I saw the word “lonely” in a UC Berkeley release on ScienceDaily, headlined “Scientists discover mysterious ‘lonely’ brain waves linked to memory loss,” and assumed I was about to read another study on isolation and dementia. It isn’t one. Nobody was asked about friends or eating dinner alone, because the lonely things are brain waves. In older adults with more tau, the tangle protein of Alzheimer’s, in the frontal cortex, more of the deep-sleep waves fired on their own instead of sweeping across the brain. The next morning those people recalled fewer of the word pairs they had studied the evening before.

FROM THE DOCUMENT
Tau pathology leads to lonely non-traveling slow waves that mediate human memory impairment
The preprint's original title. The published paper swapped "leads to" and "mediate" for "associated with" and "linked to." Source: 2024 preprint on bioRxiv
What we cannot yet say is which leads.
Omer Sharon, UC Berkeley

The participants didn’t have dementia. “These people did not have Alzheimer’s,” lead author Omer Sharon of Berkeley’s Center for Human Sleep Science said in the release, and their memory decline was within the normal range for their age. The Nature Neuroscience paper came out September 11, with Matthew Walker as senior author. It describes ordinary aging as it starts to slip, and it points to a specific place where the slipping may begin.

Waves that are supposed to move together

Slow waves are the large, rolling rhythms your cortex produces in deep non-REM sleep. In the young adults, who were in their early 20s, the waves moved in groups. They started at the front of the brain and spread across about the width of a hand on the scalp. Alzforum’s coverage of the preprint says they traveled “en masse” from frontal to parietal cortex, much like a wave rolling around a stadium.

The older group, aged from the mid-60s to the mid-70s, looked different. Their waves were smaller, came less often and covered shorter distances, and more of them started and spread nowhere.

Why should it matter whether a wave travels if it fires either way? Sharon told Alzforum that slow waves coincide with an influx of calcium into neurons that may create “sequential plasticity.” The way I picture it, a traveling wave tags one region after another in order, passing the day’s memories along the cortex like a relay, and a lonely wave drops the baton after the first handoff. That picture is my gloss, not a finding. It does explain why the team measured how far the waves traveled, not only how many there were, and compared that with memory.

The tau sits where the waves begin

The waves start in the frontal cortex, and that’s where the tau showed up. On PET scans of the older participants, more frontal tau went with more breakdown of the long traveling waves. “It’s how much pathology you have in frontal areas where global waves originate,” Sharon said. The matching location makes the pattern fit together, but it doesn’t show that tau causes the change.

Age alone didn’t decide who had which kind of wave. Some older adults had waves that looked like a 20-year-old’s, according to Alzforum. I take that as the hopeful result: some brains keep their deep sleep coordinated into their 70s, and someone should find out how.

The release reports two further results. In a subset of participants followed for several years, the people whose frontal tau rose also showed less coordinated waves and worse overnight retention of new information. A separate group of older adults, studied with neurologist Yo-El Ju at Washington University in St. Louis, showed the same direction using spinal fluid instead of a scanner: people with more isolated slow waves had higher tau-to-amyloid ratios. “It’s collected in a totally different place and in a different cohort,” Sharon said. That counts as converging evidence, not a repeat of the PET result. A spinal tap can’t show where in the brain the tau sits, and a ratio can rise either because tau goes up or because amyloid in the fluid goes down.


The two Alzheimer’s antibodies approved in the US in recent years, lecanemab and donanemab, both target amyloid plaques. In this human data, the protein that tracked with disrupted deep sleep was tau. In the preprint, which Alzforum reports included 61 young adults and 74 older adults, the researchers “saw no relationship between slow-wave activity and Aβ.” Alzforum also notes that memory decline in Alzheimer’s “correlates much more tightly with tangles than plaques.”

Marc Aurel Busche of University College London disagreed in a comment on Alzforum. In his group’s mice, impaired wave travel did cause memory deficits, but it was “instead caused by soluble Aβ” rather than plaques, and the deficits “could be rescued by enhancing GABAergic inhibition.” The PiB PET scan used in this study images plaques, so this dataset can’t resolve his point. It is still a human result that points to tau, in a field whose approved drugs target plaques.

“Leads to” became “associated with”

The 2024 preprint on bioRxiv was titled “Tau pathology leads to lonely non-traveling slow waves that mediate human memory impairment.” That title claims a causal chain in which tau disrupts the waves and the disrupted waves produce the forgetting. The published title says tau is “associated with” the waves, which are “linked to” memory impairment. The verbs got weaker between bioRxiv and Nature Neuroscience, and the weaker verbs match what the data can show.

Sharon says the same. “We can see tau and lonely waves rising together over time,” he said. “What we cannot yet say is which leads.” Matthew Pase of Monash University, who wasn’t involved, told Alzforum “it’s like the chicken and egg.” He noted that something upstream, such as the ApoE4 gene variant, could drive both the tau and the disrupted sleep. So there are at least three possibilities: tau disrupts the waves, the waves fail first, or a third factor causes both. That caveat will reach far fewer people than the word “lonely” did.

There’s no test for lonely waves at your doctor’s office, PET tau scans are expensive, and no trial has shown that restoring traveling waves protects memory. Pase gave the hopeful version: “maybe if we improve sleep, we could potentially improve memory in people with very early AD.” That still has to be tested in a trial. The variable at the center of the study, though, is deep sleep, which nobody can patent and which a lot of us give up for one more email.

I won’t buy a headband to count my lonely waves, but I will stop spending the first, deepest hours of my night on another episode.

Sources

  1. Nature Neuroscience paper
  2. 2024 preprint on bioRxiv
  3. UC Berkeley release on ScienceDaily
  4. Alzforum’s coverage of the preprint