Sleep spindles normalise negative affect: 3 experiments by Bar et al.

Sleep spindles – brief bursts of EEG activity during deep sleep – do more than just “consolidate memory.” Recent research by Bar et al. (bioRxiv, April 2026) has shown that they actively recalibrate the emotional valence of what you remember. Without them, the world systematically seems worse than it is. This isn't just a metaphor for “things will look better in the morning.” It's concrete neurophysiology.

What have [they/you] done

Three experiments combining behaviour, fMRI and EEG during sleep. 29 participants (18 female). Protocol:

1. Training. The participant sees three neutral faces. Each is paired with an outcome – positive (monetary reward), negative (shock), or neutral. The brain creates associations.
2. Generalisation testing. The participant sees morphing face — a mixture of the originals in various proportions (for example, 60% % positive + 40% 40% negative). The task is to assess which of the originals it is “closer” to emotionally.
3. Delay. Half of the participants slept overnight in the laboratory with EEG monitoring. The other half stayed awake for a long time during the day.
4. Retesting after a delay.

What did you find

Immediately after training Generalisation leans towards the negative. An unknown face that slightly resembles a “negative” one is assessed as negative.

After a night's sleep Generalisation shifts to positive. The same unknown face is now perceived as closer to positive. The negative bias decreases.

After a day of exertion — On the contrary. Generalisation becomes even more negative. Fatigue heightens the inclination towards threat.

fMRI: Amygdala and limbic system activity during learning predicted the strength of the negative deflection immediately afterwards – and the strength of the positive deflection after sleep. In other words, the amygdala “remembers” strongly, and then sleep “repackages” the memory.

EEG activity of sleep spindles during slow-wave sleep positively correlated with:

• Better generalisation of positive faces
• A larger shift from negative to positive

So sleep spindles are a neurophysiological marker of the work the brain does at night.

Sleep spindles are brief bursts of oscillatory brain activity that are characteristic of stage N2 sleep. They are thought to play a role in memory consolidation, learning, and protecting sleep from external stimuli.

Sleep spindles are short bursts of EEG activity at 11–15 Hz, lasting 0.5–3 seconds. They primarily appear during stage N2 of slow-wave sleep. They look like spindles on an EEG, hence the name.

Classic role: Memory consolidation. Sleep spindles coordinate the hippocampus (new memory) and the cortex (long-term storage). During spindles, information is literally rewritten from the hippocampus to the cortex, where it is stored long-term.

This research adds a new layer to the old story. Spindles don't just “store memory”. They recalibrate the emotional tone What is stored. What seemed threatening in the evening, in the morning receives a more balanced weight.

It is not “sleep erases the negative.” It is sleep. normalises — returns to a more realistic average, without skew.

Why does it matter

For anxiety and PTSD. Both conditions are characterised by two things: overgeneralisation of negativity (everything reminds you of the trauma) and sleep disturbance. This research shows that these are not two separate symptoms but one leading to the other. Poor sleep → sleep spindles are weak → the brain can't repackage the negativity → the condition becomes fixed.

Hence the hypothesis: targeted sleep improvement (particularly N2 stage) is a therapeutic target for anxiety disorders. Some drugs (Z-drugs, benzodiazepines) reduce sleep spindles. This may explain why they provide short-term symptom relief but do not cure long-term.

For a healthy person under stress. After a difficult day of conflicts, everything seems worse than it is. This isn't “character” – it's the current state of the brain: amygdala pumped, cortex tired, negativity unaddressed. A night's sleep literally changes the perception of the same day.

This is where the old rule “things are clearer in the morning” isn't a metaphor. It's neurophysiology.

For athletes and recovery. A stressful training block without adequate sleep = the brain's inability to repackage discomfort. Motivation drops, training feels harder. It's not “I'm physically overtrained” — it's Cognitive-emotional overload, which is resolved by sleep, not massage. The same principle also governs Recommended cold recovery strategy — The tool is one, the application context is different.

What destroys sleep spindles

• Alcohol before sleep. A classic finding – alcohol leads to rapid sleep onset, but sharply reduces N2 and REM. Spindles practically disappear. Sleep is present - no reprocessing.
• Fragmentary dream. Waking up every hour destroys the next stage. The spindles don't have enough time to form.
• Stimulants in the evening. Caffeine has a half-life of approximately 5 hours — a cup of coffee drunk at 4 pm is still active at 9 pm.
• Chronic sleep deprivation. Less than 7 hours = less full cycle = less N2 = fewer sleep spindles.
• Some sleeping pills. Z-drugs reduce deep sleep and sleep spindles, despite subjective feelings of being well-rested.
• Heatwave. High temperatures in the bedroom (>22°C) fragment sleep.

Sleep spindles are supported by the thalamocortical system, specifically by interactions between the thalamus and the cortex. They are thought to be generated by thalamic reticular nucleus (TRN) neurons, which then influence activity in the rest of the thalamus and the cortex.

• 7–9 hours of sleep. I will sleep on the weekends. Spindles happen in certain stages – a full cycle is needed.
• A cool bedroom (16–19°C). A drop in body core temperature is needed for deep sleep.
• Darkness. Any light (including a phone screen's background glow) reduces melatonin and fragments sleep.
• Regular schedule. Go to bed and wake up +/- 30 minutes from your usual time. The circadian rhythm requires predictability.
• Morning physical activity. It makes N2 and slow-wave sleep deeper. Just not in the 2 hours before sleep.

Restrictions

29 participants is not a small sample size for an fMRI/EEG study, but it's not a large one either. An experimental model with faces is a simplification of more complex real-life emotional reactions. Morph-faces are not equivalent to real trauma.

The correlation between sleep spindles and shifts is not evidence of causation in a human experiment. Spindles could be a marker for something else that's actually doing the work.

But the direction is consistent with the animal literature (where it's possible to stimulate spindles directly and see the effect) and the clinical picture of PTSD. This isn't a single finding out of nowhere; it's part of a mosaic that's already forming.

Conclusion

Sleep is not “the absence of wakefulness”. It is active neurophysiological work, part of which is Recalibration of the emotional weight of memory. Without it, the world systematically seems worse than it is.

This isn't advice that “a good night's sleep will make you more positive”. This is a fact: your perception of reality depends on whether your brain has done its nightly work. Sleep spindles are not abstract neurobiology. They are the reason why “things look clearer in the morning” – it's literally true.

---

Source: Bar E, Bringmann M, Belonosov G, et al. Sleep Renormalizes Negative Emotional Generalization. bioRxiv 2025/2026. DOI: 10.1101/2025.08.18.670780

Additionally: Walker MP, The role of sleep in cognition and emotion, Annual Review of Psychology 2017; Halonen R et al., Sleep-related processing of emotional memories, Cognitive Affective & Behavioural Neuroscience 2023.