Observe your thoughts without judging them: 5 concrete exercises

You observe a thought without judging it by treating it as an event that is happening, not as information to be processed. Spot that it is there, give it a short name, do not answer its content, come back to a sensation in the body. Four moves, in that order, as often as needed.

The nuance that changes everything sits in the word “judging”. This is not about finding the thought acceptable or interesting: it is about not entering the register of evaluation at all, neither for it, nor against it, nor against yourself for having had it. Most people fail on the second floor: they manage not to judge the thought, then blame themselves for having one.

The five exercises below run from the most mechanical to the widest. Count on three to five minutes.

Because the instrument and the object are the same thing. A thought is not in front of you like an object on a table: it takes you from the inside, and by the time you notice, you are already thinking it rather than watching it.

There is a second, more mechanical reason. Fighting a thought hands it your attention, and therefore room. That is why an instruction like “stop thinking about it” produces the opposite result, and why the beliefs that make people quit meditation are so expensive.

Brain imaging has documented two distinct ways of relating to your own experience: a narrative mode, where you tell yourself across time, and a present-centred mode, which recruits other regions and strengthens with training (Farb et al., 2007). In untrained participants, the two remain largely fused. In other words, the difficulty you run into is not an attention deficit: it is the ordinary starting point.

Sit down, close your eyes if that suits you, and wait for the first thought. When it arrives, give it a label: “worry”, “planning”, “memory”, “judgment”. One word. Never a sentence.

The one-word constraint is not a stylistic flourish. The moment you expand into a sentence, you are back inside the content, and you believe you are observing when in fact you are commenting. Putting a word on an inner state also reduces its intensity, a mechanism studied under the name affect labeling and tested as far as clinical protocols (Burklund et al., 2024).

If no word comes, take “thinking” and move on to the next one. There is no correct label.

Now take a thought that carries some weight. Not the shopping list: the one that has been circling since this morning. Say it to yourself as it is, then say it again with “I notice I’m thinking that…” in front.

“I’m going to mess up this presentation” becomes “I notice I’m thinking that I’m going to mess up this presentation”. Nothing is contradicted, nothing is corrected. You have only added a narrator, and the narrator is not the thought.

This move has a name in acceptance and commitment therapy: cognitive defusion. It belongs to a family of processes that researchers group under the term decentering, and a reference review proposes a three-component model for it, including meta-awareness and disengagement from content (Bernstein et al., 2015).

Repeat it three or four times with the same sentence. By the third repetition, most people feel the sentence lose some of its charge without having done anything to refute it. That is exactly the point.

The first two exercises work thought by thought. This one changes scale.

Settle in, and instead of following what passes, put your attention on what does not: the silence between two thoughts, the sensation of sitting, the awareness that someone here is noticing. That background does not change nature according to what crosses it, and a pleasant thought arrives in the same place as a painful one.

The classic image is the sky: clouds pass, and nobody has ever had to clean the sky for it to be sky again. A whole article explores that comparison and its limits, and shows why watching thoughts drift past like clouds (in French) works better when you do not force the metaphor.

Three minutes is enough. When a thought arrives, do not push it back: widen your attention to include what surrounds it, rather as you step back from a painting.

One honest caveat. This exercise works badly on agitated days, when the invitation to sense a wide space rings hollow. Go back to exercise 1 then, which asks for less. To take this further, our steps on how to stop trying to control everything cover the same ground from another angle.

Here is the trap that makes most people give up. You observe a thought, all is well, then a second thought appears that talks about the first one: “nice, I caught that one”, “I think I’m starting to get it”, “right, I’m hopeless at this”. The comment has become the main event.

The instruction is easy to state and tiresome to hold: treat the comment exactly like the rest. A label, one word, move on. “Evaluation”. “Self-criticism”. Then back to the breath.

Each extra floor is one more thought, not a higher level of awareness. It may be the only thing I repeat in every first class.

You will drift off. Twenty times, fifty times. That is not a failure of the practice, that is the practice.

The only move asked of you: when you notice you were elsewhere, come back. No comment, no taking yourself in hand, no starting over from the beginning. And if you want a number to hold on to, count those returns rather than the minutes. A return is a moment of lucidity, and it is the only unit that really counts here.

Last February, on a bench at Gare de Lyon with twenty minutes to spare and a train announced late, I counted thirty-one returns in eight minutes. By every usual standard the sit was bad. It is still the one I remember best.

The first sits mostly deliver an unpleasant piece of information: you measure how carried away you were without knowing it. That is a result, even if it does not look like the one you were hoping for.

Beyond that, it is worth staying careful about promised timelines. Available data suggest that a short, regular practice loosens certain cognitive automatisms: one study observed less rigidity on a classic mental-set task among people practising mindfulness (Greenberg et al., 2012). Promising a precise timeline to any given person is a step further than the evidence allows.

One field marker instead: most people notice the first change outside the sit, not during it. One day, in the middle of a tense conversation, you notice the sentence rising before you say it. Nothing spectacular. That is where the practice starts to earn its place, and it is close to what our guide on getting out of your head describes.

Observing your thoughts assumes you can stay in the presence of whatever turns up. That is not always advisable.

If the recurring thoughts are dark thoughts, traumatic memories or scenarios tied to a violent event, sitting alone in front of them can make things worse rather than calmer. Those situations call for support from a professional, not for an exercise done in your living room. The same holds during an acute panic episode: what helps at that moment is immediate sensory grounding, not an observation of the mind.

And there are ordinary days when the right decision is not to practise at all. The control you exert over your own practice is itself a form of control, and it deserves the same treatment as the rest.