Understand how we got here — and how to move forward.


February 10, 2017

Why do we sleep? This theory may surprise you.

Daily Briefing

    Editor's note: This story was updated on Sept. 25, 2019.

    Scientists have found evidence to support the theory that we sleep to forget some of what we learn throughout the day, Carl Zimmer writes for the New York Times.

    Get our ready-to-use slides on the latest neurosciences market trends

    Giulio Tononi and Chiara Cirelli, biologists at the University of Wisconsin-Madison, in 2003 suggested that synapses—the connections between neurons—grow "so exuberantly during the day that our brain circuits [get] 'noisy,'" Zimmer writes. Memories are stored in these networks of connections and the biologists argued that during sleep the brain shortens the connections to help sift through the noise, a theory formally known as the "synaptic homeostasis hypothesis."

    Early evidence

    In a new study published in Science, Tononi and Cirelli tested their theory by examining the synapses in the brains of sleeping mice. The researchers found that synapses were 18 percent shorter when mice were asleep. "That there's such a big change over all is surprising," Tononi said.

    In a separate study also published in Science, Graham H. Diering, a postdoctoral researcher at Johns Hopkins University, and colleagues investigated the synaptic homeostasis hypothesis by studying proteins, particularly one called Homer1a, in mouse brains. In earlier research, Diering found that "hundreds of proteins increase or decrease inside of synapses during the night," Zimmer writes, but Homer1a "stood out" because of its important role in cutting back on synapses.

    To examine the relationship between sleep and Homer1a, the researchers genetically engineered some mice so that they were unable to make Homer1a proteins and then compared them with non-genetically engineered mice. The researchers found sleeping patterns between the two mice groups were similar, but the synapses in the genetically engineered mice did not change the proteins the same way as those in ordinary mice did.

    The research indicates that "sleepiness triggers neurons to make Homer1a and ship it into their synapses," Zimmer writes. "Homer1a turns on the pruning machinery."

    Examining the effects on memory

    Diering and his team then tested how synapse pruning affects memory by having the mice—half of which received a chemical shown to block synapse pruning—complete a memory test. The researchers found that the mice who received the chemical that blocked the pruning process had blurred memories.

    And in the other Science study, Tononi and colleagues observed that not every neuron was affected by the synapse pruning, with about one-fifth of synapses remaining unchanged. "It's possible that these synapses encode well-established memories that shouldn't be tampered with," Zimmer writes. As Tononi puts it, "You can forget in a smart way."

    But some researchers have said that the findings are not definitive proof of the synaptic homeostasis hypothesis. Markus Schmidt of the Ohio Sleep Medicine Institute said though the brain might prune synapses during sleep, he questioned whether the phenomenon was the primary explanation for why we sleep. "The work is great," Schmidt said of recent research, "but the question is, is this a function of sleep or is it the function?" (Zimmer, New York Times, 2/2).

    Get our ready-to-use slides on the latest neurosciences market trends

    Download the slides to learn everything you'll need to know about the neuroscience market in 2019, from growth outlook and financial considerations to new care management priorities and technology innovations.

    Get the Slides

    Have a Question?


    Ask our experts a question on any topic in health care by visiting our member portal, AskAdvisory.