You held your ground in that meeting. You answered every question. You didn’t flinch, and everyone in the room saw it. By the time you’re back at your desk, colleagues have already messaged to say you were brilliant.

And you’re not thinking about any of that. You’re thinking about the pause before you spoke. Whether your voice was too tight. Whether the way you phrased the second point made you sound defensive rather than certain. You’ll think about it again at your desk an hour from now, and again at eleven that night, lying awake running the same three seconds on a loop.

You’ve tried talking yourself out of it. You know, intellectually, that it went well. You can list the reasons. None of it stops the loop, and by the third or fourth night of this you start wondering if the problem is that you’re too sensitive, too self-critical, not built for the level you’re now operating at.

What’s actually happening

But here’s what’s actually happening, and it has nothing to do with sensitivity.

In 1999, two psychologists studying self-focused attention found that turning your attention inward isn’t one thing. Some of it is driven by curiosity, a genuine interest in understanding yourself, and this kind reliably produces insight. But some of it is driven by threat: something has happened that puts your sense of self at risk, and your attention keeps returning to it. This second kind doesn’t produce insight. It just repeats. Trapnell and Campbell called the first reflection and the second rumination, and the distinction matters because they look identical from the inside, both feel like thinking, but only one of them goes anywhere.

What you’re doing at eleven at night has the shape of thinking. It isn’t. It’s rumination, and rumination is threat-driven by definition. Your nervous system has flagged the meeting as unresolved, and it keeps sending you back to check on it.

What replaying it can’t do

Here’s the part that explains why checking doesn’t help. One increasingly influential account in neuroscience, still actively debated, not settled fact, holds that perception itself works less like a camera and more like a best guess: the brain forms a prediction about what’s happening, and that prediction only updates when new information arrives. If that account is even roughly right, it explains something specific about the replay. Running the same three seconds again isn’t new information. It’s the same data your nervous system already has, presented for the fortieth time. A prediction can’t update on data it’s already seen. The scene you’re replaying was never going to look different on the fortieth pass than it did on the first, and no number of additional passes will change that.

Why your nervous system flagged it in the first place

Which raises the obvious question: why did your nervous system flag a meeting where you were, by every external measure, excellent? This part is on firmer ground. One of the most robustly supported findings in psychology is that bad registers more heavily than good . Bad feedback, bad impressions, and threatening information all get processed more thoroughly than their positive counterparts, and this holds up even when researchers control for the obvious confounds. It isn’t a flaw in your particular wiring. It’s what the system does with everyone. Your brain gave more weight to the half-second pause than to the four days of colleagues telling you that you handled it well, because that is what a threat-monitoring system is built to do. It was never going to weigh those two things equally.

None of this means the loop is permanent. It means the loop won’t end by running it again, however many times, however carefully. It ends when something changes what the system has to work with, which is a different activity than the one you’ve been doing at eleven at night, even though from the inside they can feel exactly the same.