Published on 2026-07-25
Why can't you resist 'you won't believe what happened next'?
A 1994 theory of curiosity explains the itch behind that headline, and a 2021 study found the trick barely beats a plain, honest one.
You are scrolling. A headline says 'you won't believe what happened next', and your thumb moves before your brain gets a vote. This is not a failure of willpower on your part. It is a lever, built to move you exactly like that, and the lever has a name.
The itch only a click can scratch
In 1994, the Carnegie Mellon psychologist George Loewenstein published a theory of curiosity that still shapes how researchers think about attention today. His argument: curiosity is not simple interest, it is a kind of deprivation. The moment you notice a gap between what you know and what you want to know, your mind treats the missing piece almost like an unscratched itch, and closing that gap feels like relief.
The strange part, the part that headline writers lean on hardest, is that total ignorance barely triggers curiosity at all. You need to already know something. A review of curiosity research published by the US National Library of Medicine describes this as a 'priming dose': a small piece of information sharpens your sense of exactly what you are missing, and that sharpened sense is what pulls you forward. 'What happened next' only works because you already know what happened first.
A headline that withholds on purpose
Once you see the mechanism, the whole genre reads differently. A curiosity gap headline is a small, deliberate withholding: enough setup to make the missing detail feel close and specific, never enough to let you guess it and move on. The gap is not an accident of bad writing. It is the entire design brief.
The headline never lies to you about the itch. It just never tells you where the relief actually leads.
Does the trick still work as well as it used to?
Here is where it gets interesting, because researchers actually measured this rather than assuming it. In 2021, a team from Penn State, Michigan State and the University of Maryland, led by S. Shyam Sundar, ran two experiments testing eight kinds of headlines, including classic curiosity gap phrasing, against plain descriptive ones. The first study had 150 participants, the second had 249. In both, clickbait headlines did not dramatically outperform the traditional ones, and in several comparisons people were actually more drawn to the plain, informative headline than to the teasing one.
That does not mean the gap has stopped working on any given person in any given moment: plenty of us still click. It suggests something else, that after years of exposure, a lot of readers have started recognising the shape of the trick, and recognising a trick is often enough to blunt it. The gap format has been used so often that it now advertises itself. 'You won't believe' has become a kind of watermark.
There is a small irony buried in the Penn State findings. The researchers also tried to teach artificial intelligence models to spot clickbait automatically, feeding them 175 headlines to sort. The four models agreed with each other only 47 percent of the time. If software trained specifically for the job cannot reliably tell a gap headline from a straight one, it is easier to forgive yourself for occasionally falling for a well built one. The mechanism is subtle enough to confuse a machine built to catch it.
Where the click actually goes
It helps to ask, calmly, what the gap is actually for. Most of the time the payoff on the other side is thinner than the promise: a mundane detail dressed up as a twist, an anticlimactic list entry, an article padded to keep you scrolling past a few more ad slots. The gap rarely exists to inform you faster. It exists to hold your attention exactly long enough for something else on the page to be counted, viewed, or sold.
None of this means curiosity is a design flaw in you. It is one of the more useful things your mind does, the same drive that pushes you to finish a good mystery or learn how a magic trick was done. The gap headline just borrows that drive without asking, and spends it on your attention rather than your understanding. Once you can name the itch, you get to decide for yourself whether the promised relief is worth the click, instead of finding out you already agreed before you noticed you'd been asked.