The day was fine. The moment her head hit the pillow, it started.
Meera runs a mid-sized logistics company in Bangalore. Her days are structured and full. She handles decisions, conversations, and deadlines without significant difficulty.
By primary measures, she is functioning well. Her team considers her measured and reliable under pressure. She considers herself reasonably calm during working hours.
At night, something different happens. The moment she lies down, a loop begins. The conversation with a vendor that resolved fine this morning gets replayed, with the version where she said something different.
The email she sent yesterday at 5 p.m. gets second-guessed. An upcoming presentation that is two weeks away starts generating scenarios: questions she will be asked, moments where the argument weakens, the faces of the people in the room. None of these scenarios are new.
She has run versions of them before. New ones appear to take their place.
She tries different things. She reads before sleep. She avoids screens.
She keeps a notepad nearby to write down whatever feels unresolved, hoping to empty her mind. The notepad sometimes helps for a few minutes. The loop returns.
She has been sleeping five to six hours a night for two years. She is not anxious during the day. She is exhausted at night.
What Meera is describing is a pattern that uses the night as its uninterrupted operating window. During her workday, tasks and real conversations compete with the loop for her attention. She does not notice the engine running as clearly because something else is always occupying the foreground.
When that foreground disappears at night, the engine is the only thing left. The loop does not start at night. It runs all the time. At night, nothing is covering the sound of it.
The standard advice for how to stop overthinking at night addresses the sound. It does not address the engine. This is why Meera keeps waking up after two years and a stack of techniques that each work for a week before the loop reasserts itself.
Why the simulation engine runs harder when input disappears
The simulation engine is the mind's anticipatory function. It was installed to scan ahead, model outcomes, and prepare responses before situations arrived. In the environments where it developed, this function had real utility.
The person who could anticipate what was coming had an advantage. The pattern that built this engine did so because the engine was useful.
The problem is not the engine. The problem is that the engine has no input throttle connected to the importance of the current content. It generates scenarios based on availability and emotional weight, not based on whether the scenarios serve any purpose.
At 2 a.m., the primary emotionally weighted material available is whatever felt unresolved during the day. The engine processes it. Then it processes adjacent material.
Then it constructs variations. The generation is automatic and does not require a decision to begin.
During the day, Meera's prefrontal cortex has genuine competing inputs. A vendor call requires attention. A Slack message needs a reply.
A team member walks in. Each of these events temporarily displaces the engine's output from conscious attention. The output is still being generated; it simply does not reach the foreground because something else is there first.
This is why overthinking before sleep is not always the experience of a particularly stressful person. Meera is not more anxious than average. She is average.
The engine runs for everyone. At night, without competition, she notices it clearly.
Racing thoughts at night have a specific character. They are not random. They concentrate on situations that have some element of unresolved evaluation attached: something where the outcome was uncertain, something where she might have handled it differently, something where a future event carries weight.
The engine is not selecting these at random. It is selecting for material where a simulation seems like it might produce useful output. The engine is wrong about this.
The simulations at 2 a.m. do not produce useful output. They produce the next simulation.
Kevin, in the video above, demonstrates the same engine operating in a different domain. His social anxiety was not stronger on high-stakes days only. The pattern ran continuously, amplifying as his career grew and provided more material for the engine to generate scenarios about.
The pattern did not slow down when things were going well. It accelerated because the stakes were higher and the engine had more consequential material to work with. This is the same dynamic Meera experiences: more success, more material, more overthinking at night.
What the conventional approach gets right and where it stops
Journaling works because externalising a thought removes it from the active buffer. When a scenario is written down, the engine registers it as processed and temporarily releases the demand to keep it in active generation. Sleep hygiene works because a cooler room, no caffeine after 3 p.m., and reduced screen time lower the physiological activation level, which makes it harder for the engine to sustain a high generation rate.
Progressive muscle relaxation works because sustained muscular tension is one of the signals the engine reads as evidence that vigilance is warranted. Reducing that tension reduces the signal that something needs to be anticipated.
Each of these approaches is real. The reason they do not stop overthinking at night permanently is that they all operate at the level of managing the engine's conditions, not at the level of the engine itself. You can reduce the rate of scenario generation.
You can remove some of the primary emotionally charged material from the buffer. You can lower the physiological baseline. None of these change the fact that the pattern keeps running because the pattern was installed to run, and nothing has interrupted it at the level where it was installed.
Antano Solar John's team addresses this in the context of patterns like Kevin's social anxiety: the more you work on the surface output, the more familiar the loop gets. Familiarity does not mean it runs less. Kevin was in social situations for a decade and the loop ran harder, not quieter.
Meera has been journaling for two years and the loop resumes within the week after each method stops feeling novel. The engine is not responding to the techniques because the techniques are not reaching the engine.
The first reorientation in how to stop overthinking at night is recognising that the question itself contains a framing error. Stopping the output at night, which is what all standard advice targets, is a different problem from stopping the engine that generates the output. These require different approaches.
The symptom management is real and worth doing if it allows you to sleep. It is not the same as changing what runs the loop.
What changes when the engine stops generating at night
When the loop clears, the change is specific and observable. The person does not require any technique to fall asleep. The transition from wakefulness to sleep happens without a set of tools being applied first.
The absence of the loop is not experienced as suppression or as successfully redirecting attention. It is experienced as the absence of generated content. The mind is available for sleep because nothing is competing with sleep for its operation.
Kevin's description after his pattern was addressed captures this structure: the shift extended beyond the original triggering context. Social situations that used to activate the pattern simply did not activate it anymore. The shift was not contextual.
The pattern changed in Kevin, not in the specific situations. Meera, if her simulation engine is addressed at the right level, would not need to apply techniques situationally at night. The scenarios would not be generated. The loop would not need to be managed because it would no longer be running.
This is a different category of outcome from what symptom management produces. Symptom management produces better sleep on nights when the techniques work. Pattern resolution produces access to sleep without the techniques, because the generating function has been addressed, not just the output.
The night becomes quiet not because something is suppressing the noise but because the source of the noise is no longer active.
If you recognise Meera's pattern, the practical implication is to identify what level of intervention you are currently operating at. If journaling and sleep hygiene are providing some relief, they are worth continuing. They are not the full picture.
The engine that generates the scenarios at 2 a.m. runs below the level that those tools reach. Addressing it requires working at a level where unconscious patterns are accessible. When that happens, you stop needing to manage the night at all.
Watch Antano work with this pattern live
The video series shows the session dynamic in full, including exactly where the intervention lands and what changes in the person in the room.
Watch: Stop Overthinking for Good