One of the most commonly used methods in QEEG is frequency domain analysis, which is based on the fast Fourier transform. It allows for the assessment of EEG signal power within specific frequency bands, which reflect the activity of various neural circuits, particularly thalamocortical and cortical networks (Smailovic and Jelic, 2019).

Studies conducted over the past 10 years using traditional electroencephalography in patients with depression indicate significant methodological issues that hinder the comparability of results. The main limitations include the diversity of diagnostic criteria within study groups, the lack of appropriate control groups, and substantial variability in EEG techniques used across studies (Thoduparambil et al., 2020; Pollock, 1990). Despite these limitations, increasing attention is being given to the identification of specific biomarkers that could aid in earlier diagnosis.

Soares de Aguiar Neto and Garcia Rosa emphasize that alpha waves in EEG recordings have significant diagnostic relevance in the context of depressive disorders (de Aguiar Neto and Rosa, 2019). Alpha waves occur in the frequency range of 8–12 Hz and are observed with eyes closed, in a state of relaxation and wakefulness. They most commonly appear in the occipital region. In individuals suffering from depression, researchers observe alpha wave asymmetry in the frontal part of the brain (van der Vinne et al., 2019). Kana et al. identified differences in alpha waves between a group of healthy individuals and a group of individuals with depression. It was found that alpha waves in the MDD group were lower compared to the control group. Additionally, these abnormalities were observed in the frontal, parietal, occipital, and temporal regions (Kan and Lee, 2015). An international team of researchers specializing in neurobiology and psychiatry, while analyzing the role of alpha waves in major neuropsychiatric disorders, recorded reduced activity of these waves in depression, particularly in the prefrontal cortex. These findings suggest that such a wave pattern may lead to difficulties in emotion regulation and impaired cognitive functioning (Ippolito et al., 2022). According to some studies, qEEG can be used to predict the effectiveness of pharmacological treatment in MDD. Female patients who do not respond to treatment are characterized by greater alpha wave power in the left frontal lobe compared to the right (Ip et al., 2021). Similarly, in the study by Arikan et al., it was shown that patients with MDD who respond to treatment exhibit significant changes in cortical activity in the theta, alpha, and high beta bands, with qEEG improvement ranging from 15 to 67% (Arikan et al., 2025). In turn, treatment with citalopram, a selective serotonin reuptake inhibitor (SSRI), results in an increase in alpha power (Pariante et al., 2012).

In a study involving 1,008 patients with MDD and 336 individuals in the control group, traditional EEG was performed, and it was observed that patients with MDD had elevated theta values in the frontal cortex. The authors concluded that increased theta waves in the frontal region were associated with a lack of response to treatment. On the other hand, other researchers noted that low theta wave values in the frontal cortex occur in individuals who respond poorly to antidepressant therapy (Arns et al., 2015). Theta waves fall within the frequency range of 4–8 Hz and are most commonly recorded during drowsiness, relaxation, or meditation (Kopańska et al., 2023). In MDD, a significant increase in spectral power in the theta frequency bands can be observed in the parietal and occipital regions, both with eyes open and closed (Grin-Yatsenko et al., 2009). Recent studies suggest a correlation between depression and pathological theta wave values. In their study, Melinda L. Morgan et al. examined 104 individuals suffering from depression using quantitative electroencephalography. The obtained results were compared with a control group. The authors noted that theta wave correlation in the prefrontal regions may serve as a biomarker of depression, as it is associated with the pathophysiology of the disorder and with predicting treatment response. Interestingly, absolute prefrontal power was higher in women than in men (Morgan et al., 2005).

Suzuki et al. conducted a quantitative EEG analysis on three groups of participants: 27 patients with melancholic depression, 21 patients in remission, and 17 individuals without mental disorders. The latter two groups served as the control group. After completing the study, the authors observed that individuals suffering from depression were characterized by higher theta power in the bilateral frontal regions compared to the remission group. Moreover, they noted that individuals who had experienced depression showed lower theta and alpha power in the bilateral frontal regions and lower alpha-1 power in the parietal regions compared to the control group, which may indicate specific neural dysfunctions related to emotional regulation and information processing in this population (Suzuki et al., 1996). Other authors also point out that patients with MDD are characterized by anomalies in the theta frequency range. In patients with depression, lower theta activity and faster activity in the frontal areas of both hemispheres and the posterior part of the right hemisphere can be observed compared to the control group (Begić et al., 2011; Ohashi, 1994; Yamada et al., 1995; Fingelkurts et al., 2007). Interestingly, patients with MDD show elevated theta wave values in frontal regions (Suzuki et al., 1996).

Begić et al. observed increased beta wave activity in patients with depression (Begić et al., 2011). Therefore, Cai H. et al. conducted an EEG study in patients with MDD, focusing, among other things, on beta wave analysis. The authors noted that beta wave activity, which is associated with mental activity and focus, may have atypical patterns in patients with depression, indicating disturbances in cognitive activity (Cai et al., 2018). In the study of Cai et al., the concept of simplified three channel EEG collection covering frontal area has been implemented. This is in concordance with canonical prefrontal EEG asymmetry pattern as the hallmark of affective disorders, including depression (Davidson, 1998). The anatomical attribution of beta-wave relevant changes may be however more complex, including not only frontal, but also temporoparietal sites, including their bilateral interconnectivity (Li et al., 2017).

An intriguing finding is that Internet addiction combined with depression is associated with specific brainwave patterns. Studies have shown that these patients exhibit increased relative theta power and decreased alpha power across all brain regions (Lee et al., 2014). As early as the end of the 20th century, researchers noticed a connection between the occurring traditional EEG abnormalities in the delta wave range in people suffering from depression. Kwon et al. showed that delta wave activity is increased in people with MDD and that the dominance of these waves is more common in the right hemisphere (Kwon et al., 1996). Depression patients are characterized by a higher delta and theta spectral power in the left superior temporal gyrus at the level of Brodmann areas. Increased activity of these waves may reflect difficulties in concentration, reduced ability to process emotional stimuli and problems with mood regulation (Spironelli et al., 2021). In contrast, a study of 60 menopausal patients diagnosed with depression observed anomalies in EEG activity. These included an increase in the relative values of delta/theta and beta waves, a decrease in alpha band activity, and a slowing of the center of gravity of delta/theta wave activity (Saletu et al., 2010). An interesting example of EEG research is the analysis of the effect of music therapy on functional brain connectivity in patients with MDD. In a study involving 8 patients with depression and 8 control subjects, a greater delay in the delta wave phase was observed in the patients with depression, which may indicate characteristic changes in brain activity in this group (Simmatis et al., 2023). It is worth noting that lower delta wave activity has also been identified as a potential indicator of susceptibility to depressive episodes in people suffering from this disease (Buysse et al., 1997). An EEG study conducted during sleep in hospitalized patients diagnosed with depression who were not taking medication showed that both low and high delta sleep index values are characteristic of people suffering from depression. This could potentially be a useful diagnostic marker in identifying this disorder (Nissen et al., 2001).

According to neurobiological research, a reduced level of gamma waves is associated with depression, and people suffering from this disorder often show poorer concentration than healthy people. Therefore, gamma waves can significantly affect the effectiveness of classification processes (Jiang et al., 2021). Gamma brain wave activity will help diagnose depression more than other frequency bands (Malik et al., 2018). Researchers at the University of Wisconsin-Madison examined 17 patients using a 128-channel EEG. After conducting the study, they noticed that people suffering from depression are characterized by a significantly reduced gamma frequency density (Pizzagalli et al., 2006). Numerous studies indicate that gamma rhythms in people with unipolar depression differ from those observed in bipolar disorder, including episodes of bipolar depression (Isomura et al., 2016; Liu et al., 2012; Lee et al., 2010). Listening to an instrumental piece of music by 19 people with diagnosed unipolar depression caused an increase in gamma wave activity in the prefrontal area. This result suggests that such an intervention can modulate cortical activity, which potentially indicates the therapeutic effect of music on brain function in depression (Mosabbir et al., 2022). The analysis of gamma and delta bands shows increased gamma activity in patients with MDD, especially in response to emotional stimuli. Other authors, on the other hand, observed a reduced resting gamma density in people with depression, which may indicate a reduced activity of tonic processes in the anterior regions of the cingulate gyrus (Akdemir Akar et al., 2015). Individuals with MDD exhibited significantly increased current density in the delta, theta, alpha, beta1, and beta2 frequency bands compared to control subjects in the anterior cingulate cortex and prefrontal cortex (Korb et al., 2008). Roemer et al., on the other hand, used traditional EEG to examine elderly people diagnosed with depression. The patients were observed to have lower delta values and higher theta wave values, which has also been noted by other researchers, as shown in Table 1 (Roemer et al., 1992).

One of the frequently studied anomalies in MDD using EEG is asymmetry in brain activity, particularly in the alpha wave range. The researchers suggests that the relative difference between the left and right frontal brain regions in alpha wave activity may be a predictor of depression vulnerability. Withdrawal and negative behaviors are often related to high levels of right cortical activity, while positive reactions and mood have been related to increased left cortical activity. Therefore, frontal alpha asymmetry (FAA), with greater right-sided frontal activity, may be an important biomarker in diagnosing depression and anxiety (Barros et al., 2022). Alessandra Monni et al. analyzed a latent factorial approach to measure FAA. The research was conducted among 139 non-clinical participants. The authors distinguished a frontal alpha asymmetry factor (FAAf) and a parietal factor (PAAf) subjecting all asymmetry indices to a varimax-rotated, principal component analysis. Next the researchers explored among others the associations of latent factor and raw FAA scores with symptoms of depression and anxiety to determine which correlations were driven by FAA after variance from parietal activity was removed. Next, after correcting for false discovery rate, it turned out that only FAAf at the low alpha band was negatively correlated with depression symptoms (a latent CES-D factor) and significantly diverged from PAAf's association with depressive disorder. Therefore, the latent factor approach indicates beneficial effects for isolating functionally distinct resting-state EEG signatures (Monni et al., 2022).

The studies conducted by Ambrish Dharmadhikari et al. among 24 participants with Mild Depression indicates that results by measuring of FAA at resting stage was inconsistent. Researchers suggesting that it is necessary to revisit our approach from conventional search of diagnostic marker. Frontal Alpha Asymmetry might reflect component of depression but not the syndrome depressive disorder (Dharmadhikari et al., 2019). In turn, the meta-analysis by Luo et al., which includes 23 studies involving over 1,900 patients with MDD, showed that frontal alpha asymmetry measured with traditional EEG has a limited but statistically significant diagnostic value, which may support the clinical assessment of depression (Luo et al., 2025). It should be noted that frontal alpha asymmetry recorded during emotional stimulus processing differentiates individuals with MDD from healthy subjects more effectively than measurements taken at rest (Stewart et al., 2014; Périard et al., 2024). Arns et al. also investigated whether alpha wave asymmetry in the occipital and frontal regions of the brain could distinguish between outpatients with MDD and healthy individuals, assess the predictive power for antidepressant treatment outcomes, and account for the influence of gender on these relationships. The results did not show significant differences in alpha wave levels in the occipital and frontal cortex. However, for FAA, a specific interaction effect between gender and the type of medication used was observed. The findings suggest that future research on EEG alpha markers in depression should stratify participants by gender (Arns et al., 2016). Many researchers indicate a link between depression and EEG asymmetry, especially in the alpha and theta wave ranges in the frontal regions. In a study aimed at comparing the occurrence of regional brain asymmetries in severe depression with or without anxiety disorders, traditional EEG analysis of 44 patients revealed alpha asymmetry in individuals with MDD, indicating lower activation in the right posterior region compared to the left posterior region (Bruder et al., 1997). Another study showed that EEG spectral asymmetry (SA)—based on differences in power across frequency bands and their sources—differentiated women with depression from healthy individuals. This suggests that SA may serve as a useful marker of treatment outcome specifically in female patients (Hinrikus et al., 2010).

Abnormalities in traditional EEG activity are associated with various mental disorders, including depression, suicide and aggression. An EEG study conducted by Graae F. et al. showed differences in alpha wave asymmetry compared to the control group. Alpha asymmetry in the posterior regions of the brain was associated with suicidal ideation, but not with the severity of depression. The results suggest reduced activation of the left posterior brain as a factor associated with suicidal or aggressive behavior (Graae et al., 1996). Also, a reduction in the power of alpha wave activity in the left frontal lobe, which plays a key role in coping with emotional stress, may contribute to suicidal tendencies (Rasouli et al., 2024). This was also noted by Roh et al., who observed a reduced alpha wave power in the left frontal lobe in patients with MDD with suicidal thoughts compared to patients with MDD without such thoughts. Therefore, it can be seen that suicidal thoughts are an important moderator of alpha wave asymmetry in the frontal part of the brain in patients with MDD, as shown in Table 2 (Roh et al., 2020).

Functional connectivity refers to the statistical relationships between EEG signals recorded from different brain regions, allowing for the assessment of the degree of their mutual synchronization and functional integration (Smailovic and Jelic, 2019).

Studies indicate that patients with depression may exhibit abnormalities in EEG signal coherence, particularly in high-frequency bands. In one experiment, Li et al. compared a group of healthy individuals and patients with MDD during an emotional face recognition task. The patient group showed significantly higher gamma-band coherence compared to healthy participants (Li et al., 2015). Similar results were observed in a study analyzing activity in the 35–45 Hz range among healthy individuals, those with depression, and those with schizophrenia—individuals with depression demonstrated increased gamma wave activity throughout the emotional task (Siegle et al., 2010). In addition to task-related activity, resting-state EEG also provides valuable diagnostic insights. In a study by Sun et al., the Phase Lag Index (PLI) was used to assess resting-state functional connectivity. The results showed that patients with depression had significantly disrupted intrahemispheric connectivity, particularly in the left hemisphere. The application of PLI achieved a classification accuracy of 82.3%, indicating its potential as a biomarker for MDD (Sun et al., 2020). Changes in connectivity are also observed during pharmacological treatment. Studies have shown that MDD patients who respond poorly to SSRI treatment exhibit stronger connectivity in the right frontotemporal network within the delta and theta bands (Lee et al., 2011). On the other hand, serotonin-norepinephrine reuptake inhibitors (SNRIs) tend to reduce theta-band coherence (Bares et al., 2008; Cook et al., 2002; Bares et al., 2015). In a review of studies, Armitage highlighted that individuals with depression show reduced interhemispheric coherence, particularly during sleep. The authors point out that computer-assisted EEG analysis can detect subtle changes in neural communication patterns that may underlie depressive symptoms and treatment outcomes, as shown in Table 3 (Armitage, 1995).

Source mapping of the brain's electrical activity responsible for the distribution of potentials on the scalp allows for the mathematical solution of the EEG inverse problem. One of the popular methods used is LORETA—a functional imaging technique that estimates the cortical sources of EEG signals in a three-dimensional brain model (Smailovic and Jelic, 2019). In depression, patients show reduced source current density in the left hemisphere. Differences in brain volume distribution are observed in the delta band, which is characterized by increased source current density, suggesting specific alterations in brain activity in this patient group (Flor-Henry et al., 2004). Bachmann and Lass identified EEG-based biomarkers of depression using a combination of two analysis methods: the linear Spectral Asymmetry Index (SASI) and the nonlinear Detrended Fluctuation Analysis (DFA). Results showed the most significant differences in SASI values in channel Pz and DFA values in channels Pz and O₂, indicating that analyzing signals from a single parietal channel using these parameters may enable high classification accuracy between healthy individuals and those with depression (Bachmann et al., 2017). Additionally, patients with MDD exhibit higher Lempel-Ziv complexity (LZC) and lower power spectral density (PSD). The highest classification accuracy (up to 92.4%) was observed in the frontal, temporal, and central sources (Yang et al., 2023).

One study involving 36 participants analyzed brain activity changes before and after 12 weeks of antidepressant pharmacotherapy. It demonstrated that responders had lower multiscale entropy (MSE) at small scales and higher MSE at large scales, particularly in the fronto-central region (Jaworska et al., 2017). Korb et al. examined 74 individuals using 36-channel EEG and found significantly increased values in the delta, theta, alpha, beta1, and beta2 frequency bands in the anterior cingulate cortex and prefrontal cortex of patients with depression compared to the control group (Korb et al., 2008). In another study, Li et al. examined individuals with MDD using Event-Related Potential (ERP) measurements at frontotemporal sites. The results suggested that depressed participants had lower P300 amplitudes and significant differences in both fast and slow neural responses in the frontal and parietal lobes (Li et al., 2023). Bilateral projections of frontal activity and right parietotemporal activity were observed in participants with depression and comorbid disorders (Mathersul et al., 2008). It is worth noting that the use of simple EEG methods in combination with machine learning may serve as an effective diagnostic aid for depression. Ahmadlou et al. observed that individuals suffering from depression exhibit elevated gamma wave values in the frontal region. Their study also used Higuchi's fractal dimension (HFD) and Katz's fractal dimension (KFD) analyses (Ahmadlou et al., 2012). In another study, a simple 3-channel EEG system combined with linear and non-linear feature analysis and machine learning algorithms enabled discrimination between depressed and healthy individuals with a maximum classification accuracy of 76.4%, offering a potentially effective and accessible diagnostic tool for depression (Cai et al., 2018). In the study by Cai Hanshu et al., an EEG with three electrodes was used, placed at the points Fp1, Fp2 and Fpz. The collected data was analyzed using machine-learning based soft computing techniques. This approach enables automated QEEG signal denoising and categorization and in the cited study K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Approximate Nearest Neighbor (ANN) and Deep Belief Network (DBN) classifying algorithms were used. The results showed that the DBN method outperformed traditional approaches using shallow algorithms. In addition, the researchers suggest that the absolute power of beta waves is one of the helpful indicators for detecting depression (Cai et al., 2016). Li et al. also noted after conducting an analysis that the average classification accuracy for the beta band was higher than for the alpha and theta bands in patients with depression which suggests that the beta band may be more suitable for machine-learning based detection of MDD features in QEEG than the other bands. However, this study was performed in individuals with mild form of depression, making its conclusion not necessarily applicable to MDD subjects (Li et al., 2019).

Traditional EEG studies conducted during depression treatment increasingly highlight characteristic changes in brain activity patterns influenced by different forms of pharmacotherapy. Analyzing these changes helps to better understand the neurophysiological mechanisms of antidepressant medications and identify potential markers of therapeutic efficacy (Leuchter et al., 2009a). Bruder et al. also observed that SSRI treatment resulted in increased alpha power among responders, particularly in the occipital region (Bruder et al., 2008). Detailed spatial EEG features and machine learning applications are presented in Table 4.

Depressive disorders are among the most commonly diagnosed mental disorders worldwide, but they remain difficult to diagnose accurately, especially in cases with a mild or atypical course. Current traditional diagnostic methods, which rely on clinical interviews and rating scales, are prone to errors due to subjective opinion. Therefore, they do not provide a complete picture of the neurobiological basis of depression. For this reason, increasing attention is now being paid to objective methods which include the analysis of bioelectrical brain activity by means of EEG and QEEG (Kopańska et al., 2023; Williams, 1954).

This article analyses the results of a study that looked for patterns of brain wave activity characteristic of patients with major depression. The analysis focuses on specific EEG bands which include alpha, beta, theta, delta and gamma waves. The most common phenomenon observed by other researchers was the asymmetry of alpha waves in the prefrontal regions (with dominant activity on the right side), which seems to be related to the severity of depressive symptoms and the presence of suicidal thoughts (Graae et al., 1996; Rasouli et al., 2024; Roh et al., 2020). In the case of waves, their increased activity has been shown in some studies to be a positive predictor of response to treatment, especially drug treatment (Grin-Yatsenko et al., 2009; Suzuki et al., 1996). In contrast, gamma waves, which are associated with the integration of cognitive and emotional information, tend to decrease in patients with MDD (Jiang et al., 2021; Malik et al., 2018; Pizzagalli et al., 2006; Akdemir Akar et al., 2015).

It is worth noting that the results of traditional EEG studies in depression are not entirely consistent (Arns et al., 2015). An additional complication is that many studies rely on small samples and different analytical techniques (including band power analysis, coherence, source activity mapping and non-linear indices). Despite these limitations, EEG/QEEG has not only diagnostic but also prognostic potential. Changes in specific wavebands prior to treatment can serve as predictors of treatment efficacy—for example, higher theta activity and lower gamma activity prior to SSRI treatment were associated with better response to pharmacotherapy (Leuchter et al., 2009a). Schiller noted that the use of QEEG in depressed patients may help to adjust pharmacotherapy, especially in treatment-resistant cases (Schiller, 2019). Additionally, the increasing importance of combining QEEG with advanced computational tools, such as machine learning algorithms or signal source localization methods, deserves special attention. These approaches not only improve the accuracy of classification between MDD patients and controls, but also open new perspectives for medicine in psychiatry (Cai et al., 2018; Ahmadlou et al., 2012). In conclusion, quantitative electroencephalography is a promising tool to aid the diagnosis of depression.