Long COVID and How it Impacts Neurological Functions

Debilitating for some, long COVID can lead to cognitive impairment. It persists more than three months after a person experiences the SARS-CoV-2 infection associated with COVID-19. Known as “brain fog,” the post-viral condition encompasses chronic neurological symptoms such as memory failure, headaches, and fatigue, which linger after the infection’s acute stage has passed.

COVID remains a new phenomenon. Therefore, the medical community’s knowledge is rapidly evolving as it researches it. In the early months of the pandemic, direct infection of the brain was one of the suspected reasons for neurological symptoms in patients. However, the medical community could not establish clear evidence of the link between COVID-19 persisting or proliferating in the brain.

Microclots might contribute to long-term COVID-19, impacting blood vessel linings and low serotonin levels. An amino acid-derived chemical, serotonin acts as a hormone and monoamine neurotransmitter. It carries messages in the central nervous system (CNS) that pass among nerve cells within the brain and throughout the body via the peripheral nervous system.

Most serotonin resides in the gastrointestinal tract, and platelets accept it after the tract releases it into the blood system. As part of the food digestion process, the cells lining the intestines produce 90 percent of serotonin, and the brain produces the remaining 10 percent. In addition to regulating bowel function and wound healing capacity, serotonin affects mood and sleep, and the latter may be implicated in brain fog.

Researchers at the Charite Neuropathology Institute in Berlin studying long COVID found that, although the virus may be transmitted by immune cells into the brain, the virus does not directly infect brain cells. They hypothesized that neurological symptoms are a side effect of the massive immune response the body marshals in countering the virus.

In particular, molecular processes appear to change in certain brain cells, activating interferon signaling pathways. This is a typical response when viral infection occurs, as neurons react to an inflammation in a different region of the body. Affected neurotransmitters in the brainstem may be responsible for fatigue, as these are the same cells that regulate mood, motivation, and drive.

Researchers have found a prevalence of reactive nerve cells associated with COVID in the vagus nerve nuclei. The vagus nerve nuclei sit in the brainstem and link with cells that extend to the intestine, lungs, and heart. Although the infection does not directly impact the vagus nerve nuclei, the inflammation in the peripheral organs does disrupt its normal function.

A study published in Nature Neuroscience in February 2024 focused on vascular disruption and disruption of the blood–brain barrier (BBB) function as another possible reason for long COVID. Undertaking a transcriptomic analysis of peripheral blood mononuclear cells among patients with brain fog, researchers found impairment of the coagulation system and adaptive immune response.

As a result of sustained systemic inflammation, paired with BBB dysfunction, brain endothelial cells became inflamed, and the persistence of S protein and other viral components of COVID occurred. With the long-term influence of S protein on cerebrovascular functions still unknown, researchers believe it merits additional investigation.

How Trauma Changes Neurophysiology

Research shows that trauma impacts neurophysiology, or the way the nervous system, which spans the spinal cord, brain, sensory organs, and peripheral nerves, functions. The neurological network encompasses whole organs, single cells, subcellular compartments, and cellular networks, which function in a highly calibrated way. The generation and transmission of electrical impulses, within and across neurons, tie the functions together.

Trauma, whether an accident, sports injury, loss of a loved one, or other unexpected life event, becomes embedded in the body and mind. It influences perceptions, behavior, and how individuals navigate future encounters and life events. Trauma can disrupt neural patterning and self-regulatory capacities, reflecting the nature of the traumatic stress response, where the brain and nervous system work in tandem to mobilize neurophysiological processes that help deal with the cause of stress. Prolonged activation of the stress response in the brain regions reorganizes the brain-mind-body system in ways that may compromise responses to subsequent trauma and navigating obstacles in daily life.

The mechanisms center on the prefrontal cortex, which makes choices and decisions, remembers information and plans rational responses. Extreme fear, shock, or trauma may trigger the “fear circuitry,” bypassing the prefrontal cortex. It leads to the “fight or flight” sensation or the deer caught in the headlights freezing, the most common physiological response. Alternatively, neurological systems may respond with tonic or collapsed immobility, as when a possum goes limp after being frightened. In humans, this may take the form of passing out or feeling extremely sleepy, an unconscious choice hardwired into the physiology.

In the aftermath of trauma, there may also be a sense of dissociation or feeling disconnected from the body. Rather than actively making decisions using the prefrontal cortex, the individual relies on habitual, socialized modes, with the person operating on autopilot.

A 2022 study by researchers at the University of Rochester’s Del Monte Institute for Neuroscience focused on trauma exposure and mechanisms the brain uses in distinguishing between the safe and the potentially dangerous. After trauma, specific mechanisms may no longer function properly, mainly when emotion is present, including the salience network that the brain employs for survival and learning. Such changes occur whether psychopathologies such as depression, anxiety, and PTSD emerge or not.

The researchers applied fMRI imaging and recorded participants’ brain activity as they viewed images of various-sized circles, including one associated with a threat or small shock. Participants who had trauma-exposed brains without psychopathologies reacted differently than the others. Their neural circuits compensate for brain processing changes by activating the executive control network, one of the brain’s primary networks. It regulates cognitive processes that work together to enable goal-directed thoughts and behaviors.

In a parallel study, fMRI scans of participants with PTSD revealed lower signaling between the hippocampus, which controls memory and emotion, and the salience network, which enables learning and survival. In addition, less signaling occurred between the amygdala (also emotion-linked) and the default mode network, which is activated when a person does not focus on the external world. The finding suggests that individuals exposed to PTSD-causing traumas have particular challenges discriminating between actual and perceived threats when emotion is involved.

Emotions overload the abilities of cognitive systems to distinguish between danger, safety, and reward, resulting in an overgeneralization towards danger. The impaired fear generalization syndrome is common with several psychopathologies, but the extent to which trauma-exposed people encounter such issues requires more in-depth study.

Understanding the Present and Future States of AI

Artificial intelligence (AI) uses computer science and large data sets to generate automated problem-solving. More specifically, different theories and strategies for AI development strive to create machines capable of reproducing human thought processes, allowing machines to perform tasks that previously required some human oversight.

In recent years, several AI-based initiatives and calls for governments to regulate AI. Potential threats range from cyber security risks to replacing upwards of 2.4 million American jobs via AI automation by 2030. While investing in effective regulatory measures is important, it may surprise individuals to learn that true AI does not exist and may never be developed by humans.

AI falls into three primary groupings: artificial narrow intelligence (ANI), artificial general intelligence (AGI), and artificial superintelligence (ASI). As the names suggest, machines operated by ANI can perform and problem-solve for one or a few simple, related tasks. AGI, by comparison, is an AI with problem-solving skills on par with a human mind. Finally, ASI systems can problem-solve like a human using the total computing power of a machine, resulting in superior intelligence.

As of the start of 2024, human scientists have only been able to produce ANI or weak AI. Developers build systems with specific goals in mind. Then, the AI might achieve minor innovations and deviations but remain constrained to a single function.

Facial recognition technology provides a clear example of ANI. The technology can identify human faces from photographs and videos, an impressive engineering feat, but cannot perform any actions beyond the designated function. Other examples of ANI include speech recognition technology and self-driving vehicles. Although ANI is a type of AI, it does not replicate human intelligence. Instead, it simulates human behavior in a highly specific context.

AGI, or strong or deep AI, might be within the reach of human engineers but does not yet exist. Despite AGI not existing in reality, it is the version of AI often depicted in popular culture. If real, AGI machines would possess a complete intelligence roughly on par with the human brain. AGI would apply to any problem or task instead of operating under a strict set of conditions. In this way, strong AI could achieve independent thought, reflection, and action and shift through these stages like a human mind.

It bears repeating that strong AI requires extensive oversight and regulation. However, no one knows for sure how much more time scientists need to achieve AI comparable to the human mind. Developers would need to program a complete set of cognitive functions and subject those functions to a depth of experiential learning. Even a functioning AGI system would need time to develop beyond a limited series of tasks before applying general intelligence to a more diverse set of problems. To put this challenge in context, scientists and physicians do not fully understand the human brain, which makes it nearly impossible to replicate how the brain works in a different system.

Lastly, science fiction often depicts AI systems capable of surpassing human intelligence. In truth, humans are nowhere close to developing machines capable of self-awareness. A key distinction between strong AI and superintelligence is that the latter would recognize its needs and act toward those interests, even if they conflict with the system’s initial programming. Research indicates that an ASI developing human emotions is highly unlikely, and it remains unclear whether humans will ever develop systems beyond weak AI.