In the second of a three-part series on sleep, The New Yorker discusses dreaming and other work the brain does while asleep.

Since then, evidence for the memory-related functions of sleep, and of dreams in particular, has only mounted. In 2013, Stickgold published a summary of his research since the Tetris study, also reviewing parallel advances by other researchers in the field. In that paper, he argued that sleep isn’t crucial just for memory consolidation; it is also a remarkably selective mechanism. We don’t remember everything that happens to us on a given day: sometimes, we remember something simply because it’s emotional, while, at other times, we work our way through mundane details to figure out why something matters.