There is a specific thing that happens in the meeting the day after a short night, and it is not what most people would guess. It is not that you feel more. Tiredness usually flattens feeling rather than sharpening it. It is something quieter and stranger than that. The colleague who normally reads as neutral seems, today, faintly cool. The feedback that would ordinarily just be feedback has an edge on it. The room feels a half-degree more hostile than the room actually is.

The environment did not change. The instrument you use to read it did.

An earlier post in this series looked at what a short night does to the pause between feeling and reacting, the internal circuit that lets you notice irritation without acting on it. This post is its outward companion: not what bad sleep does to your regulation, but what it does to your perception of other people. The inward cost has an outward twin, and the outward one is easier to miss, because it does not feel like something happening in you. It feels like the world.

The calibration of threat

The clearest human evidence here is older than it sounds, and it is precise. In a 2015 study in the Journal of Neuroscience, researchers kept people awake and then asked them to read faces, sorting threatening expressions from affiliative ones.[1] Sleep loss did not simply make them worse at the task in a general way. It shifted the calibration in a particular direction: sleep-deprived people lost the ability to reliably discriminate a threatening face from a friendly one, and the error ran toward threat. Neutral and ambiguous signals got pulled toward hostile.

This showed up in the brain and in the body at once. The regions that integrate a social signal with a gut reading of it, the anterior insula and cingulate among them, processed the cues differently. And the coupling between the heart and the amygdala, the loop that normally lets your physiology track the emotional temperature of a face, degraded.[1] The felt translation is unsettlingly literal. After a bad night, you are not imagining that people seem harder. Your threat-detection system has quietly widened its net, and more of the neutral world is falling inside it.

A comprehensive 2020 review in Trends in Cognitive Sciences set this alongside everything else that shifts.[2] Sleep loss amplifies the amygdala's response to socially negative signals while weakening the prefrontal connection that would normally temper it. It makes people less accurate at reading moderate-intensity expressions, the ordinary in-between faces that make up most of a day. And it does not stop at perception. The same review tracks the effect outward into behaviour: reduced prosocial action, more social withdrawal, more conflict in relationships and at work. A short night, read at scale, is quietly anti-social.

A measurement error, not a moral one

The popular version of sleep and relationships tends to say: tired people are less patient, less generous, a bit more selfish. That framing is not wrong, but it lands badly, because it is about character. It asks the conscientious woman to notice she was less kind, and she will file that as one more thing to feel bad about.

The perception finding is different. The problem is not that you cared less. It is that the raw data arriving at your senses was already distorted before any question of kindness came up. You were responding, reasonably, to a room that your own nervous system had mis-scored. The unkindness, where it appears at all, is downstream of a measurement error, not a moral one.

This is also not confined to the dramatic case. The 2015 work used substantial sleep loss, but a 2020 study in Nature Human Behaviour shows the milder end of the same gradient.[3] Insufficient sleep, not just sleeplessness, amplifies next-day anxiety through impaired activity in the medial prefrontal cortex, in a dose-response way that holds even at population scale: the nights you shorten predict the anxiety of the days that follow. Non-REM slow-wave sleep, specifically, does anxiety-calming work through that same circuit, and losing some of it is losing the calm. You do not have to pull an all-nighter to walk into a room miscalibrated. You only have to sleep less than your brain was counting on.

A mechanism worth naming carefully

There is a newer thread here, and I want to handle it honestly, because it is the kind of finding that is easy to oversell.

A 2026 study in Brain Research looked at what changes in the amygdala itself after sleep deprivation, at a cellular level.[4] It found that the brain's own immune cells, the microglia, became activated in the basolateral amygdala, a region central to social and emotional processing, and that this activation tracked with anxiety and impaired social interaction. When the researchers suppressed that microglial activity, the socio-emotional behaviour improved. It is a clean and suggestive result, and it points at inflammation, not just faulty wiring, as part of the story.

It is also a mouse study. There is no human replication of this specific mechanism, and the honest position is that we do not know whether the microglial pathway works the same way in people. The human evidence for the social effect, the threat mis-reading and the prosocial erosion, is solid, but it runs through a different, better-established route: connectivity, autonomic coupling, prefrontal regulation. I mention the microglia because it is genuinely new and worth knowing exists, not because it has earned the right to explain your Tuesday. The load-bearing science is the human work. The mouse work is a lead, and I would rather flag the gap than let it quietly close.

What to do with this, which is not much

I am wary, as always, of ending on a fix. The woman MoMo is built for does not need to be told to sleep more. She knows. The instruction was never the missing piece, and on many of the nights in question the short sleep was not a choice she got to make.

What is worth carrying is smaller and, I think, kinder. The next time a day feels socially expensive, when a conversation costs more than it should have, when a room reads as colder than it turns out to be, it might be worth asking what the night before was sitting underneath. Not to excuse anything. Just to hold open the possibility that the hostility you felt in the room was partly a reading error, and that the person who mis-read it was running an instrument that had not been allowed to recalibrate overnight.

That recognition, the way a hard afternoon connects back to a short night so the pattern stops feeling like evidence about who you are and starts looking like a cycle you can see, is the kind of thread MoMo is built to hold between days. But that is a footnote to the essay, not its argument. The argument is the calibration. The calibration is real, and it is not a verdict on your character. It is a measurement running on less than it needed.

If this feels like more than a pattern (if you are in distress or crisis) please reach out: Mind Sweden 90101 | Samaritans UK 116 123 | your local crisis line.

References

Goldstein-Piekarski, A. N., Greer, S. M., Saletin, J. M., & Walker, M. P. (2015). Sleep Deprivation Impairs the Human Central and Peripheral Nervous System Discrimination of Social Threat. Journal of Neuroscience, 35(28), 10135–10145. PMID: 26180190.
Ben Simon, E., Vallat, R., Barnes, C. M., & Walker, M. P. (2020). Sleep Loss and the Socio-Emotional Brain. Trends in Cognitive Sciences, 24(6), 435–450. PMID: 32299657.
Ben Simon, E., Rossi, A., Harvey, A. G., & Walker, M. P. (2020). Overanxious and underslept. Nature Human Behaviour, 4(1), 100–110. PMID: 31685950.
Xin, T., Kong, D., Shi, M., Wei, C., Chen, G., Wang, S., Zhao, S., Ma, Y., Jin, X., & Liu, Z. (2026). Microglial inhibition in the basolateral amygdala is associated with improved socio-emotional behaviors following sleep deprivation in mice. Brain Research, 1885, 150332. DOI: 10.1016/j.brainres.2026.150332.