For more than a century, scientists have known that, in humans, sleep plays an important role in memory consolidation – taking memories and storing them so they can be retrieved for future use. As far back as the beginning of the 20th century, studies have shown that if you sleep after memorizing a list, you will find it easier to recall the list than if you had remained awake. Since then, hundreds of additional studies have shown that sleep improves memory consolidation.
In 2014, Albrecht P. Vorster and Jan Born, at the University of Tübingen in Germany, published a review of studies on sleep and memory in animals. They found that sleep is found throughout the animal kingdom and that it is associated with memory formation in vertebrates and invertebrates.
Sleep and Memory Consolidation in Mammals
Vorster and Born reported that studies on humans and rodents indicate that non-REM deep sleep, or slow wave sleep (SWS), seems to play a more important role in memory consolidation than REM sleep. This is particularly true of declarative and prospective memories (memories of statements and plans). REM sleep may help with the consolidation of procedural memories (memories of how to perform tasks) and memories of emotions. This relationship between the different sleep stages and different types of memory was previously hypothesized by Jie Zhang, in his Continual Activation Theory of Dreams, in 2004.
Research suggests that during SWS, neurons replay memory representations. When you perform a task, certain neurons in your brain fire in a particular order. During SWS, these neurons fire again, in the same order. These replays have been observed during SWS and quiet wakefulness, but are normally not observed during REM sleep.
Active System Consolidation
To explain how mammals form memories during sleep, scientists have come up with the concept of “active system consolidation”. According to this model, memories are stored in two stages. During the first stage, memories are rapidly encoded in an initial storage system. In the second stage, which proceeds at a slower pace, memories are encoded in long-term memory.
The first stage occurs during wakefulness. New information is encoded in the hippocampal and neocortical networks, with the hippocampus encoding episodic memories. Episodic memories are memories of experiences that have actually happened to you. For example, you can remember going to a restaurant, being seated and giving the waiter your order. You can remember what your dinner companion was wearing, the waiter’s tone of voice, what you ate and how it tasted.
Episodic memories differ from semantic memories – memories of facts and concepts that are shared by others. Knowing how to order in a restaurant and how much of a tip you would normally be expected to leave are examples of semantic memory.
When memories are first encoded within the hippocampus, they are contextualized. That is, they are associated with a particular time and place.
During SWS, the newly encoded representations of memories are reactivated repeatedly. During this process, some of these representations are stored outside the hippocampus, where they are decontextualized. In other words, you don’t simply remember that when you ate sushi for the first time at this restaurant, you liked it. You remember that sushi tastes good. If you were taught how to use chopsticks for the first time, you don’t simply remember, “This is how I used chopsticks when eating sushi at this restaurant.” You remember, “This is how to use chopsticks.”
Sleep and memory in birds
Birds and mammals are the only animals to experience distinct SWS and REM sleep stages. However, sleep in birds differs greatly from sleep in mammals. Sleep is very fragmented in birds, occurring in short bursts lasting between one and four minutes. Birds often sleep unihemispherically, with one eye closed and one eye open. That is, one brain hemisphere goes to sleep, while the other remains awake. The only mammals that do this are cetaceans (whales and dolphins).
In birds, both hemispheres will go to sleep at the same time when conditions are safe. This suggests that SWS is more effective when it involves the whole brain. Studies on chickens show that when they are deprived of sleep, they will sleep with both eyes closed the next day.
Because reptiles and amphibians don’t have distinct SWS and REM sleep stages, it can be assumed that these stages evolved independently in birds and mammals. The large and strongly interconnected brains in mammals and birds could have resulted in a need for these sleep stages.
Sleep and Imprinting in Birds
In birds, SWS appears to play a role in filial imprinting, which takes place in the first three or four days of a chick’s life. During this time, a brief exposure to an object causes the chick to easily recognize and strongly bond with this object. This allows the chick to selectively follow the object. Normally, the chick becomes attached to a member of its own species, usually the mother. However, in scientific experiments, chicks have imprinted on humans and objects.
In 2008, a study was performed in which chicks were trained to imprint on a rotating red box. The chicks were divided into two groups. One group was allowed to sleep for six hours after imprinting training. For the following six hours, they were deprived of sleep. These chicks imprinted successfully.
The other group had their sleep disturbed for the first six hours after imprinting, and was then allowed to sleep for six hours. These chicks never became attached to the red box.
Although the brains of birds are very different, structurally, from the brains of mammals, EEG reports shows that in birds, as in mammals, neuronal representations of memories are reactivated during SWS.
Sleep and Song Learning in Birds
Sleep also seems to be necessary for song learning. Songbirds develop their songs in two stages that take place between 30 and 90 days after hatching. During the first stage, the bird listens to a song sung by adult tutor. During the second stage, the bird attempts to imitate the song and gets auditory feedback on its performance.
After a bird is first exposed to the singing of an adult tutor, it falls asleep quickly. This suggests that learning a song requires more sleep.
In adult birds, sleep is important for song discrimination. In a series of studies performed in 2010 and 2013, starlings were trained to distinguish different five-second song segments. When the starlings were allowed to sleep before a retest, their performance significantly improved.
Sleep and Memory in Invertebrates
Sleep research on invertebrates began in the 1980s, when researchers began to study sleep in bees, cockroaches and scorpions. Since then, they have also studied sleep in flies, honeybees, crayfish and the roundworm C. elegans.
Evidence suggests that most invertebrates sleep. Since invertebrates don’t exhibit EEG signals typical of mammalian sleep, researchers identify sleep in invertebrates by behavioral signs, such as inactivity, the presence of a specific body posture and difficulty being aroused.
Studies of honeybees and flies suggest that they might have different sleep stages, which serve different functions.
Sleep and Memory in Bees
Bees have highly developed brains in comparison to many other insects. They are able to perform complex tasks involving navigation and communication. This includes identifying the location of pollen and nectar in relation to cues in the landscape and the position of the sun in the sky, so that the food source can be found again. They must then communicate this information to their hive- mates. To perform these tasks successfully, they must develop episodic memories of events they have experienced when seeking food.
In a 2009 study, bees were given sucrose when they extended their proboscises when presented with a neutral odor. They learned to extend their proboscis in response to the odor. Sleep deprivation did not affect the bees’ learning this response.
However, sleep deprivation did have an effect on extinction learning. This is learning that occurs after reinforcement – in this case, the sucrose – is removed. The researchers stopped giving a group of bees sucrose and then allowed them to sleep. The following day, the bees stopped extending their proboscises in response to the odor. However, another group of bees, which was not allowed to sleep after extinction training, continued to extend their proboscises the next day.
Sleep Helps Honeybees Form Spatial Memories
Research indicates that bees need sleep to help them form spatial memories. In a study performed in 2012, honeybees were removed from their hive and moved to an unknown site. They then had to find their way back to the hive. 58% of the bees were able to find their way home.
The following night, some of the bees were allowed to sleep, while others were deprived of sleep. The bees where then taken to the release site once more. This time, 83% of the bees that had been able to sleep made their way back to the hive. However, less than 50% of the sleep deprived bees were able to find their way back.
The bees that were allowed to sleep after the first exposure to the release site slept longer than unusual, another indication that sleep is important for performing a highly demanding spatial memory tasks.
Sleep and Memory in Fruit Flies
When fruit flies (Drosophila) are exposed to enriched environments and when they learn specific behaviors, they sleep more. This shows that sleep is important for learning.
Mutant flies that sleep much less than wild flies have reduced abilities to form short term memories.
A 2012 study showed that sleep deprivation affects long term memory consolidation. Fruit flies were exposed to an odor and given electric shocks. They were also exposed to a second odor, but they did not receive shocks when exposed to this odor. Afterwards, some flies were allowed to sleep but others were kept awake.
The flies were then placed in a T-maze, where they had to choose between the two arms, one of which contained the odor that had been associated with shocks, and one of which contained the other odor. The flies were supposed to avoid the arm containing the odor associated with shocks. Sleep-deprived flies had much greater difficulty performing the task correctly.
In another experiment, sexually naive male fruit flies were placed with male flies that produced female pheromones. The naive male flies chased the flies with the female pheromones, but when they caught up to them, they were unable to mate. When the naive flies were allowed to sleep, they stopped chasing the female-scented males. However, sleep deprived flies continued to chase them.
References:
Beyaert, L. Greggers, U. & Menzel R. (2012). Honeybees consolidate navigation memory during sleep. Journal of Experimental Biology, 218, 15.
Brown, T.P., Nusbaum, H.C. & Marboliash, D. (2010). Sleep-dependent consolidation of auditory discrimination learning in adult starlings, Journal of Neuroscience, (30)2, 609-613.
Brown, T.P., Nusbaum, H.C. & Marboliash, D. (2013). Sleep consolidation of interfering auditory memories in starlings. Psychological Science, 24, 439.
Ganguly-Fitzgerald, I., Donlea, J. & Shaw, P.J. (2006). Waking experience affects sleep need in Drosophila. Science, 313(5794), 1775-1781.
Hussaini, S.A., Bogusch, L., Landgraft, T. & Menzel, R. (2009). Sleep deprivation affects extinction but not acquisition memory in honeybees. Learning & Memory, 16, 698-705.
Jackson, C., McCabe, B., Nicol, A.U., Grout, A.S., Brown, M.W. & Horn, G. (2008). Dynamics of a memory trace: effects of sleep on consolidation. Current Biology (18)6, 393-400.
Le Glou, E., Seugnet, L., Shaw, P.J., Preat, T. & Goguel V. (2012). Circadian modulation of connsolidated memory retrievel following sleep depriviation in Drospophila. Sleep. (35)10, 1377-1348.
Vorster, P. & Born J. (2014). Sleep and memory in mammals, birds and invertebrates. Neuroscience & Behavioral Reviews, 50, 103-119.