A wandering mind is the normal setting, not a fault. In one large sampling study, people were thinking about something other than what they were doing in 46.9% of moments.

Stress changes what that wandering is for. The body raises its guard, and thinking becomes the tool it uses to hunt for the threat and for a way out of it.

That is why the thoughts feel urgent and repetitive rather than creative. The search is looking for danger, so it keeps returning to the same unfinished thing.

You rarely notice it starting. It arrives as a next thought, not as a decision, which is why people describe it as a mind that will not stop rather than one they turned on.

The cost is physical, and it has been measured. Pooling 60 studies of healthy people, worry and going over things raised blood pressure and heart rate and lowered a marker of the body’s calming brake.

In daily life the same shows up hour by hour. Diary studies found heart rate rose during worry episodes, and work worry moved it most of all.

The loop keeps itself going. The scan finds no answer, so the guard stays up, and the raised guard is exactly the state that starts the next round of scanning.

That is why it outlasts the event. Repetitive thinking keeps the stress response running before something happens and long after it is over, which is the part worth breaking.

Start with the baseline, because it changes how this reads. A smartphone study sampled 2,250 adults at random moments and found their minds were elsewhere in 46.9% of them, and in at least 30% of moments during every activity.

So a busy mind is not the problem. What separates ordinary drifting from a racing mind is not how much thinking happens, it is whether the thinking closes. Wandering moves on. Stress thinking returns.

Researchers group worry about what is coming and going over what already happened under one heading, because they behave the same way in the body. Both keep the stress response switched on when nothing is actually happening.

That single idea explains most of what follows. The unfinished search is the engine, and it runs on whether or not the thing being searched for is still in the room. It is also why the same evening replay can feel like problem solving and achieve none.

Two systems carry the stress signal. A fast nerve path lifts heart rate and alertness in seconds, and a slower hormone path keeps the body ready for much longer. Thinking sits on top of both.

You do not need a real threat to start it. In one study, 89 people gave a short speech to an evaluative panel and nothing else happened to them. That alone produced more going-over afterwards than the non-evaluative versions of the task.

It was the thinking, not the speech, that predicted the size of the response. The people who went over the talk most in the quiet ten minutes afterwards had the highest and longest-lasting rise in a stress hormone.

A second study of 59 people found the same shape and added one detail worth knowing. Bringing the memory back two weeks later restarted part of the response, which is the clearest evidence that the thought, on its own, is enough to move the body.

The common objection is that this is only real for people in crisis. The daily-life measurements say otherwise. In 73 teachers wearing monitors for four days, ordinary worry episodes raised heart rate by about two beats a minute against neutral hours.

Work worry raised it by about nine, and heart-rate variability, a marker of the body’s calming brake, fell during the same hours. None of these were crises. They were normal working days logged hour by hour.

There is also a lag, which is the part people miss. In the same group, worry raised heart rate in the following hour and was still measurable two hours later, while the stressful events by themselves showed no such effect.

So the size of the worry is a poor guide to the size of the cost, and the moment you stop worrying is not the moment your body stops. Low-grade looping through a whole afternoon is not a smaller version of the problem, it is a longer one.

Most racing happens in gaps. Commuting, showering, waiting for a reply, lying still before sleep. The task drops away, the guard is still up, and the search fills the space that just opened.

Then the amplifiers, which matter because they are adjustable. Caffeine is the clearest. In a crossover study of 12 volunteers, 500 mg produced more tension, nervousness and restlessness than 250 mg, and worse performance with it.

Sleep sits underneath everything. In 100 students, both short sleep and a late sleep pattern went with more repetitive negative thinking, so late nights and late screens load the next day before it starts.

Hormones belong here too. Tracking 2,956 women for ten years, those who were calm before menopause were more likely to report high anxiety symptoms through the transition. Stress is one cause among several, and the others are worth ruling in.

Repetition is the mechanism, which is the hopeful part. A pattern the nervous system learned is a pattern it can learn differently, and the trials aimed squarely at repetitive thinking are the strongest evidence we have here.

The clearest one added a therapy built specifically around going-over to usual care for 42 people with stubborn low mood. Symptoms and remission rates improved, and the improvement tracked the drop in going-over rather than anything else.

Parking beats pushing. Pooling seven trials and 999 people, setting a short fixed worry period each day and postponing worry to it reduced daily worrying, which is a remarkably small change for the effect.

Broader calm practice helps less than the marketing suggests but more than nothing. Across 47 trials and 3,515 people, meditation programmes had moderate evidence for anxiety and were not better than other active treatments.

One belief deserves retiring: that you can push a thought out of your mind. In the original experiment nobody managed it, and the people told to suppress a thought later reported more of it than people who had never been asked to.

Pooling the controlled studies, that rebound is real but small to moderate, and it varies with the thought and the way it is measured. So a racing mind is not proof your brain is faulty. It is a search being fed.

Then the finding that cuts the other way. In 2023, 120 adults across 16 countries were trained for three days to suppress fearful thoughts. No rebound appeared, and anxiety and low mood fell, with the mood benefit still there three months later.

The everyday tools are unsettled too. One online worry-postponement trial with 361 finishers did not beat simply logging the worry. Hold the mechanism firmly and any single technique loosely, including ours.