Introduction

The rhythmic and repetitive brain activity, which can be measured through EEG, is ranging from delta (0–4 Hz) to gamma (30–100 Hz). The study of brain oscillations is attracting substantial amount of scientific attention and is one of the fastest growing research areas in neuroscience. Oscillations represent a major mechanism of communication within the brain and have been consistently related to cognitive functions. An example of such an association is the link between frontal-midline (fm) theta oscillations and executive control. Power increases of fm-theta have been associated with enhanced cognitive processing and can predict successful behavioral performance.

Definition of neurofeedback: Neurofeedback is a technique, for dealing with brain-based functional disorders without the use of medication or invasive procedures, in which brain activity is recorded using electrodes and presented visually or audibly so that the patient can know the state of the function he or she is trying to control.

Mechanism of Action

The goal of neurofeedback is the self-regulation of endogenous neural oscillations. Neural parameters of ongoing neural activity are fed back to the participant on a trial-by-trial fashion to up- or downregulate one’s own brain activity. Thereby implementation of neurofeedback is realized by a software system and a processing pipeline consisting of five basic elements, including data acquisition, online data processing, online feature extraction, online feedback generation, and the learning participant.

Enhancement of Cognition by Neurofeedback

A considerable amount of literature reported associations of alpha oscillations with cognition. Alpha brain oscillations are associated with working memory, covert attention and behavioral performance. For example, upper alpha frequency training led to enhanced strategic and top-down processes as reflected in associative
memory, whereas training of the sensory motor rhythm (SMR, 13–15 Hz) led to enhanced performance in less-effortful and less-strategic memory task as reflected in improved item memory. The results in research provide support that the modulation of endogenous oscillations is possible by neurofeedback and that such self-regulation transfers to enhanced cognition.

Effects of Neurofeedback on Everyday Life Performance

Studies investigating the transfer of neurofeedback on everyday life performance show, for instance, that participants demonstrated improved surgical skills after learned self-regulation of SMR. Another study in the field of sports showed that self-regulation of SMR in neurofeedback transferred to enhanced SMR power during action preparation while golfing. In the domain of creativity in the arts, research shows that alpha-theta training increased performance of professional and novice musicians and increased dancing performance of professional dancers. 

Conceptualization of Self-control of Brain Activity

Control of brain activity during neurofeedback is more than merely learning to regulate the activity in one specific neural network that is targeted directly by neurofeedback. It is rather the result of conjugated labor of different brain networks tuned to optimize the control of the specific brain signals under training by means of feedback, thereby giving rise to different forms of brain plasticity.

Neuroplastic Effects of Neurofeedback

For neurofeedback, and in analogy to general learning, plasticity implies a progressive and long-term change - of at least >20–30 min - of a measure during or after training.

A collection of studies confirmed that the plasticity of oscillatory patterns may be Hebbian. Another body of research points to the existence of a complementary form of plasticity which is anti-Hebbian, or homeostatic. This appears to be the consequence of intrinsic regulatory mechanisms that prevent brain activities reaching extremes, such as pathologically high/low synaptic strengths or oscillatory states.

Results show that neurofeedback may lead to plastic changes in cortical regions responsible for cognitive control such as the anterior cingulate, associated with improvements in attention-deficit or on-task mind wandering. It may also impact white matter pathways, in addition to changes in gray matter volume. 

Interindividual Differences in Neurofeedback: Responders and Nonresponders

While review papers are full of examples of positive neurofeedback effects, only a few studies so far have investigated negative effects of neurofeedback systematically. One may distinguish at least four reasons for individual variability in the responsivity to neurofeedback:

1. Physical reasons (poor signal detection)

2. Physiological reasons (initial signal intensity of the brain feature)

3. Cognitive reasons (reserve capacity)

4. Metacognitive reasons (training instructions and strategies of self-regulation)

Specificity and Efficacy

What factors constitute a training that maximizes the pre- to post-changes in neural parameters and behavioral performance measures? The following factors could be considered:

• Since neurofeedback usually aims at the enhancement of a specific cognitive function, it seems straightforward to optimize those neural systems and processes that give rise to these cognitive processes.

• it is likely that induced plasticity closely relates to those neural mechanisms providing the underpinnings of cognitive processes in the first place.

• Increased neurofeedback specificity may be achieved by enriching standard EEG frequency features through spatial filters.

• The majority of studies seem to indicate that reliable training effects occur after about ten training sessions

Conclusion

By feeding back neural parameters of ongoing neural activity to the participants on a trial-by-trial fashion, self-regulation of brain activity can be achieved. Different brain networks might be engaged to adjust control over a brain signal during neurofeedback training. Regarding the physiological mechanism responsible for neurofeedback-induced plasticity, which might even impact brain morphology; two forms are in focus, Hebbian/associative plasticity and a complementary form, which is known as anti-Hebbian/homeostatic plasticity. 

Regarding the responsivity to neurofeedback large individual variability has been reported and four different reasons have been suggested to play a role. The responsiveness to neurofeedback and hence its efficacy may further be moderated by methodological factors. One such group of factors considers how to best address a given neural system and its means of communication. Here, the combination of frequency-specific feedback with EEG source analysis offers one approach. A further group of factors focuses on the optimization of training designs that follow the principles of basic learning mechanisms. 

Whereas many factors can be derived from our knowledge on the neural underpinnings of cognition, available measurement techniques, as well as basic learning mechanisms, much more systematic work needs to be conducted to optimize neurofeedback protocols for basic research and clinical applications.