Neurofeedback training can improve self-esteem in those with major depressive disorder


A study published in the journal Neuroimage: Clinical found that patients with Major Depressive Disorder (MDD), who had recovered from symptoms, were able to strengthen some of their brain connections whilst thinking about guilt-evoking memories, thereby increasing their self-esteem.

The research showed that connectivity between certain brain regions – previously found to be decreased when feeling guilt in people with a history of depression – could be strengthened in a single session of neurofeedback training through functional magnetic resonance imaging (fMRI), captured before and after the procedure.

The study was conducted by the Brazilian organizations D’Or Institute for Research and Education (IDOR), Federal University of Rio de Janeiro (UFRJ) and Federal University of ABC, in collaboration with King’s College London, in the United Kingdom.

Major Depressive Disorder (MDD), also known as depression, is a mental disorder caused by a set of social, psychological and biological factors.

Its symptoms are characterized by the continuous loss of interest and pleasure in daily life and the prevalence of negative feelings such as low mood, self-blame, and low self-esteem. According to data from the World Health Organization (WHO), last year, depression already affected more than 300 million people in the world, becoming the most disabling disease of contemporary times. Considering the seriousness of the problem in global public health, the medical and scientific communities are increasingly seeking to understand the depressive disorder, aiming at the development of new therapies and improvement of patients’ quality of life.

Studying MDD through fMRI – a technique that allows researchers to analyze brain structure and function in a noninvasive way-, the recently published paper was based on the scientific finding that people with depression, even when recovered from symptoms, showed less connectivity between two specific brain areas while experiencing feelings of guilt: the right anterior superior temporal (ATL) and the anterior subgenual cingulate (SCC). By connectivity, the study refers to the exchange of information between these structures, as they are directly linked to the interpretation of social interactions.

Based on this “neural signature” on patients’ brains, the study tested the possibility of strengthening these connections through neurofeedback, a program that allows participants to observe and modify their brain activities in real-time.

Although at the early stages, the result was quite remarkable: in just one training session, participants already demonstrated a stronger connection between the mentioned areas and reported an increase in self-esteem after the neurofeedback interaction.

How the research was done?

The study’s first author, Dr. Roland Zahn at King’s College London, explains why the study was carried out in people who had recovered from symptoms:

“The brain signature of excessive self-blame was discovered in patients with Major Depressive Disorder whose symptoms had remitted, suggesting it could precede the symptoms of depression, making people more vulnerable to the disorder. Secondly, for safety reasons we wanted to make sure people’s depression wouldn’t get worse after the treatment, and people with remitted MDD are less at risk to worsen significantly than people with current symptoms”.

Based on this “neural signature” on patients’ brains, the study tested the possibility of strengthening these connections through neurofeedback, a program that allows participants to observe and modify their brain activities in real-time.

To conduct the research, the 28 participants with remitted depressive symptoms were randomly divided into two groups.

One group was exposed to a control neurofeedback exercise where they had to maintain the same level of their brain connections, while the other group was instructed to increase these connections during training.

This procedure was accomplished through visual feedback on a screen that indicated if people were doing the brain exercise in the instructed way.

“The participants had to imagine a specific memory from their past which made them feel guilt or indignation towards others.

On the screen, they had to change the way they felt about this so that the color display that reflected their brain connections would also change.

The marker was a thermometer that, when filled to the top, would be a signal that the participants were doing well in the training”, explains the first author.

Although neurofeedback exposure time was the same in both groups, in the fMRI results, participants who were instructed to increase activity on their brain wirings showed strengthening in the connectivity between the exercised areas. At the same time, it was observed an increase in their self-esteem that wasn’t found in the group that kept their connections at the same initial level, results that proved the effectivity of the training.

The study also required the development of a specific neurofeedback software, the “Functional Real-Time Interactive Endogenous Neuromodulation and Decoding”, or simply, FRIEND. Dr. Jorge Moll, a neuroscientist at IDOR and corresponding author of the study, led the group who created the program.

“FRIEND is a toolbox developed for any kind of neurofeedback study using fMRI.

The current implementation geared towards this aspect of MDD pathophysiology, but other designs, cognitive states, emotions, and patient populations can also be targeted in future researches”, he explained.

At IDOR, Dr. Moll has conducted other studies involving the induction of brain wiring changes through neurofeedback training.

For him, fMRI is one of the most powerful tools for noninvasively analyzing brain functions, but clinical applications related to neurofeedback are still at an early stage. “There is a road ahead. We need to establish the pathophysiology, side effects, therapeutic effects, and ideal target patient populations, as well as cost-effectiveness. Despite fMRI time being expensive, it is not much more than other treatments, and this can potentially offer an alternative for patients who are poor responders to conventional therapies. Yet, lots of fundamental research needs to be done, but potential clinical applications are starting to emerge.”

Intending to allow further research development on the area, Dr. Jorge Moll informs that the neurofeedback software, FRIEND, is available online for free, accessible to any other interested researchers.

Its use is not limited to the current study scope and can be applied more widely in neurofeedback research.

Authors inform that this research is the first step in developing a novel treatment for recurrent depression, but it was not intended to prove the efficacy of this approach, which will need to be investigated in future larger studies with longer follow-up observations.

Stroke is a debilitating neurological condition caused by interruptions in the blood supply to the brain. One in six people worldwide will suffer from a stroke throughout their lives. Being the main cause of adult disability, stroke usually results in the changes in electrical brain activity and the impairment of sensation, motor or cognition functions (1). In this context, as a non-invasive and relatively low cost technique that contributes to neuroplasticity, EEG neurofeedback training (NFT) might be a promising tool for stroke rehabilitation. By addressing excesses and deficits in particular EEG activities through real-time inhibition or augmentation training, NFT enables the patient to modulate their neuronal electrical activity and thus on cortical metabolism.

Following NFT, some stroke patients showed modest improvements in cognitive functions such as attention, memory, concentration, reading, and coordination speech (2). For instance, following upper alpha NFT, two single chronic stroke patients with memory deficits showed memory improvement meanwhile cortical “normalization” was found in a stroke patient with pathological brain activation patterns (3). Compared to traditional rehabilitation, thirty sessions of sensorimotor rhythm (SMR, 12–15 Hz) or mid-beta (15–18 Hz) NFT was more effective to improve concentration and visual perception for the patient with hemi-paralysis from stroke within the previous 3 months to 1 year (4). In summary, previous studies have shown the potential of NFT for cognition rehabilitation in stroke survivors.

Ischaemic stroke is a result of blockage of a blood vessel supplying the brain. Attenuation of alpha power may occur following ischaemic stroke (5). Alpha power plays important roles in not only cognitive functions but also physiological conditions. It has been found that alpha power had a relatively strong positive correlation with the regional cerebral blood flow (rCBF) (5). Moreover, alpha power enhancement was associated with improvements in motor functions and activities in daily living (67).

Although NFT has demonstrated benefits on stroke rehabilitation, whether it still works on the chronic stroke patients who had suffered from stroke with long period (above 3 years) and failed to gain further improvement using traditional therapies is not clear yet. Therefore, we applied NFT on two patients with chronic stroke, in order to examine whether they were able to increase their alpha activity by NFT and consequently gain improvement in cognition or motor ability.


Both participants finished the NFT without experiencing obvious adverse effects. For the across session effects, it was found that Participant B could learn to increase her IAB amplitude over sessions, which also led to enhancement in EO baseline. The same is no longer true for Participant A who had no obvious increase trend in either NFT or EO baseline across sessions. Similarly, regarding the within session effects, Participant A did not show obvious increase trend in IAB over 5 NFT blocks, whereas Participant B was much better than Participant A. Nevertheless, Participant A could learn to elevate his IAB in 9 NFT sessions compared to EO baseline, indicating that he could show higher IAB activity in NFT blocks than resting state within short term but the maintenance of such increase was not easy over longer periods. Similarly, Participant B also learnt to increase her IAB in most NFT sessions when compared with EO baseline. These results suggested that NFT yielded some plastic changes in brain activity, but such change had inter-individual difference.

The inter-individual differences in neurofeedback learning have also been found in previous work, regardless of the NFT protocol and subject population (1623). With regard to stroke patients, Kober et al. (24) which investigated the NFT effects on cognitive functions also found that not all patients could learn to enhance alpha amplitude within and across NFT sessions. In healthy population the ability to enhance alpha has been found to be positively related to resting alpha activity, i.e., the individual who has higher resting alpha activity is more likely to gain successful enhancement of alpha by NFT (25). Such relation might be generalized to the patients with chronic stroke, as Participant B who had much higher EO alpha than Participant A (1.42 in Participant B and 0.98 in Participant A before the first session) showed much larger alpha increase within sessions and across sessions.

Following alpha NFT, the two participants have slight improvement in their emotional well-being. The levels of anxiety and depression dropped to a more relaxed behavior. For the adapted tests, a slight evolution in speech performance was found in Participant A who had symptoms of conduction aphasia. Likewise, improved speed of the march was observed in Participant B. The NFT sessions were performed at patients’ home, and there was no change of their exercise routine. Furthermore, only NFT was performed without other therapy during NFT intervention. Thus, it is speculated that the improvement was only due to NFT. Furthermore, it has been shown that in the course of recovery within 6 months after stroke, alpha increase is associated with improvement in motor performance and activities of daily living (6). Our results suggested that with the help of NFT, the chronic stroke patients were able to enhance their alpha activity, and achieve improvement on rehabilitation, even though the time from stroke onset was rather long (at least above 3 years) and the patients cannot be further improved using traditional therapy.

Prior case reports also showed positive effects of NFT on stroke rehabilitation but with different NFT protocols. For instance, Rozelle and Budzynski (26) employed beta1/theta NFT with the purpose of beta1 enhancement and theta reduction on a male stroke patient. After NFT, the patient reduced his slow-wave activity, depression, anxiety, and tinnitus. Additionally, he improved his speech fluency, word finding, balance and coordination, attention, and concentration. Mroczkowska et al. (27) conducted SMR/theta NFT in order to enhance SMR and decrease theta on a female stroke patient with aphasia, and obtained positive results in terms of concentration, visual perception, categorizing, as well as the regulation of affect and reduction in the aphasia symptoms. While the present study suggested that alpha NFT also could contribute on stroke rehabilitation, especially in the patient with chronic stroke who could not gain further improvement from traditional therapy.


In conclusion, alpha NFT appeared to have induced some brain plasticity in chronic stroke patients, which was associated with improvement of emotional state, cognitive, and motor functions. It should be noted that even a slight improvement is promising, since the patients had no further improvement using traditional therapy. Our result suggested the potential of NFT for chronic stroke rehabilitation.

D’Or Institute for Research and Education
Media Contacts:
Maria Eduarda Abreu – D’Or Institute for Research and Education
Image Source:
The image is in the public domain.

Original Research: Open access
“Blame-rebalance fMRI neurofeedback in major depressive disorder: A randomised proof-of-concept trial”. Roland Zahn et al.
NeuroImage: Clinical doi:10.1016/j.nicl.2019.101992.


Please enter your comment!
Please enter your name here

Questo sito usa Akismet per ridurre lo spam. Scopri come i tuoi dati vengono elaborati.