Evidence check

In the early 1960s, the psychiatrist Aaron Beck was studying depression at the University of Pennsylvania. He expected to find hidden, unconscious conflicts. Instead, his patients kept describing something much closer to the surface.

Alongside what they said out loud ran a quick, almost unnoticed stream of thoughts: "He's bored of me." "I always get this wrong." Beck called them automatic thoughts. They felt like facts. When he and his patients looked at them carefully, many of them didn't hold up.

That observation became cognitive therapy, now usually called CBT, and checking the evidence for a thought is one of its oldest tools.

An evidence check treats a thought as a hypothesis to test. You gather what supports it and what doesn't, the way a fair-minded friend or a good detective would.

The hard part is being strict about what counts. Worry is good at passing off other things as evidence:

• Feelings: "I feel like they're annoyed" is a feeling about the evidence, not evidence.
• Mind reading: "She must think I'm useless" is a guess about someone else's head.
• One memory standing in for many: "Last time it went badly" can crowd out the five times it went fine.

The goal isn't to force a positive thought. It's to land on a view that fits all the facts. Often that view is calmer simply because it's more accurate. Sometimes the evidence does support the worry, and that's useful too: now you know what to fix.

Do it on paper if you can. In your head, worry repeats itself. On paper you can see when the "evidence" is really a feeling or a guess.

Theo suggested a weekend hiking trip in his friends' group chat. Two hours later, nobody has replied, though two people have reacted to someone else's message about a game. He's sure: "They don't actually want me around." Last month the same group invited him to a birthday dinner, and one of them texted him yesterday to ask for a book recommendation. Last year, one of them did cancel plans with him at the last minute.