For people struggling with post-COVID “brain fog,” new findings suggest that cognitive recovery is possible, though it may take years. A large-scale study recently published in Brain, Behavior, & Immunity – Health tracked cognitive changes over three and a half years and found significant improvements in most mental functions. Yet even with this progress, some participants continued to experience deficits in mental speed and flexible thinking.
Since the early days of the COVID-19 pandemic, patients have reported lingering symptoms such as mental slowness, forgetfulness, and trouble concentrating—often described collectively as “brain fog.” While several studies have examined the short-term impact of COVID-19 on brain function, little was known about how these symptoms evolve over several years.
Previous research has offered a mixed picture. Some studies have suggested that cognitive symptoms may improve within a year, while others point to longer-lasting effects, especially in people who were hospitalized. Many of these studies, however, relied on self-reported symptoms or brief online tests, and few used comprehensive in-person assessments with validated neuropsychological tools. Moreover, most did not include diverse or younger populations, making it harder to draw conclusions about how widespread or enduring these impairments might be.
To fill these gaps, researchers from the Mount Sinai Health System launched a long-term prospective study with a large, diverse sample of adults who had confirmed COVID-19 infections. Their goal was to track cognitive changes using validated testing methods and to determine which factors might influence the pace and extent of recovery.
“The focus of my research program pre-pandemic was on neurocognition in the context of chronic medical illnesses,” explained study author Jacqueline H. Becker, a clinical neuropsychologist and assistant professor of medicine at the Icahn School of Medicine at Mount Sinai.
“Being at Mount Sinai, right at the epicenter of the pandemic, I witnessed firsthand the toll COVID-19 was taking on patients and families. At the time, no one anticipated something like Long COVID, but it was immediately clear that the effects of the pandemic would be long lasting, especially for brain health. I felt a strong responsibility to contribute to the recovery effort in some way, which eventually became a natural extension of my work.”
“At the inception of Mount Sinai Health System’s (MSHS) Post-COVID registry, we decided to include a neurocognitive battery to track the effects of infection on cognition over several years. This turned out to be crucial, as we became one of the first and only centers globally to collect objective, in-person neuropsychological measures as early as April 2020. That early work positioned us to better understand the long-term cognitive consequences of COVID-19 and how they evolve over time.”
The research team analyzed data from 1,553 participants in the Mount Sinai Post-COVID-19 Registry. These adults were recruited from a pool of patients who had tested positive for COVID-19 and received care at one of Mount Sinai’s facilities in New York City, starting in April 2020 and followed through January 2024.
Participants completed a comprehensive battery of neuropsychological assessments designed to measure various domains of cognition, including attention, working memory, verbal learning, memory recall, language fluency, processing speed, and executive functioning. These tests were administered in either English or Spanish, depending on the participant’s preference, and results were standardized to account for age, sex, and education.
“Unlike many prior investigations that relied on self-reported ‘brain fog’ or online testing, we used validated, in-person neuropsychological assessments, strengthening the validity of the findings,” Becker said.
Participants were tested once a year, up to four times, depending on when they joined the study. In addition to cognitive testing, they completed surveys about their medical history, mental health symptoms such as anxiety and depression, and fatigue levels. Other factors such as vaccination status, body mass index, and where participants received care during their initial infection (outpatient, emergency room, or inpatient) were also recorded.
To ensure the integrity of the results, the study excluded anyone with a prior diagnosis of cognitive problems or those who showed signs of suboptimal effort on performance validity tests.
At the start of the study—when participants were assessed within six months of recovering from COVID-19—many showed mild to moderate impairments across multiple cognitive areas. On average, their test scores were about half to one and a half standard deviations below the normative mean.
“Like earlier studies, our results confirm that cognitive impairment is common after COVID-19 and can persist well beyond the acute phase,” Becker told PsyPost.
Over the following 42 months, most cognitive domains showed signs of improvement. In particular, verbal learning and memory recall demonstrated the largest gains. For example, verbal learning scores improved from an initial average of 1.26 standard deviations below the mean to within the normal range by the end of the study period. Language abilities, such as phonemic and semantic fluency, also improved over time.
Processing speed and executive functioning showed more modest gains and were still below average after three and a half years. While there were measurable improvements, the average scores for these domains remained about 1.5 standard deviations below the norm, suggesting lingering impairments for some participants.
“Most earlier studies stopped at 12–24 months; our study was among the first to show trajectories out to 42 months (3.5 years), providing stronger evidence of both recovery and plateauing,” Becker said.
Interestingly, attention and working memory showed relatively little change over time. These areas were only mildly impaired at baseline, which may have left less room for measurable improvement.
“The most important takeaways are that, while cognitive recovery may be gradual, it is a very real possibility for most people with Long COVID,” Becker told PsyPost. “However, there may be a subset that remain with deficits even several years later. Overall, this suggests that people with Long COVID need more support for recovery.”
Participants who had a body mass index under 25—a range generally considered normal—tended to experience greater cognitive improvements. No other factors, including age, sex, vaccination status, or the severity of the initial COVID-19 infection, significantly influenced the pace of cognitive recovery.
“It was surprising that COVID-19 severity did not predict cognitive recovery trajectories and that lower BMI was the only factor associated with cognitive recovery over time,” Becker said.
The study offers one of the most comprehensive long-term assessments of cognitive recovery after COVID-19 to date. Still, the authors acknowledge some limitations. One key issue is that the sample may not fully represent all people with Long COVID. While the study recruited from a broad base, those with the most severe symptoms or those who had fully recovered might have been less likely to participate, potentially skewing the results.
The average participant was relatively young and well-educated, which could limit how generalizable the findings are to older adults or those with fewer cognitive or educational resources. It’s also possible that some of the observed improvements were partly due to practice effects, although the use of alternate test forms and annual spacing of assessments likely helped minimize this.
“Questions that remain unanswered are, what biological mechanisms (e.g., inflammation, vascular health, neuroplasticity) drive both persistence and recovery of deficits and which interventions (e.g., cognitive rehabilitation, lifestyle modification, medical therapies) can accelerate or optimize recovery,” Becker explained.
The researchers suggest that future studies could explore targeted interventions to promote recovery, such as cognitive rehabilitation or lifestyle changes. They also advocate for research into biological markers—such as inflammation levels—that might help predict who is most at risk for long-term impairment and who is likely to recover more quickly.
“Currently, I am working on developing and piloting interventions (including cognitive rehabilitation) to accelerate cognitive recovery in people with Long COVID,” Becker told PsyPost. “My long-term goals are to increase awareness of Long COVID and other infection-associated chronic conditions, inform health policy to better support affected individuals, and ultimately improve quality of life for patients through evidence-based care.”
The study, “Neurocognitive Trajectories in Long COVID: Evidence from Longitudinal Analyses,” was authored by Jacqueline H. Becker, Jia Li, Jenny J. Lin, Alex Federman, Emilia Bagiella, Minal S. Kale MD, Daniel Fierer, Logan Bartram, and Juan P. Wisnivesky.
