A bottlenose dolphin can sleep and stay awake simultaneously. One hemisphere rests while the other runs the animal — monitoring for predators, timing breaths, keeping the body moving. The eye on the sleeping side closes. The other stays open.
This is called unihemispheric slow-wave sleep, and it solves a problem that obligate aquatic mammals face that terrestrial mammals do not: breathing is not automatic. A sleeping dolphin that loses cortical control will sink and drown. The solution, arrived at through evolution rather than engineering, is to keep half the system running.
The foundational work
Lev Mukhametov, a Soviet neurophysiologist, established the basic picture in 1977 with EEG recordings from bottlenose dolphins (Tursiops truncatus) in captivity at the Moscow Dolphinarium. He attached electrodes to both hemispheres and recorded brain activity over extended periods. What he found was striking: while one hemisphere showed the slow, synchronized waves characteristic of deep non-REM sleep, the other showed the fast, irregular pattern of wakefulness. The hemispheres took turns. Neither was ever fully asleep at the same time.
Mukhametov confirmed a behavioral correlate: the eye contralateral to the sleeping hemisphere — the eye controlled by the awake side — remained open and tracking. The eye on the sleeping side was closed. A dolphin resting at the surface with one eye open and one eye closed is not half-awake by analogy. It is neurologically precise about which half.
Duration and switching
Dolphins sleep roughly four to eight hours per day in total, accumulated across multiple bouts. A given hemisphere sleeps for approximately one to two hours before the pattern switches and the other hemisphere takes the rest period. During sleep bouts, dolphins typically rest at the surface, swimming slowly in tight circles or log-floating motionless while occasionally surfacing for a breath — this is called logging behavior.
Young calves present a striking exception. For the first weeks of life, dolphin calves remain continuously active with no detectable slow-wave sleep in either hemisphere. Their mothers match this schedule, going weeks without apparent sleep during the early postpartum period. How they manage this without the cognitive deficits that total sleep deprivation produces in other mammals is not well understood. One hypothesis is that the developing brain does not yet require sleep in the same way, or that the mother and calf are both running on a kind of extended wakefulness that has no equivalent in terrestrial mammal biology.
What each hemisphere does during the other's rest
The awake hemisphere handles continuous monitoring: echolocation processing, predator detection, social signals from nearby animals, and the timing of the respiratory reflex. Dolphins breathe voluntarily, not reflexively — each breath requires a cortical decision, even during sleep. Losing bilateral cortical function would interrupt this.
The sleeping hemisphere shows the restorative functions associated with non-REM sleep in other species: slow oscillations, synaptic downscaling, clearance of metabolic waste products. The architecture of what sleep accomplishes at the cellular level appears to be conserved even when only half the brain participates at a time.
Convergent evolution
Dolphins did not invent unihemispheric sleep. Several bird species show the same phenomenon, and in contexts that illuminate the underlying pressure. Mallard ducks sleeping at the edge of a group preferentially position the outward-facing eye to remain open, connected to the awake hemisphere, while the inward-facing eye closes. The birds at the group's periphery are more exposed and sleep more lightly — the proportion of time spent in unihemispheric versus bilateral sleep shifts with predation risk.
Frigatebirds show unihemispheric sleep during multi-day transoceanic flights. EEG recordings from birds tagged with miniature loggers confirmed they can sleep — one hemisphere at a time, and occasionally both — while flying. They sleep far less during flight than on land, suggesting that unihemispheric flight-sleep is a partial substitute, not a complete one.
The convergent appearance of this adaptation in mammals and birds, in aquatic and aerial contexts, suggests that the pressure toward it is strong wherever sustained vigilance conflicts with sleep need.
What remains uncertain
Three questions remain genuinely open. Whether unihemispheric sleep is as restorative as bilateral sleep — whether one hemisphere sleeping while the other works produces the same recovery as both sleeping together — has not been established. The behavioral evidence suggests dolphins are cognitively functional after their typical sleep pattern, but the comparison is hard to make directly.
How hemisphere switching is triggered — what signal determines that it is time for the hemispheres to trade states — is not known. It is presumably regulated at the level of individual-hemisphere sleep pressure, but the mechanism has not been traced.
And the quality of cognitive function in the awake hemisphere during the other hemisphere's sleep is difficult to measure. What does it mean to think with half a brain? The dolphin appears to navigate this question without apparent difficulty. We do not yet have the instruments to ask it more precisely.
---
Find more writing at anethoth.com. Browse indie SaaS projects at builds.anethoth.com.