Scientists from the United States have learned that dolphins are able to stay alert and active for 15 days or more at a time by sleeping only one-half of their brain. The experts believe that this ability to stay alert by using only half of their brain is a key to the survival of sea mammals. The scientists believe that the ability helps the dolphin to surface to breathe and remain vigilant for predators such as sharks.

The scientists conducted the research in California by testing the ability of two bottlenose dolphins to echolocate accurately over a certain period. The scientists say that the dolphins could echolocate accurately over a period long enough to have left other animals sleep-deprived. During the tests, the two dolphins swam around their enclosure looking for Phantom sonar targets.

Each of the eight test devices used with the dolphins was made up of a device that could pick up the dolphins sound pulses and return a phantom target. The dolphins then responded to the phantom targets by pressing a paddle. The dolphins were rewarded with a special tone with success and a fish. A false alarm gave the dolphins no tone and no reward.

The researchers report that over three sessions of five continuous days the Dolphins had a success rate of up to 99%. The female dolphin outperformed her male partner during the test. The scientists tested the female dolphin with the same experiment over a period of 30 days.

http://www.slashgear.com/dolphins-can-sleep-one-half-of-their-brain-at-a-time-say-researchers-18252614/

Hi,

I heard dolphins do. I guess it can be assessed with EEG...

Hope that helped

Scientists from the United States have learned that dolphins are able to stay alert and active for 15 days or more at a time by sleeping only one-half of their brain. The experts believe that this ability to stay alert by using only half of their brain is a key to the survival of sea mammals. The scientists believe that the ability helps the dolphin to surface to breathe and remain vigilant for predators such as sharks.

The scientists conducted the research in California by testing the ability of two bottlenose dolphins to echolocate accurately over a certain period. The scientists say that the dolphins could echolocate accurately over a period long enough to have left other animals sleep-deprived. During the tests, the two dolphins swam around their enclosure looking for Phantom sonar targets.

Each of the eight test devices used with the dolphins was made up of a device that could pick up the dolphins sound pulses and return a phantom target. The dolphins then responded to the phantom targets by pressing a paddle. The dolphins were rewarded with a special tone with success and a fish. A false alarm gave the dolphins no tone and no reward.

The researchers report that over three sessions of five continuous days the Dolphins had a success rate of up to 99%. The female dolphin outperformed her male partner during the test. The scientists tested the female dolphin with the same experiment over a period of 30 days.

http://www.slashgear.com/dolphins-can-sleep-one-half-of-their-brain-at-a-time-say-researchers-18252614/

When talking about the ability of an animal to sleep "with one hemisphere only", people usually mean what is also called unihemispheric slow wave sleep. The advance of unihemispheric sleep is that an animal can rest parts of its brain, while still being able to pay attention to its environment, e.g., avoid predation, and to keep moving in a coordinated manner, which is essential for marine animals that need to go up to the water surface to breath and for migrating birds that need to keep flying for very long stretches of time. So, for a few examples, Beluga whales seem to do it, bottlenose dolphins and other dolphins, swifts, sparrows, and blackbirds, too.

The go-to method for studying sleep, especially in (terrestrial) mammals and birds, is EEG, i.e., electroencephalogram, and sometimes ECoG, i.e., Electrocorticogram. To conduct an EEG, a number of electrodes are distributed across the head, and these electrodes can pick up gross electrical activity in the underlying brain tissue, especially in regions near the surface, e.g., the cortex for mammals. For ECoG, these electrodes are placed into direct contact with the membranes of the brain, below the skull bone, which increases signal-to-noise ratio, but renders the procedure invasive, while EEG is entirely non-invasive.

As electrodes can be spaced evenly across the skull, it is possible to record from both brain hemispheres at the same time. The EEG signal in the brain of a sleeping person or animal is distinctly different from the signal of an awake person or animal (further subdivisions exist), so unihemispheric sleep can unequivocally be detected with EEG.

For many animals, e.g., marine mammals, conducting an EEG is technically very difficult or impossible, and most sleep studies on these animals are done behaviourally, e.g., by presenting tones or visual targets and measuring the response time. Here, researchers assume that response goes up with a rising arousal threshold which accompanies the sleep state, and a rise in response time is thus used as a marker for sleep. For one example of such an approach in dolphins see Ridgway et al 2006 "Dolphin continuous auditory vigilance for five days"  (https://www.ncbi.nlm.nih.gov/pubmed/16943502). In behavioural unihemispheric sleep studies (I just know of such studies in chicken), the stimulus is presented to one eye (or ear) at a time only and the response time then taken and compared between both eyes (or ears). Assuming that one half of the brain controls the other side of the body, e.g., the left hemisphere 'controls' the right eye, differences in response times are taken as an indicator that one half of the brain has been asleep.

One needs to keep in mind though that, unless an EEG can pick up a distinctly "sleepy" pattern in one hemisphere and a distinctly "awake" pattern in the other, the evidence for the existence of unihemispheric sleep in any animal is usually indirect aka 'if an animal can go without a significant drop in response time for a number of days, assuming that it needs to sleep every day or so, it must be sleeping with one part of its brain only'.

I know of two main reviews on the topic. More focused on marine mammals: Rattenbourg, Amlaner, Lima, 2000, "Behavioral, neurophysiological and evolutionary perspectives on unihemispheric sleep". More focused on birds: Rattenborg, 2006, "Do birds sleep in flight?".

Hope this helps.