For years, "long COVID brain fog" has been a real, widely-reported experience without a clear biological explanation — which made it easy to dismiss as vague or psychological. A study published in October 2025 changed that, by directly imaging what's actually different in the brains of people experiencing it.
What the researchers actually measured
A team led by Fujimoto and colleagues used a specialized PET imaging technique ([11C]K-2 tracer) to directly visualize the density of AMPA receptors — a type of receptor central to how neurons communicate — in the living brains of 30 long COVID patients with cognitive impairment, compared against 80 healthy volunteers.1
What they found
Long COVID patients showed widespread increases in AMPA receptor density throughout the brain. Critically, the degree of receptor increase tracked closely with how severe a person's cognitive symptoms were — this wasn't a diffuse, unrelated finding, it lined up with the actual brain fog people were reporting. Higher inflammatory markers were also associated with higher receptor density, pointing toward neuroinflammation as a plausible driver.
The degree of AMPA receptor increase tracked with symptom severity — a measurable biological finding lining up with what patients were actually reporting, not a diffuse, unrelated result.
Accurate enough to be a real diagnostic candidate
The imaging approach distinguished long COVID patients from healthy controls with 100% sensitivity and 91.2% specificity in this study — numbers strong enough to suggest this could become an actual diagnostic tool, not just a research finding, pending replication in larger and more diverse samples.
That last clause is doing a lot of work, and it's worth being blunt about why. Sensitivity and specificity of that size in a first study, in 30 patients against 80 healthy controls, almost always shrink on replication. The comparison that matters clinically isn't long COVID versus healthy — it's long COVID versus everyone else who turns up with fatigue and poor concentration, which includes untreated sleep apnoea, thyroid disease, anaemia, depression and chronic insomnia. Until the tracer is tested against that group, the 91.2% figure tells you how well it separates sick from well, not how well it would separate one cause of brain fog from another.
A second mechanism, from a different direction
The AMPA finding isn't the only biological signal. In 2024, a team publishing in Nature Neuroscience used dynamic contrast-enhanced MRI to measure blood-brain barrier leakage directly, comparing 11 people who had long COVID with cognitive impairment, 11 who had long COVID without it, and 10 who had recovered fully.2 Only the brain fog group showed significantly increased whole-brain leakage. The long COVID patients without cognitive symptoms had normal barrier integrity, despite still being unwell in other ways.
The leakage was greatest in the temporal lobes on both sides and in the frontal cortex, and temporal-lobe disruption correlated with how long each person's loss of smell had lasted. The brain fog group also showed raised transforming growth factor-beta, raised glial fibrillary acidic protein — a marker of astrocyte activation — reduced brain volume and cortical thinning. The authors named their own limits clearly: no cerebrospinal fluid samples to confirm barrier breakdown, no longitudinal scans to show whether the leakage resolves, and a sample far too small to generalise from. Two small studies pointing at inflammation from different angles is encouraging. It is not the same as a settled mechanism, and it's a useful reminder that "inflammation" is both a real mechanism and a marketing frame depending on who is using the word.
How big is the effect, in people rather than in scanners?
The largest attempt to quantify this tested 112,964 adults in England and compared cognitive performance across illness histories.3 People whose symptoms had resolved within four weeks scored 0.23 standard deviations below the no-COVID group — roughly the equivalent of 3 IQ points. People with symptoms still unresolved at the time of testing scored 0.42 SD lower, about 6 points. Those who had been in intensive care were 0.35 SD lower, about 9 points on the authors' framing. Reinfection added roughly another 2 points. Earlier variants produced larger deficits than Omicron.
Those numbers cut both ways. They are real, measured in a very large sample, and they validate the experience. They are also small at an individual level — a 3-point average shift is invisible in one person's day and detectable only across a hundred thousand of them. The authors said plainly that the long-term persistence of these deficits and their clinical relevance remain unclear. Anyone quoting "COVID costs you 9 IQ points" is quoting the intensive-care subgroup and dropping the rest of the table.
How common is it, actually
A 2024 systematic review pooled 17 studies covering 41,249 people with long COVID. Brain fog or concentration problems came out at 23.7%, but with a 95% confidence interval of 10% to 51% — an enormous spread that tells you how inconsistently different studies defined and measured the thing. Memory complaints came out at 21%, anxiety at 21.9%, depression at 21.4%, insomnia at 11.6%.4 The authors' main stated limitation is the one that matters most here: most included studies had no non-COVID control group, so these figures describe how often symptoms are reported, not how much of that is attributable to the infection.
The differential, and the tells
Persistent cognitive symptoms after an infection are not self-evidently long COVID, and the conditions that mimic it are the ones that are actually treatable. Thyroid dysfunction produces slowed processing that resolves with treatment. Iron deficiency and low B12 do the same — and the B12 trial evidence is messier than the supplement aisle suggests. Undiagnosed obstructive sleep apnoea is common and easy to miss: if you snore, wake unrefreshed or have witnessed breathing pauses, the STOP-BANG questions take two minutes. Depression presents with concentration and memory complaints as core features, not as an afterthought. The practical move is the boring one — ask for the four blood tests before concluding anything, and read the wider list of causes before settling on the one that was in the news.
What actually helps right now — including what didn't work
There is no approved drug for long COVID cognitive impairment, and AMPA-receptor targeting is a research direction rather than a treatment. The most rigorous test of non-drug approaches to date, the RECOVER-NEURO trial, randomised 328 adults across 22 US sites to five arms including online brain training, a structured cognitive rehabilitation programme and transcranial direct current stimulation.5 None of them beat the comparison groups. Every arm improved modestly, which is exactly the pattern you get when improvement is driven by time, attention and repeated testing rather than by the intervention. The lead investigator's summary was that none of the rehabilitation approaches proved effective.
That leaves pacing, treating what is treatable, and time. It does not leave a supplement. No Proco product has been tested in long COVID, we are not going to imply otherwise, and if a company tells you their capsule addresses post-viral cognitive impairment, they are making a claim no trial supports.
Why this matters beyond the mechanism itself
For a long time, "brain fog" without an obvious lab abnormality has been hard for patients to get taken seriously about, precisely because there wasn't a visible biological marker. These studies are a real step toward establishing long COVID brain fog as a measurable, physical condition — with specific biological systems as potential treatment targets for future drug development, not just a symptom to describe and wait out.
What this doesn't mean, yet
These are small studies that haven't been replicated at scale. They don't yet translate into an available clinical test or an approved treatment. If you're experiencing persistent cognitive symptoms after a COVID infection, the honest next step is a conversation with your doctor about what's currently available for evaluation and management, not waiting for this specific research to become clinical practice.
References
- Fujimoto Y, et al. Systemic increase of AMPA receptors associated with cognitive impairment of long COVID. Brain Communications, Vol 7, Issue 5, October 2025. academic.oup.com
- Greene C, et al. "Blood-brain barrier disruption and sustained systemic inflammation in individuals with long COVID-associated cognitive impairment." Nature Neuroscience, 2024. nature.com
- Hampshire A, et al. "Cognition and Memory after Covid-19 in a Large Community Sample." New England Journal of Medicine, 2024. nejm.org
- "Prevalence of mental health conditions and brain fog in people with long COVID: A systematic review and meta-analysis." General Hospital Psychiatry, 2024. eprints.whiterose.ac.uk
- RECOVER-NEURO randomised controlled trial results, RECOVER COVID Initiative, 2025. recovercovid.org
This article is for informational purposes only and has not been evaluated by the FDA. It is not intended to diagnose, treat, cure, or prevent any disease, including long COVID. If you're experiencing persistent cognitive symptoms after a COVID infection, speak with a healthcare provider for evaluation.
Related reading: What Actually Causes Brain Fog? and Grief and Your Brain: What the Research on "Grief Brain" Shows
See Cognitive Support's full ingredient list at procohq.com.