The Fog That Won’t Lift: What Long COVID Is Really Doing to the Human Brain
Imagine being sharp and efficient, someone who prides herself on tracking a dozen details at once, and then catching COVID-19 with an infection that is only moderate, with no ICU stay and no ventilator. Weeks later you cannot remember whether you took your own medication. You lose words mid-sentence. You stare at a grocery list you wrote yourself and it reads as though it were in a foreign language. Patients with long COVID describe exactly this kind of experience.
This experience is not a curiosity. It is, researchers now believe, one of the defining medical phenomena of the decade. Long COVID, formally termed Post-Acute Sequelae of SARS-CoV-2 infection (PASC), affects an estimated 65 million people worldwide, according to figures published in Nature Reviews Microbiology in 2023. Among the most debilitating and least understood of its symptoms are the neurological ones: the crushing fatigue, the memory failures, the inability to concentrate, the personality shifts, and the sensation patients uniformly describe as brain fog. Years of intensive research have begun to reveal what is actually happening inside the skulls of these patients, and the picture is both more concrete and more alarming than early skeptics imagined.
Defining the Problem: Who Gets Brain Fog and How Often
Before scientists could explain what was going wrong in the brain, they needed to establish how widespread the problem actually was. Early in the pandemic, neurological symptoms were often dismissed as anxiety, as understandable psychological reactions to a frightening illness. That dismissal has not aged well.
A landmark study published in The Lancet Psychiatry in 2021, drawing on data from more than 236,000 COVID-19 survivors, found that roughly one third experienced at least one neurological or psychiatric condition within six months of infection. Cognitive deficits, the clinical term for what patients call brain fog, appeared in a significant subset. A 2022 study in Nature, led by researchers at Oxford, found that people who had been infected with SARS-CoV-2 showed measurable reductions in gray matter thickness and overall brain size compared with uninfected controls, even in those who had experienced only mild illness.
The 2022 Oxford study was particularly striking because its participants had undergone brain imaging before their infections as part of the UK Biobank project, meaning researchers had genuine before-and-after comparisons rather than relying on estimates. The affected regions included the orbitofrontal cortex and the parahippocampal gyrus, areas involved in memory, emotional regulation, and executive function. Participants also performed worse on cognitive tests after infection.
Crucially, severity of the initial illness was not a reliable predictor. Many patients with severe long COVID neurological symptoms had never been hospitalized. This detail demolished a comforting early narrative that lasting brain effects were only a risk for the sickest patients.
Inside the Inflamed Brain: What Neuroinflammation Actually Means
The word neuroinflammation appears constantly in long COVID research, but its meaning is sometimes obscured by jargon. Put simply, it refers to an activated inflammatory state within the brain and central nervous system. The brain has its own immune cells, called microglia, which normally act as custodians, clearing debris and pathogens. In neuroinflammatory conditions, those cells become persistently activated, releasing cytokines and other signaling molecules that, over time, damage the very neural tissue they are supposed to protect.
Evidence that this is occurring in long COVID patients came from several directions at once. A 2024 study published in Brain, Behavior, and Immunity, led by Michael VanElzakker and colleagues at Massachusetts General Hospital, used PET brain imaging with a specialized radioactive tracer to detect neuroinflammation in living long COVID patients. In a small sample of 12 patients compared with 43 healthy controls, they found elevated inflammation across several brain regions, including the cingulate cortex, thalamus and basal ganglia, and it correlated with blood markers of vascular dysfunction.
Other researchers have pointed to the role of mast cells, immune cells found throughout the body and brain, that appear to become dysregulated after SARS-CoV-2 infection. When mast cells degranulate inappropriately, they release histamine and other inflammatory compounds that can cross or disrupt the blood-brain barrier, the protective layer that normally keeps pathogens and large molecules out of the central nervous system. Disruption of this barrier is one proposed mechanism for long COVID’s neurological effects, effectively allowing inflammatory signals that should stay in the body’s periphery to penetrate and agitate brain tissue.
Complementing the imaging studies, autopsy work has added to the picture. Researchers at the National Institutes of Health, publishing in Nature in 2022, examined 44 autopsies of people who died with COVID-19 and found SARS-CoV-2 RNA in many parts of the body, including throughout the brain, in one case as late as 230 days after symptom onset. That suggests the virus or its remnants may persist in the central nervous system, though the same study found little evidence of inflammation outside the respiratory tract, so what the persisting virus does remains an open question.
The Viral Persistence Hypothesis and Its Rivals
Not everyone is convinced that viral persistence drives all long COVID neurological symptoms, and the scientific debate here is genuinely unresolved. Several competing hypotheses exist, and there is growing evidence that long COVID may not be a single disease at all, but a cluster of distinct syndromes with overlapping symptoms but different underlying mechanisms.
One prominent alternative hypothesis focuses on autoimmunity. Several research groups have found elevated levels of autoantibodies in long COVID patients, proteins produced by the immune system that mistakenly attack the body’s own tissues, including neural structures. A 2021 study in the Journal of Translational Autoimmunity identified autoantibodies targeting G protein-coupled receptors, which regulate the autonomic nervous system. Dysfunction of the autonomic nervous system, which controls heart rate, blood pressure, and breathing without conscious effort, could explain many of the symptoms that overlap with a condition called dysautonomia, including the fatigue, cognitive difficulties, and exercise intolerance that characterize long COVID.
A third hypothesis centers on the reactivation of latent viruses. Epstein-Barr virus, which causes mononucleosis and lies dormant in most adults, has been found at elevated reactivation levels in some long COVID patients. The idea is that SARS-CoV-2 disrupts immune surveillance sufficiently to allow these dormant pathogens to reactivate and produce their own inflammatory effects. Some studies have found markers of this reactivation to be more common in people with long COVID than in controls.
Each of these mechanisms, viral persistence, autoimmunity, and latent virus reactivation, may be simultaneously true in different patients or even in the same patient at different times. This biological complexity is part of what has made treatment so elusive and what has frustrated patients who are often told, even years into their illness, that there is no proven therapy.
What Brain Imaging Is Teaching Clinicians
Advances in neuroimaging have arguably been the most important methodological development in this field. For years, patients reporting cognitive symptoms would undergo standard MRI scans and be told everything looked normal. That conclusion now appears to have been an artifact of the tools being used rather than evidence that nothing was wrong.
Functional MRI (fMRI), which measures brain activity by tracking blood oxygen levels, has revealed abnormal connectivity patterns in long COVID patients even when structural scans appear unremarkable.
Even more sensitive are techniques like diffusion tensor imaging (DTI), which maps the integrity of white matter tracts, the highways that connect different brain regions. Several research groups have found white matter abnormalities in long COVID patients that standard MRI would miss entirely. These subtle lesions may explain why patients who appear neurologically intact on conventional exams still perform poorly on tasks requiring sustained attention or working memory.
Spectroscopy-based MRI, which can measure the chemical composition of brain tissue, has been used in smaller studies to look for markers of neuronal injury and inflammation in long COVID patients.
These imaging findings carry both diagnostic and therapeutic implications. If neuroinflammation is quantifiable, it can in principle be tracked over time, tested against interventions, and used to stratify patients into subgroups likely to respond to different treatments. That is the direction researchers are pushing toward, though the clinical translation remains in early stages.
The Lived Reality: Fatigue, Cognition, and the Cost to Society
Numbers and mechanisms can obscure what this illness actually means for the people living inside it. Brain fog is not merely forgetfulness or tiredness. Patients describe an inability to read a paragraph and retain what they read, difficulty tracking a conversation, errors in simple arithmetic they would never previously have made, and a profound exhaustion that worsens with cognitive exertion in a phenomenon known as post-exertional malaise. This last feature links long COVID neurological illness to myalgic encephalomyelitis and chronic fatigue syndrome (ME/CFS), a condition that long COVID researchers have increasingly drawn parallels with.
The economic consequences are substantial. A 2022 Brookings Institution analysis estimated that roughly 1.1 million Americans were not working because of long COVID. The cognitive dimension of that disability is central: many patients can walk and perform basic activities of daily living but cannot sustain the mental demands of professional work.
For patients managing these symptoms at home, practical accommodations matter enormously. Noise-canceling environments, reduced screen time, and careful pacing of cognitive tasks are all commonly recommended. Some patients have found that noise-canceling headphones help reduce the sensory overload that worsens their cognitive symptoms. For those dealing with the sleep disruption that frequently accompanies neurological long COVID, a cooling weighted blanket has anecdotal support in the ME/CFS community as an aid for nervous system regulation. And patients who need to track symptoms and cognitive patterns for medical appointments have found that smart journals or health tracking notebooks help them communicate more precisely with their clinicians.
These are modest tools against a serious illness, and they should not be mistaken for treatments. But for people navigating a condition that medicine has not yet solved, they can make the difference between productive engagement with care and complete disorientation.
Where the Science Is Headed: Treatments, Trials, and Cautious Hope
The field has moved from documenting that neurological damage occurs to beginning systematic trials of interventions targeting suspected mechanisms. Several directions show genuine promise, even if none has yet produced a proven, widely available treatment.
Low-dose naltrexone, a drug thought to modulate the immune system, has been tested in a number of long COVID trials, though it is not an established treatment. Antihistamines, particularly H1 and H2 blockers used in combination, have been reported by some clinicians to provide modest symptom relief in patients whose presentations suggest mast cell involvement. Anticoagulation therapy, targeting the microclots some researchers believe contribute to reduced cerebral blood flow, has been explored in smaller studies, primarily by South African researcher Resia Pretorius and colleagues, with intriguing preliminary results that await larger replication.
Longer-term, researchers are hoping that drugs currently in development for other neuroinflammatory conditions, including some being tested for Alzheimer’s disease and multiple sclerosis, may have application in long COVID. The mechanistic overlaps are real. The chronic microglial activation, the blood-brain barrier disruption, the white matter changes; these are not unique to long COVID, and therapeutic approaches proven in other contexts may translate.
What remains most urgent is the recognition that this is a biological disease demanding biological investigation and biological treatment. The patients who were told their symptoms were psychosomatic, who lost jobs and relationships and years of their lives while awaiting a diagnosis that the medical system was slow to provide, deserve more than validation after the fact. They deserve a research infrastructure and a clinical system commensurate with the scale of what has happened.
For many patients, recovery is incomplete and hard-won, and the ceiling is hard to see. For millions of people worldwide, that ceiling is still the central unanswered question. Science, moving faster than it has for comparable post-viral conditions in the past, is working urgently to raise it.
As an Amazon Associate, The Rough Idea earns from qualifying purchases.