Learning new information while we sleep

By Karyn O'Keeffe 31/08/2012


This week brings some exciting findings about sleep and learning.  In particular, in a study to be published in the journal Nature Neuroscience, Arzi and colleagues have shown that it’s possible to learn new information during sleep.

To date, research has largely focussed on information storage during sleep rather than new information processing during sleep.  For example, we know that the learning of varying types of information is enhanced with sleep, and that different types of information are better learnt during different sleep stages.

However, up until now we haven’t seen many studies that have successfully demonstrated that we might be able to learn new information while we are asleep.  Using very simple forms of associative learning, researchers have previously demonstrated that animals, such as rats, and human infants are able to process some types of information during sleep.  In these studies, the brain centres activated during sleep were the same as those we use when we unconsciously process information (eg, motor skills when learning to ride a bike). 

To date, no studies have successfully shown that we can actively process during sleep information we have learnt consciously (eg, remembering the trip you took at Christmas 2003, knowing that your bright red sweater needs to be washed separately).  However, in a group of young adults, Arzi has shown that we might be able to process this type of information during sleep.

Different tones were played to sleeping adults and each time a particular smell was wafted past their nose.  Particular tones were linked to particular smells – pleasant (deodorant, shampoo) or unpleasant (rotten fish or flesh) – and the sniff response to each measured.  (A strong sniff response is linked to a pleasant smell and a weak response to an unpleasant one.)  The tones were repeated many times during the night and the  researchers were then able to test whether an association had been learnt between the tones and a particular smell.

Arzi and colleagues showed two fairly exciting things.  We seem to be able to learn during our sleep and demonstrate that learning during the same sleep period.  When tones only (no odour) were played in the second half of the night, the matching sniff response was demonstrated.  Information learnt during sleep continues into wakefulness.  When the tones only were played during the following day, the matching sniff response was demonstrated.

This is very much a first step in this area and it will be interesting to see where this study leads.  It is definitely unclear whether more complex learning is possible during sleep, and then if it is, what function or role this type of learning could serve.