On the Evidence — Cognitive Dissonance

Sep 29, 2026

Learning & Change

By Kavi Arasu

Ink drawing of a blank coin balanced on a wooden peg on a white background.

One dollar.

That was the fee. In the late 1950s, some students were paid one dollar to lie. They had to tell a stranger that a boring task was fun. Others were paid twenty dollars to tell the same lie.

Later, both groups were asked what they really thought of the task. The twenty-dollar group said it was boring. The one-dollar group said it was quite enjoyable.

The smaller the bribe, the more the liar came to believe the lie. That result made Leon Festinger’s name, and it still unsettles anyone who manages people.

The man and the flying saucers

Festinger was a social psychologist trained under Kurt Lewin, the founder of modern group research. He had a taste for bold predictions and awkward evidence.

Before the famous experiment, he did something stranger. In 1954 he and two colleagues quietly joined a small American group convinced that a flood would end the world on 21 December, and that a flying saucer would rescue the faithful. They wanted to watch what happens when a belief meets hard reality.

The saucer never came. The world did not end. Instead of abandoning the prophecy, the most committed members became more certain. They began phoning newspapers and recruiting. Festinger published the account as When Prophecy Fails (1956), and a year later set out the full theory in A Theory of Cognitive Dissonance (1957).

The idea

Cognitive dissonance is the mental discomfort of holding two ideas that clash. The clash might be two beliefs. More often it is a belief and a behaviour: “I am honest” and “I just lied.”

Festinger argued that people are driven to reduce this discomfort, much as hunger drives them to eat. There are three main routes. You can change the behaviour. You can change the belief. Or you can add new thoughts that soften the clash (“everyone does it”).

His key insight was that behaviour is often the hardest thing to undo. It has already happened. So the belief moves instead.

How the experiment worked

The proof came in 1959, in a study with his student James Carlsmith. Male undergraduates spent an hour on two deliberately tedious jobs: loading spools onto a tray, then turning 48 square pegs a quarter-turn at a time.

The experimenter then asked each student to help out. Would they tell the next participant, actually a planted assistant, that the task was enjoyable? Some were offered one dollar. Some were offered twenty, a substantial sum at the time. A control group did the task and told no lie.

Later, a separate interviewer asked everyone how enjoyable the task had been, on a scale from −5 to +5. The control group and the twenty-dollar group rated it slightly negative. The one-dollar group rated it clearly positive.

The logic is neat. Twenty dollars is a good enough reason to fib. One dollar is not. Those students had no external excuse, so they found an internal one: perhaps it wasn’t so boring after all. Festinger called this insufficient justification.

The limitations deserve naming. The sample was small, around 60 young men at one university. A few were excluded for being suspicious. And “enjoyment” was a single self-reported rating.

What happened next

The theory drew fire almost at once. In 1967 Daryl Bem proposed self-perception theory. People, he said, feel no inner tension at all. They simply watch their own behaviour and infer their attitudes, the way an observer would.

Later work pushed back. In 1994 Andrew Elliot and Patricia Devine measured participants’ feelings directly and found real psychological discomfort after acting against their views. The tension appears to exist.

Other researchers refined the conditions. Joel Cooper and Russell Fazio (1984) argued dissonance bites hardest when people feel personally responsible for a harmful outcome. Elliot Aronson added that the sharpest clashes threaten our self-image as decent, competent people.

Then came an uncomfortable twist. In 2024 a large, pre-registered replication led by David Vaidis tested this family of experiments across 39 laboratories in 19 countries, with nearly 4,900 participants. People wrote essays arguing against their own views, some feeling free to refuse and some feeling they had no choice. Theory says only the free choosers should shift their attitudes. Both groups shifted. The broad idea that acting against a belief can move the belief survives. The precise mechanism the textbooks teach looks far shakier. There is a certain irony in a field defending its most beloved finding for sixty years.

The pattern since

The organisational version usually involves commitment. Once people have acted publicly, they defend the act.

Sport supplies the purest case. Lance Armstrong denied doping for more than a decade before confessing in 2013. Each denial raised the cost of the next admission. He was no longer defending a few races. He was defending a cancer survivor’s comeback, a charity, and a public identity. He later said the “perfect story” had kept him defending the lie.

This is Festinger’s mechanism at full scale. The behaviour could not be undone, so the story around it had to hold.

Before anyone notices

From the inside, dissonance feels like conviction. A leader who championed a project finds the critics oddly ill-informed. A team that worked weekends on a launch decides the lukewarm market simply needs time. Nobody feels they are rationalising. They feel they are being loyal and resilient.

The tell is this: the reasons for continuing get stronger as the evidence gets weaker.

This is the sixteenth post in a series on research that changed how we understand organisations and the people in them. Each one takes a single paper and asks what a working leader might do with it, drawing from management, organisational behaviour, sociology and psychology. The selection criterion is simple: it has to have been right about something important, and mostly ignored in the places that needed it. The previous post on self-efficacy is here.