges. You expect the answer to be reachable. It may help you act, relieve an unresolved state, connect to something important or deliver the pleasure of discovery. Those benefits begin competing with the time, effort and possible emotional cost of finding out.

The irritation is most plausibly one part of that competition: the deprivation side of curiosity. It cannot explain the delight of exploration, the avoidance of bad news or the many unknowns that never become mentally visible.

The stranger finding is that knowledge does not merely close gaps. It creates them. Learning enough about a subject reveals distinctions and missing pieces that a novice could not see. The expert may possess far more answers and experience far more precise ignorance.

So the better question is not “why does the brain hate not knowing?” It is “why did this one gap earn a feeling?”