Sleep loss mostly drains positive mood, not inflates negative mood
In laboratory studies where healthy adults stay awake for about 38 hours, the most consistent emotional change is a fall in positive affect — feeling less cheerful, interested, or energized. In one study of 96 adults, positive affect declined steadily across the waking period, while negative affect rose only slightly and not significantly [1]. That asymmetry matters: sleep deprivation looks less like becoming irritable and more like losing the bright side of mood.
Individual differences sit on top of that pattern. People who score higher on a trait called emotion reactivity — roughly, how strongly they react to emotional events — showed somewhat more negative affect during sleep deprivation, but this effect was small and negative affect stayed near the floor of the scale [1]. The same trait did not predict the drop in positive mood at all [1]. So knowing someone's emotional sensitivity tells you a little about how cranky they may get, but not much about how flat they will feel.
Why the clock changes what you see
Emotion is not steady across a day. Alertness, arousal, and the body's internal circadian rhythm rise and fall, and these shifts can change both how someone feels and how well they read emotions in others. A study of older adults compared emotion recognition in the morning versus the afternoon and found that time of day mattered for decoding some emotions from dynamic visual cues [3]. In other words, the measuring stick itself moves with the clock.
That means a sleep-deprivation test at 9 a.m. and the same test at 9 p.m. are not measuring the same starting point. If positive mood is already lower in the early morning, sleep loss may add less on top; if alertness is higher in the evening, the same sleep loss may show up differently. The 96-person study handled this by sampling mood every two to four hours across the day and night, which lets the time-of-day pattern be separated from the time-awake pattern [1]. Positive mood fell as a function of test bout — a combined measure of hours awake and clock time — showing that the decline is not a single event but a trajectory across the day [1].
What the evidence does not settle
The direct test of whether time of day changes the size of sleep-loss effects on emotion is still thin. The mood study tracked time across the day but did not report a separate comparison of, say, morning versus evening effects of sleep loss on emotion [1]. The emotion-recognition study in older adults points to time-of-day effects on decoding emotions, but it is about reading faces, not about how sleep deprivation changes the reader's own mood [3]. And a neural study of 12 women found that 24 hours without sleep slowed responses and changed brain activity during an emotional go/no-go task, but it did not vary the testing time, so it cannot say whether the clock would alter those effects [2].
Nearby evidence from decision-making suggests the clock does reshape how sleep loss plays out, even if it is not about emotion. In a 39-hour constant-routine study, risk-taking peaked around midday before sleep deprivation and dropped after about 29.5 hours awake, and reaction times were slowest in the morning after sleep loss [4]. That is a different behavior, but it shows the general principle: the same sleep loss can produce different performance depending on when it is measured. For emotion specifically, the honest answer is that time of day clearly matters for emotional processing, and sleep loss clearly matters for mood, but how much the two interact — and in which direction — is not yet pinned down.
What this means if you are judging your own mood
If you are sleep-deprived and feel flat rather than furious, that fits the pattern: the strongest and most reliable change is a loss of positive mood, with negative mood rising only modestly and mostly in people who are already emotionally reactive [1]. If you notice that your mood is worse at some hours than others, that is also expected, because emotional processing shifts with time of day [3]. The practical caution is not to compare a bad morning after poor sleep with a good evening after good sleep and conclude that sleep is the whole story — the clock is doing part of the work.
What would sharpen the answer is a study that deliberately tests the same people at different clock times under both rested and sleep-deprived conditions, measuring mood and emotion processing directly. The existing studies come close — one samples mood around the clock [1], one compares morning and afternoon emotion recognition [3], and one maps risk-taking across a constant routine [4] — but none combines all of those elements for emotion. Until then, the best-supported statement is that time of day shapes emotional processing and sleep loss lowers positive mood, and the two likely interact, but the size and direction of that interaction remain open.
Sources used in this answer
0143 Trait Emotion Reactivity Predicts Change in Negative but Not Positive Affect During Total Sleep Deprivation
During 38 hours of total sleep deprivation, positive affect declined significantly across test bouts while negative affect rose only slightly and non-significantly; higher trait emotion reactivity predicted more negative affect but did not predict the drop in positive affect.
Neural effects of sleep deprivation on inhibitory control and emotion processing
After 24 hours without sleep, 12 women showed slower response times, more errors, higher resting theta power, and altered N2 and P3 brain responses during emotional and shape go/no-go tasks, suggesting sleep loss impairs inhibitory control and emotional task performance.
The combined effect of sleep and time of day on emotion decoding from dynamic visual cues in older adults
In older adults, time of day (morning versus afternoon) affected decoding of some basic emotions from dynamic visual cues, indicating that emotional processing varies with testing time.
Time of Day and Sleep Deprivation Effects on Risky Decision Making
During a 39-hour constant routine, risk-taking and reaction time varied with circadian timing, with greater risk propensity around midday before sleep deprivation and slower morning reaction times after sleep loss.
A time for learning and a time for sleep: the effect of sleep deprivation on contextual fear conditioning at different times of the day.
In rats, post-training sleep deprivation impaired contextual fear memory consolidation at both training times, but the impairment required less sleep loss during the resting phase than during the active phase, consistent with the amount of sleep lost mattering more than time of day itself.