There's a cool psychological test on loss-justification.

Basically: you give people a multiple choice online form, asking them stuff.

On one question, regardless of how they answer, the answer is always switched to another answer.

So, for instance, if they are asked on a scale of how much they like X, the form will make them accidentally say the opposite.

The questions and answers don't matter so much as what is actually being observed: after believing themselves to have made an untrue statement, how do people respond?

What's found is interesting:

- many people, upon realizing the form put down the opposite of what they'd intended to express, begin to hedge their original opinion. They are more likely to change their opinion later.

It's believed that this arises from trying to resolve the cognitive dissonance produced. They justify their misdirected answer, as "there must be a reason I'd say that."

In previous such experiments back before the internet, you'd get people to read lies aloud. Then, test them to see whether those lies influenced their original beliefs.

Result: repeating a lie, even knowing it is a lie, makes those original beliefs more malleable.

Like: preface the lie with a statement that it is a lie. Stick it in dialogue, where the person saying it says it's a lie. The original attribution vanishes, the act of saying the falsehood is enough to begin justification.

In modern tests, you can dial in better on what is actually going on.

And what seems to happen, is that a falsehood requires cognitive effort to disbelieve.

Cheap belief, expensive doubt.

Being in the proximity of a lie, reading or hearing a lie, that has little justificatory effect.

*Repeating* the lie, physically expressing I, (even if by accident), is what begins the justification process.

Even accidentally expressing the falsehood, such as misreading text and accidentally saying "does" instead of "does not," seems to increase the amount of effort necessary to doubt.

You *want* to believe what you say and do, and do not want to doubt yourself.

So you use a dark pattern and get someone to accidentally click YES instead of NO. Or vice versa.

You design your form to be inscrutable, making people unsure whether they are answering the opposite way.

Getting people to express things they do not believe, even seemingly by accident, causes them to be more willing to believe those things.