A new study published in Nature Human Behaviour in January 2019 found that during light sleep, the brain selectively processes information during short periods of time.
This is a press release from the Centre National de la Recherche Scientifique (CNRS). Content made in collaboration with our partners at esanum.fr
But how do you make sure you don't miss your stop? Sleep seems to be accompanied by a loss of our ability to perceive and interact with our environment. However, previous experiences have shown that some sounds are perceived during sleep. Thus, a person's first name tends to wake a sleeper more easily than any other first name.
Until now, work has focused on the ability of the sleeping brain to process isolated sounds. However, this scenario is not representative of what happens on a daily basis. We frequently sleep, day and night, in rich acoustic environments where various sounds overlap and mix. On awakening and automatically, individuals tend to focus on the meaningful flow.
In this study, the researchers identified, during their sleep, the brain responses of several participants exposed simultaneously to two very similar voices in their acoustic properties but radically different in their meaning: one pronounced extracts from dialogues or articles while the other pronounced a stream of words similar to French but meaningless. The scientists then used a technique to reconstruct what sleepers hear from their brain activity.
Thus, they were able to confirm that during light sleep, participants prefer the message that makes sense to them. As a result, even when immersed in sleep and unconscious, the brain records the surrounding sounds, separates the different auditory sources and selects the one that is most understandable.
This ability to focus on what is relevant is temporary since it only concerns slow and light sleep. In addition, during this sleep phase, the brain seems to be able to process information from the outside world only during short time windows.
Sleepers track informative speech in a multi-talker environment. Guillaume Legendre, Thomas Andrillon, Matthieu Koroma and Sid Kouider, Nature Human Behaviour, 2019. https://doi.org/10.1038/s41562-018-0502-5