How to do it
- Sit, eyes closed, attention on your breath. The scan study used breath focus.
- When it wanders, notice, then come back. The noticing and the coming back are the exercise, not a failure.
- Ten minutes, most days, for two months before you judge it. Brief daily practice showed at eight weeks, not four.
- No app needed. A timer and a chair.
In your brainTap a word for what it means
- Default modeDefault modeA network of linked brain areas, mostly along the middle of the brain, that gets busier when your attention turns inward: memories, plans, daydreams. It quietens when you focus on a task in front of you.Read morethe mind wanders
When experienced meditators drifted off the breath, the brain's resting network, the one busy with daydreams and worries, was more active.
- Saliencenotices the drift
At the moment they caught themselves, areas that flag what matters right now were more active.
- PrefrontalPrefrontal cortexThe front part of the brain's outer layer, behind the forehead, sometimes called the brain's CEO. It handles planning, decisions, self-control and keeping emotions in check, and works worse when you are short on sleep.Read morebrings you back
Returning to the breath and staying there used the brain's control network, including parts of the prefrontal cortex.
- Youa bit less anxious
Over eight weeks, anxiety fell a little. That the wander-notice-return cycle is what does it is an inference.
Why it works
Meditation is a workout for the part of your attention that decides what to hold and what to drop. Evidence: worth trying Across a large review of trials it gives small to moderate reductions in anxiety and low mood. Where it was compared with other active treatments, it didn't beat them. Claims beyond that are mostly oversold. Evidence: decent
No app needed. A timer and a chair. The wandering is the exercise, not the failure.
The long readEvery step of the chain, each study with its numbers, and what's still unproven. 3 min read.
What happens in your brain
Your mind wanders. In a scan study, 14 experienced meditators focused on their breath and pressed a button whenever they realised their mind had drifted. In the seconds before the press, while they were drifting, the default mode network, the brain's resting network that runs daydreams and worries, was more active. Evidence: decent
Something notices. Around the button press, areas of the salience network, which flags what matters right now, were more active. Evidence: decent
The front of the brain brings you back. As they shifted back to the breath and held it there, parts of the executive network, including the prefrontal cortex, were more active. Evidence: decent Scans track blood flow, so these areas were busier, not switched on or off. These were experienced meditators, and the study shows what happens in one sitting, not what changes with practice. Evidence: decent
You get a little less anxious. Over eight weeks of training, anxiety and low mood fell a little in trials. Evidence: decent That repeating the wander-notice-return cycle is what causes it is an inference. Evidence: worth trying
The proof
- A review of 47 randomised trials with 3,515 people, all with an active comparison, found mindfulness programmes gave small improvements in anxiety and low mood at eight weeks, a bit smaller at three to six months. Evidence for less stress was weaker, and for better attention, sleep or positive mood it was thin or absent. No programme beat drugs, exercise or other therapies. Evidence: decent
- Most people in those trials had a health problem: anxiety, depression, insomnia, chronic pain, smoking or drinking, and a range of medical conditions. The programmes typically ran 20 to 27 hours of training over eight weeks. Evidence: decent
- People new to meditation did 13 minutes a day of guided meditation or listened to a podcast for eight weeks. At eight weeks the meditation group had better mood, attention and memory and less anxiety in a stress test. At four weeks there was no difference yet. One study. Evidence: worth trying
- 218 people new to meditation were randomly given an eight-week mindfulness course, a matched health course, or a waiting list. Brain scans before and after found no structural change from the mindfulness course. Evidence: decent
How to do it
- Sit, eyes closed, attention on your breath. The scan study used breath focus. Evidence: decent
- When it wanders, notice, then come back. The noticing and the coming back are the exercise, not a failure. Evidence: worth trying
- Ten minutes, most days, for two months before you judge it. In the brief-practice trial, 13 minutes a day showed at eight weeks, not four. Evidence: worth trying
- No app needed. A timer and a chair.
Watch out
Ten minutes a day is less than was tested in most trials; the big review's typical programmes ran 20 to 27 hours over eight weeks. Evidence: decent The gains are small and were mostly measured in people with a health problem. Evidence: decent
You'll read that meditation rebuilds your brain or shrinks the amygdala. The amygdala claim comes from 26 stressed people whose drop in stress went with a change in amygdala density: a correlation, no control group. Evidence: decent The largest controlled trial, with 218 people, found no structural change at all. Evidence: decent