7.9 — When Meditation Goes Wrong

A woman with a history she has never discussed with anybody starts a mindfulness app. Ten minutes, eyes closed, attention on the breath.

On the fourth day, in the silence, a memory arrives with a force she has not felt in fifteen years. She stops the session shaking, concludes that something is wrong with her, and does not tell anyone — because everything she has read says this practice is calming and safe.

Nothing was wrong with her. What happened is documented, has a mechanism, and is common enough that adapted forms of the practice exist specifically for it. She was not told, because the material she was given did not mention it.

The scale of it

Adverse effects from meditation are under-reported for two reasons: the studies frequently did not ask, and participants tend to attribute a bad experience to their own inadequacy rather than to the practice.

When researchers have asked systematically, unpleasant experiences turn out to be common. Reviews and surveys find a meaningful minority of practitioners reporting effects such as increased anxiety, low mood, depersonalisation, intrusive images and disturbed sleep. Reported rates vary widely with the population and the intensity of practice, and in most cases the experiences are transient. A smaller number are lasting and disabling.

This does not mean meditation is dangerous. It means it is an active intervention, and active interventions have adverse effects.

What actually happens, and why

Four distinct mechanisms, each with a different remedy.

Removing the distraction

The most common. Ordinary life is full of external input, and that input keeps difficult material out of awareness. Sitting still with your eyes closed removes it, and what has been held back arrives.

For most people this is mild — awareness of a low mood they had been outrunning, or a worry that was being managed by staying busy. For people with unprocessed trauma it can be severe, because what has been kept at bay is a great deal more than a worry.

Attention turned on distress

For someone in a severe depressive episode, turning attention inward and observing thoughts can amplify rumination rather than interrupting it (5.5). This is why Mindfulness-Based Cognitive Therapy was designed for people in remission between episodes, not for people who are currently severely depressed — a detail almost always lost when the technique is recommended generally.

Depersonalisation

Practices that involve observing your own thoughts and the constructed nature of the self (7.5, 2.7) can produce a sense of detachment from yourself, your body, or reality. In small doses this is the intended insight. Sustained, it is distressing and it is a recognised condition — the feeling of watching yourself from outside, of things being unreal, of not quite being present.

People with a prior tendency to dissociate are most at risk, and intensive practice increases it.

Intensity

A ten-day silent retreat is a much stronger intervention than a daily ten minutes, and adverse effects cluster there. Many hours of practice, little sleep, no external contact, no distraction — that is a set of conditions that would produce unusual mental states in most people even without the technique.

Who should be careful

Anyone with a trauma history. The highest-risk group, and the most likely to have been recommended meditation as a coping strategy. Standard instructions — eyes closed, still, attention inward, no talking — replicate several features of being unable to escape, which is precisely the wrong set of conditions.

Anyone currently in a severe depressive episode. Behavioural activation and treatment first (14.2); this later.

Anyone with a history of psychosis, or a family history of it. Intensive practice can destabilise, and this is a genuine contraindication rather than a caution.

Anyone prone to dissociation or depersonalisation.

Anyone with panic disorder, specifically for breath-focused practice, where attending to breathing frequently makes it feel effortful and irregular, and those sensations feed the panic loop (4.2). Use a different object.

None of these people are excluded from practice. They need a different form of it, and someone to tell.

Trauma-sensitive practice

The adaptations are specific and they work. If any of the above applies to you, start here rather than with the standard instructions.

Eyes open, with a soft gaze on a point in the room. This keeps you oriented in the present and it is the single most important adaptation.

Choose the object deliberately. External anchors — sound, the feeling of the chair, the soles of the feet — are safer than internal ones. Avoid the breath and avoid the torso in the body scan; hands and feet are far safer starting points.

Short. Three to five minutes. Length is what allows material to surface.

Movement instead of stillness, where stillness is difficult. Walking practice (7.7) is a full practice and it is much safer for many people.

Permission to stop, stated in advance to yourself. I can open my eyes and stand up at any point. Having the exit available is what makes staying tolerable — the same principle as anywhere else.

And an anchor to return to. Feet on the floor, hands on your thighs, one object in the room you can look at and name.

What to do if something difficult arises

In the moment:

Open your eyes. Immediately. This is not failure.

Orient to the room. Name five things you can see, out loud if you are alone. Feel your feet on the floor. Say today's date.

Move. Stand up, walk, get water, put your hands under a cold tap.

Stop the session. It has finished.

And then tell somebody. This is the one people skip. The instinct is to conclude you are broken and say nothing, which is exactly what happened to the woman at the top of this page.

Afterwards:

If it was mild and passed, return to a shorter, adapted version — eyes open, external anchor, three minutes.

If it recurs, or if it does not settle within a day, stop the practice and speak to a professional. Something is being surfaced that needs handling with support (Part 15). Meditation is not the treatment for trauma. Specific therapies are (14.4), and they are effective.

And if depersonalisation persists, stop the practices that involve observing the self, use grounding and physical activity instead, and get an assessment. It is treatable and it is worse if it is left.

The teaching problem

Two things worth saying about how this is usually presented.

Meditation is marketed as universally beneficial and free of risk, which is not true of any active intervention and which leaves people unprepared for a normal outcome.

And "just observe it, do not react" is bad advice for a flashback. It is good advice for an itch, a distracting thought, or ordinary discomfort. Applied to traumatic re-experiencing it means staying in something you should be exiting, and the instruction to observe without reacting is exactly what keeps people sitting through it.

The correct instruction for that situation is: stop, open your eyes, orient, move. If a teacher or an app tells you otherwise, they are wrong.

What to do with this

Know before you start whether you are in one of the higher-risk groups, and if you are, begin with the adapted form rather than the standard one. Eyes open, external anchor, three minutes, permission to stop.

Treat a difficult experience as information rather than as failure. It tells you the intensity or the form is wrong, or that there is something that needs proper attention.

Never do a long silent retreat as your first serious practice. Build a daily practice for a year first, and choose a retreat where somebody experienced is available and paying attention.

And tell someone if something surfaces. The most damaging part of the story at the top of this page was not the memory. It was the fifteen months she spent afterwards believing she had failed at the one thing that was supposed to help, and telling nobody.

Next: 7.10 closes Part 7 with the only problem that actually matters for most people — not which technique, but how to still be doing it in a year.