The question

Loneliness has been repeatedly linked to worse brain health in old age, and a large meta-analysis of more than 600,000 people put the associated increase in dementia risk at roughly 31 percent.^[2] But "linked to worse brain health" hides two very different possibilities. Loneliness could pull down a person's overall memory level — a one-time gap that then stays roughly constant. Or it could act more like a corrosive, speeding up the year-on-year rate of decline. Those lead to different worries and different interventions, and most studies measure memory only once, so they cannot tell them apart. This study set out to separate the two.^[1]

What they did

The team drew on the Survey of Health, Ageing and Retirement in Europe (SHARE), a long-running study that revisits the same older adults every couple of years. They used three waves collected between 2012 and 2019 — a seven-year window — across 12 European countries grouped into four regions.^[1] The sample was 10,217 people aged 65 to 94, and the authors excluded anyone with a history of dementia or difficulty with basic daily activities, so the analysis is about ordinary aging rather than diagnosed disease.

Loneliness was measured once, at the start, using the Revised UCLA Loneliness Scale, and participants were sorted into low, average, and high loneliness. Memory was tested at every wave with a standard 10-word list, scored two ways: immediate recall (words remembered right after hearing them) and delayed recall (words remembered a few minutes later). Because each person was measured repeatedly, the researchers could model both starting scores and the slope of change, while adjusting for factors such as physical activity, social participation, depression, and diabetes.^[1]

What they found

At baseline, the highly lonely group remembered fewer words than the low- and average-loneliness groups on both tests — about 0.24 standard-deviation units lower for immediate recall and 0.21 lower for delayed recall.^[1] That is a modest but real gap: loneliness sat alongside a noticeably lower memory score even after accounting for mood, health, and activity.

The more striking result was what did not happen. Over the seven years, the memory of lonely participants declined at essentially the same rate as everyone else's. Loneliness shifted the starting line down but did not steepen the downhill slope.^[1] The authors describe this as loneliness mattering more for the initial state of memory than for its progressive decline.

Why it matters

The popular framing — that loneliness slowly erodes the aging brain — implies that a lonely 70-year-old is on a faster track toward serious memory loss. This study does not support that trajectory story. Instead it points to a level difference that is already present and then travels alongside, rather than a widening gap. Practically, that reframes loneliness less as an accelerant of decline and more as a marker of, or contributor to, memory that is already running lower. It also fits a broader pattern in which subjective loneliness carries its own signal for cognitive outcomes even after objective social isolation and depression are taken into account.^[2]

What this does not prove

This is an observational cohort, so it cannot show that loneliness causes lower memory. The reverse is entirely plausible — people whose memory is already weaker may withdraw socially and feel lonelier — and unmeasured factors (education, early-life health, personality) could drive both. Loneliness was recorded only once, at baseline, and treated as fixed, but in real life it rises and falls; the study cannot capture how changing loneliness tracks with changing memory. The seven-year window and the exclusion of people with dementia or heavy disability mean the findings speak to relatively healthy older adults, not to those already declining fast, and a longer follow-up might reveal slope differences this one could not. Memory here is one word-list task, not the full range of cognition. And the sample is European; the pattern may differ elsewhere.^[1]

What happens next

The obvious next step is to let loneliness vary over time — modeling whether people whose loneliness worsens also see memory changes, rather than freezing it at one reading. Longer follow-ups and samples that include people closer to dementia onset would test whether a rate effect emerges late, when this study's healthy sample would not show it. And because association is not causation, the causal question ultimately needs designs that can move loneliness — such as trials of interventions that reduce it — to see whether memory responds. For now, the honest summary is narrow and useful: among healthy older Europeans, loneliness marked a lower memory starting point, not a faster fall.^[1]

---

References

1. Venegas-Sanabria LC, Pineda-Mateus E, Borda MG, Satorres E, Bueno-López C, Meléndez JC. Memory trajectories in lonely individuals in Europe: an analysis of the Survey of Health, Ageing and Retirement in Europe (SHARE). Aging & Mental Health. 2026. doi:10.1080/13607863.2026.2624569
2. Luchetti M, Aschwanden D, Sesker AA, et al. A meta-analysis of loneliness and risk of dementia using longitudinal data from >600,000 individuals. Nature Mental Health. 2024. doi:10.1038/s44220-024-00328-9