On January 30, 2020, the World Health Organization [1] declared the infectious disease known as coronavirus (Covid-19) a pandemic, raising concerns for international public health [2, 3, 4]. From the first reported case of Covid-19 infection until today, over 450 million cases have been recorded worldwide [5]. Despite global efforts to control the spread of the virus and mitigate its effects, Covid-19 continues to pose a significant challenge for the scientific community [6, 7, 8]. Understanding the long-term aspects of the disease has become a fundamental priority to address its clinical, psychological, and social implications [9, 10, 11]. Following acute SARS-CoV-2 infection, approximately 10–20% of patients experience the condition known as Post-Covid or Long-Covid, characterized by the continuation or manifestation of new symptoms that develop 3 months after the initial infection [12]. The main symptoms may include fatigue, muscle pain, shortness of breath, cough, chest pain, palpitations, anosmia, ageusia, sleep difficulties, and cognitive dysfunction that persist for at least 2 months without other explanation [12]. In addition to numerous physical and neurological symptoms, Long-Covid has been associated with a range of mental health issues, such as post-traumatic stress disorder, anxiety, and particularly depression [13, 14]. Among the main risk factors associated with Long-Covid are the quantity and duration of symptoms experienced during the acute phase of the infection. A study conducted by Sudre et al. [15] found that having more than five symptoms within the first week of infection—including fatigue, headache, breathing difficulties, hoarse voice, and myalgia—increased the likelihood of developing Long-Covid. On the contrary, the severity of Long-Covid symptoms does not appear to correlate with the initial severity of Covid-19 infection [16, 17, 18, 19]. This implies that even patients with mild or moderate acute symptoms could be at risk of developing Long-Covid. Other risk factors include having a high body mass index and being female [20, 21]. Among the most common neurological symptoms associated with Long-Covid, cognitive fog or brain fog is frequently observed [22, 23]. According to a study conducted by Graham et al. [24], 81% of the participants reported experiencing this condition after contracting Covid-19 infection. Cognitive fog does not represent a distinct diagnostic category, but it includes a range of neurological symptoms that may manifest as inattention, short-term memory loss, reduced mental sharpness, and cognitive fatigue [25], negatively affecting people’s daily functioning. The causes of brain fog are not clear. Several etiological hypotheses have been proposed by researchers, including inflammation due to cytokine storm, mitochondrial dysfunction in the brain, reduced tissue oxygenation, and diminished blood flow [26]. Some authors have suggested that cognitive fog could be due to the neuroinvasion of the virus through the vagus nerve, trigeminal nerve, or olfactory tract [27]. Research has shown that SARS-CoV-2 can enter the olfactory mucosa and cause anosmia [28, 29]. Unlike transient anosmia observed in other respiratory infections, persistent anosmia is a Long-Covid symptom that may result from the lasting effects of direct viral damage to the olfactory epithelium [30, 31]. Indeed, imaging studies conducted on patients with persistent anosmia from Covid-19 have demonstrated extensive damage to the olfactory epithelium, primarily characterized by thinning of the tissue [32].

Several studies have suggested that Covid-19 infection may impact brain electrical activity, detectable through electroencephalography (EEG). Changes in EEG have been documented in post-Covid-19 patients, associated with the recovery from psychopathological symptoms, including cognitive impairments [33]. Specifically, a widespread slowing of EEG has been noted, primarily in frontal, central, and temporal regions, accompanied by an increase in theta wave activity and a decrease in alpha wave activity [34, 35]. Regarding beta waves, while some studies have reported increased activity in central and temporal regions, others have indicated a more general reduction [36, 37]. Similar EEG changes have been described in mild cognitive impairment, unrelated to viral infections [38]. Moreover, specific EEG disturbances have been well-documented in patients with severe forms of Covid-19 [39]. Building upon these observations, the hypothesis has been put forth to utilize EEG as a standard diagnostic and monitoring tool to assess persistent neurological symptoms from Long-Covid-19 [34]. Specifically, these symptoms can be assessed through neurophysiological measures such as Event-Related Potentials (ERP) and beta band rhythms [40, 41, 42, 43, 44]. ERPs constitute a powerful analytical tool for measuring and examining brain activity associated with sensory, motor, or cognitive functions triggered by specific stimuli or events. The P300 component emerges as a positive deflection in the ERP waveform, typically occurring around 300 ms after the presentation of a rare or unexpected stimulus. This element is closely linked to cognitive processes involving attention and memory. The onset latency of the P300, the time required for the P300 response to manifest following a stimulus, plays a crucial role as it reflects the speed and efficiency of cognitive processing. The amplitude of the P300 is primarily associated with attention, stimulus relevance, and the ability to discriminate between rare or unexpected stimuli and more common or expected ones. Greater amplitude generally indicates a stronger neural response, reflecting increased cognitive engagement in processing the stimulus. Moreover, the analysis of alpha, beta, and theta band rhythms is crucial because these neural oscillations are frequently associated with higher cognitive processes, including those related to attention, concentration, and information processing. In the context of Long-Covid, the persistence of symptoms such as fatigue, cognitive difficulties, and mental confusion could be linked to alterations in the frequency and organization of these bands’ rhythms.

Based on the previous evidence of EEG changes during Covid-19, we hypothesized that post-Covid-19 brain fog may be characterized by specific EEG alterations. These variations could potentially provide an objective confirmation of the diagnosis in individuals reporting cognitive disturbances. To explore this topic further, we conducted an electrophysiological analysis based on EEG in a sample of 20 patients exhibiting symptoms of post-Covid-19 brain fog, assessing them also 8 months post-recovery.

The main objective of this research was to evaluate the effects induced by a previous Covid-19 infection on brain cortical activity in patients with cognitive fog symptoms in the medium and long term. Specifically, the first aim was to determine if the cognitive symptoms reported by patients as cognitive fog correlated with actual alterations in neurophysiological parameters, namely P300 and alpha, beta, and theta bands. The second objective was to assess whether any neurophysiological alterations remained constant over time, to elucidate the long-term impact of Covid-19 infection on the nervous system and cognitive health. Lastly, the third goal was to investigate if there were significant differences in altered brain cortical activity between subjects experiencing only cognitive fog and those presenting cognitive fog associated with anosmia. In line with the hypothesis of neuroinvasion through the olfactory tract [28, 29], this study postulated that the group of subjects with cognitive fog associated with anosmia would exhibit more intense neurophysiological symptoms compared to the group without anosmia and to the control group.

Table 2 shows descriptive statistics, namely mean and standard deviation, related to P300 latency at 1-month and 8-month assessments in cognitive fog groups and the control group.

Preliminarily, the factor groups concerning the two clinical cognitive fog groups (with and without anosmia) does not show significant differences, neither with latency and amplitude parameters of P300 (respectively F(1, 18) = 0.04, p = 0.95, $\eta {}^2$p = 0.002 and F(1, 18) = 0.07, p = 0.81, $\eta {}^2$p = 0.003) nor with the alpha, beta and theta band rhythm (respectively F(1, 18) = 1.6, p = 0.88, $\eta {}^2$p = 0.082; F(1, 18) = 0.05, p = 0.82, $\eta {}^2$p = 0.003; F(1, 18) = 0.94, p = 0.88, $\eta {}^2$p = 0.05). These results indicate that the two clinical groups are almost equal and in subsequent analyses, the two clinical groups will be considered together.

After this, a MANOVA was applied considering as an independent variable the two groups together (cognitive fog group) and the control group. Table 3 presents descriptive statistics, namely mean and standard deviation, related to beta band rhythms at 1-month and 8-month assessments in cognitive fog groups.

As shown in Fig. 1, the group with cognitive fog had significantly higher P300 latencies than healthy controls (main effect of groups: F(1, 39) = 16.11, p 0.01, $\eta {}^2$p = 0.292). There was not a significant decrease in P300 latency between one and eight month observations (main effect of follow-up phase F(1, 39) = 1.68, p = 0.23, $\eta {}^2$p = 0.041), and there was not a significant interaction between group x phase interaction F(1, 39) = 0.03, p = 0.858, $\eta {}^2$p = 0.0007). These data indicate that both groups are stable in the 8-month assessment compared to the 1-month assessment, meaning that clinical group maintains slowness and control group maintains normal P300 latency.

As shown in Fig. 2, the group does not exhibit a significant effect (F(1, 39) = 1.08, p = 0.27, $\eta {}^2$p = 0.026). Similarly, the phase shows no significant differences (F(1, 39) = 3.02, p = 0.91, $\eta {}^2$p = 0.072), and there was no significant interaction between group and phase (F(1, 39) = 0.03, p = 0.88, $\eta {}^2$p = 0.0007). As shown in Fig. 3, these data indicate that P300 amplitude is not significantly influenced by cognitive fog.

As shown in Fig. 4, the group with cognitive fog had significantly lower levels of alpha band rhythms than healthy controls (main effect of groups: F(1, 39) = 15.02, p 0.01, $\eta {}^2$p = 0.278). There was not a significant increase between one- and eight-month observations (main effect of follow-up phase, F(1, 39) = 0.29, p = 0.31, $\eta {}^2$p = 0.007) and there was not a significant interaction between group x phase interaction F(1, 39) = 2.01, p = 0.31, $\eta {}^2$p = 0.049. These data indicate that both groups are stable in the 8-month assessment compared to the 1-month assessment, meaning that clinical group maintains low alpha band rhythms, and control group maintains normal ones.

As shown in Fig. 5, the group with cognitive fog had significantly lower levels of beta band rhythms than healthy controls (main effect of groups: F(1, 39) = 11.15, p 0.01, $\eta {}^2$p = 0.222). There was not a significant increase between one- and eight-month observations (main effect of follow-up phase, F(1, 39) = 1.2, p = 0.27, $\eta {}^2$p = 0.029) and there was not a significant interaction between group x phase interaction F(1, 39) = 0.48, p = 0.276, $\eta {}^2$p = 0.012). These data indicate that both groups are stable in the 8-month assessment compared to the 1-month assessment, meaning that clinical group maintains low beta band rhythms, and control group maintains normal ones.

As shown in Fig. 6, the group with cognitive fog had significantly higher levels of theta band rhythms than healthy controls (main effect of groups: F(1, 39) = 5.80, p 0.01, $\eta {}^2$p = 0.129). There was not a significant decrease between one- and eight-month observations (main effect of follow-up phase, F(1, 39) = 3.68, p = 0.31, $\eta {}^2$p = 0.086) and there was not a significant interaction between group x phase interaction F(1, 39) = 0.98, p = 0.17, $\eta {}^2$p = 0.024). These data indicate that both groups are stable in the 8-month assessment compared to the 1-month assessment, meaning that clinical group maintains high theta band rhythms, and control group maintains normal ones. Fig. 7 shows comparisons of the absolute spectral power in central electrodes between the cognitive fog and control groups.

The overweight participants (BMI between 25 and 29.9), representing 30% of the sample, exhibit more severe neurophysiological symptoms compared to the normal-weight participants (BMI between 18.5 and 24.9). In more detail, patient n.2: BMI = 27.3; P300 Latency = 318 ms; P300 Amplitude = 3.95 µV. Patient n.11: BMI = 26.9; P300 Latency = 317 ms; P300 Amplitude = 5.21 µV. Patient n.13: BMI = 28.1; P300 Latency = 319 ms; P300 Amplitude = 4.89 µV. Patient n.16: BMI = 27.8; P300 Latency = 319 ms; P300 Amplitude = 4.83 µV. Patient n.19: BMI = 25.5; P300 Latency = 316 ms; P300 Amplitude = 3.67 µV. Patient n.20: BMI = 28.6; P300 Latency = 319 ms; P300 Amplitude = 4.56 µV. While statistical analyses were not conducted on this specific subgroup due to sample size limitations, these results serve as a preliminary observation that could suggest an association between higher BMI and the severity of neurophysiological symptoms. Further research with a larger sample size would be necessary to validate and confirm this potential relationship.

In this study, the effects of a previous Covid-19 infection on brain cortical activity were evaluated in a sample of patients with cognitive fog symptoms in the medium and long term. Specifically, the main objective of this research was to investigate whether the cognitive symptom reported by patients as cognitive fog correlated with actual alterations in neurophysiological parameters of P300 and alpha, beta, and theta bands.

The results of this study are consistent with previous knowledge regarding the associations between P300 parameters and patient-reported symptoms. The role of P300 as a neurophysiological correlate of cognitive functions has been previously reported in both normal samples [50, 51] and clinical groups [52]. In the present study, this correlation was also confirmed in cases of cognitive fog, showing a longer latency and reduced amplitude of the P300 compared to the control group. This data suggests a slowdown in the brain’s response to significant stimuli, indicating possible impairment or deterioration of cognitive functions. The prolonged latency of P300 may reflect reduced efficiency in information processing or a diminished capacity of the central nervous system to promptly detect and respond to stimuli. This slowdown could negatively impact the ability to process information, maintain attention, and make decisions, thus contributing to cognitive fog symptoms.

Further changes were observed in the rhythms of alpha, beta, and theta brain waves in the cognitive fog group compared to the control group, suggesting an association with cognitive symptoms. Alpha waves are generally associated with a state of relaxation and a decrease in brain activity in the absence of external stimuli [53, 54]. Variations in alpha waves could indicate impairment in the ability to relax or concentrate, both of which are fundamental for cognitive functions. In our results, we observed a significant decrease in alpha waves in some patients with cognitive fog. This could suggest difficulty in achieving a relaxed mental state or a decrease in attention, both of which influence cognitive performance. Moreover, beta rhythms, known for their association with sensory information integration and motor activity execution [55], were reduced in the cognitive fog group compared to the control group. This reduction could indicate impairment in the brain’s ability to effectively integrate environmental information and translate it into appropriate motor actions. This result suggests that individuals with cognitive fog may have difficulty processing sensory information and coordinating motor activities efficiently, with possible impacts on various cognitive domains such as attention, memory, concentration, and decision-making skills [36, 37]. Regarding theta waves, we observed an increase in activity in the cognitive fog group. Theta waves are commonly associated with memory and decision-making processes [56, 57]. An increase in theta wave activity could indicate greater brain effort in attempting to compensate for the decrease in cognitive functions. This increase may reflect brain adaptation to recover compromised cognitive functions from previous Covid-19 infection. However, this additional effort could also lead to increased mental fatigue and a reduction in cognitive resources available for other activities. The second objective of the study was to investigate whether any alterations in P300 parameters and beta band rhythms persisted over time. From the results of this research, it emerged that the problems of slowness in P300 and the alterations in beta band rhythms in individuals affected by cognitive fog persist even after recovery from Covid-19. Although slight improvements were observed eight months after recovery, the neurophysiological parameter alterations related to cognitive fog remained present. This suggests that, despite the disappearance of acute symptoms of the disease, cognitive functioning connected to these neurophysiological indices may remain compromised [58, 59, 60].

Finally, the third objective was to examine whether there were significant differences in the alteration of brain cortical activity between subjects who presented only cognitive fog and those who presented cognitive fog associated with anosmia. Specifically, in line with the hypothesis of neuroinvasion through the olfactory tract [28, 29], we expected that the group of subjects with cognitive fog associated with anosmia would exhibit more intense cognitive symptoms. Contrary to this expectation, the results of this study showed that the values of P300 and beta band rhythms in subjects with only cognitive fog and those with cognitive fog associated with anosmia did not show significant differences. This does not exclude the possibility that cognitive fog may be due to virus neuroinvasion through other pathways such as the vagus nerve or trigeminal nerve [27].

The key findings of this research have revealed three important discoveries regarding cognitive fog symptoms due to Long-Covid: (a) a significant connection has been found between cognitive fog symptoms occurring after Long-Covid and neurophysiological alterations, particularly in the brain waves of P300 and beta frequency bands; (b) the identified neurophysiological alterations appear to persist in the long term, suggesting that the effects of Long-Covid on brain function may endure over time; (c) the presence or absence of anosmia does not seem to be significantly correlated with the observed neurophysiological alterations. This suggests that anosmia may not be a determining factor in the neurophysiological consequences of Long-Covid.

This study provides a significant contribution as it adds crucial knowledge about the relationship between cognitive fog and neurophysiological alterations in the context of Long-Covid. Understanding this connection could open new perspectives in the diagnosis and management of health-related cognitive impairments [61, 62, 63]. Moreover, recognizing the persistence of neurophysiological alterations in the long term emphasizes the importance of carefully assessing patients with Long-Covid to prevent and address potential cognitive deficits [64, 65, 66]. The main strengths of this research can be identified in its longitudinal nature, which allowed for the observation of symptoms and neurophysiological alterations over time, providing an in-depth perspective on the persistence of cognitive fog due to Long-Covid. Additionally, the inclusion of a comparison group with and without anosmia further allowed exploration of the correlation between this condition and cognitive fog, contributing to the understanding of the complex manifestations of Long-Covid. These combined aspects make this study a significant contribution to the scientific literature regarding the long-term effects of Covid-19 on the brain and provide valuable foundations for the development of targeted clinical interventions to support patients affected by Long-Covid.

However, there are also some limitations to consider. For instance, the study sample might be limited in terms of age, potentially leading to underrepresentation of specific demographic groups. This aspect should be considered in future studies to ensure greater representativeness of the results. Additionally, the relationship between cognitive fog and neurophysiological components requires further investigation and exploration. The absence of pre-infection P300 data does not exclude the possibility that EEG alterations may serve as a predisposing factor for cognitive fog, rather than representing a response to the infection itself. The data should be considered cautiously because the group of healthy subjects was drawn from a database within the hospital, but we are not aware of whether they contracted the virus or not. This leads us to conclude that all observed differences were likely to be attributed to the presence of COVID-19 along with cognitive fog. Essentially, it is not clear which attribution we can solely attribute to cognitive fog. The limitation is related to the absence of pre-infection data; for this reason, we cannot establish a causal link with absolute certainty due to the absence of pre-infection data. Future research should directly compare individuals with COVID-19-related brain fog to those without, collecting complete pre-infection data to better understand the virus’s role in neurocognitive symptoms.

It is important to note that in the current study, measures of the severity of symptoms related to cognitive fog were not available. In future investigations, it may be beneficial to consider incorporating quantitative measures of symptomatology to provide a more comprehensive and detailed assessment of the extent of the disorder. While connections between these phenomena have been observed, the exact mechanism of brain function slowing and its link with mood and other factors are not yet fully understood. Therefore, it is essential to encourage further multidisciplinary and longitudinal studies on patients affected by Long-Covid and other relevant conditions. It is important to acknowledge that the manifestation of cognitive fog may vary depending on different conditions and variables involved, which could influence its relationship with neurophysiological parameters depending on the underlying cause. Furthermore, confounding factors such as depression and chronic fatigue syndrome could independently influence brain function, emphasizing the need for more in-depth investigations into the complex interaction among these variables.

In conclusion, further research is needed to achieve a comprehensive understanding of this complex phenomenon. Future studies should include a larger sample involving a wide range of ages and other demographic characteristics to obtain more representative and generalizable results. Adopting a multidisciplinary approach would be desirable, considering psychosocial factors such as mood, stress, and quality of life.

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Material preparation, data collection and analysis were performed by AG, RS & RAF. The first draft of the manuscript was written by RAF & RS. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

Approval was obtained from the ethics committee of the Madonna della Consolazione Polyclinic Nursing Home (protocol code: 2022-269). The procedures used in this study adhere to the tenets of the Declaration of Helsinki. Informed consent was obtained from all individual participants included in the study.

Not applicable.

This research received no external funding.

The authors declare no conflict of interest.