The problem with neurofeedback training

There is a problem or two with neurofeedback training.  A doctor friend summed up one of them. He suggested that "It seems like a wonderful tool but i can't recommend it because I don't know how it works".  I pointed out that actually we don't really know how most of the medications prescribed today work, but we do this anyway because the evidence suggests that they do work (sometimes at least).

Definition of neurofeedback

Getting to grips with neurofeedback does takes some effort as it requires knowledge from across several domains and there has to be sufficient belief that it's worth making the effort. This article can't cover all aspects of neurofeedback but we can glimpse at some key aspects.

Evidence for Neurofeedback

The notion of what constitutes evidence is worthy of mention.  If we start out by thinking of an evidence metric based on how we would assess the effectiveness of a drug, we are going to be disappointed.  In a drug trial, a patient with a particular condition and representative clinical features, can be assigned to a placebo goup or an active treatment group on a random basis and both the patient and physician may well be "blind" to this assignment. 

Neurofeedback though is not a once size fits all approach and evidence for it's efficacy must often be gauged on a case by case basis.  Neurofeedback training is highly tailored by a knowledgeable person to the needs of the individual case and although the patient/subject's subconscious does some of the work, the process is one of learning (part software guided with direction from the therapist) that takes place over a number of sessions. 

At it's core, neurofeedback is an art as well as a science. It embodies a process of neuronal self-regulation - a learning process that leads the brain to find new and beneficial states. 


Surface electrodes detect microvolt level signals from the cortex area

There are a number of milestones in the development of neurofeedback training with the very first being the discovery in 1929 by Hans Berger of electrical activity arising in the brain that can be detected at the scalp.  It was realised that certain patterns could be seen in this electrical activity and that these could be related to moment by moment brain function and mind-body states. 

By the 1950's, the electroencephalogram (EEG) was in widespread use in (specialised) medical practice as a diagnostic tool.  However it was extremely challenging for EEG to be considered as anything more than a qualitative tool and this limited it's widespread practical application beyond diagnostic utility for particular conditions.

With advances in signal processing and computing technology we have the emergence of QEEG (quantitative electroencephalogram) which has allowed EEG signals to be rapidly and objectively analysed. 

The QEEG approach offers significant advantages compared to even the trained observer; subtle patterns of change can be detected in an inexpensive, reliable and safe fashion.  The emergence of QEEG has been very beneficial in a broader range of diagnostic applications than was possible before.

For example, QEEG has been shown by Thatcher et al to be very successful in identifying and discriminating Mild TBI from both normal controls and those exhibiting more severe brain injury.  However, we aren't going to focus so much on diagnosis in this article - neurofeedback takes us into the realm of learning.  Before learning


The second important milestone was the discovery of the brain's neuroplasticity.  Although the term itself is much used and even abused, it points to one of the most important discoveries in neuroscience.  The fact that the brain is not at all static, at any age, opens up some exciting possibilities as we explore how it is possible for the brain to constantly reconfigure itself. 

Neuroplasticity is sometimes described in the context of rehabilitation as if it is universally a good thing, but that is not actually true. Neuroplasticity is a property of our mind body system that can sometimes work against us.  One of the frequently heard mantras of neurorehab these days is that therapy needs to be specific, frequent and intensive to do any good but this must be interpreted with care.  Sometimes our brain learns and reconfigures in an undesirable way for our body as a whole when it adopts a new "normal" based on repeated damaging stimulation from the environment (eg PTSD). 

Here are some of the things that neurofeedback points to:-

A neurofeedback process

A neurofeedback process

  • Our brain continues to make new cells every day we live.
  • The brain can form complex synapses throughout life.
  • The connections between neurons can be strengthened against weakness.
  • There is no limit to brain repair. Previously, it was believed that the window for brain recovery was at most one year after injury;  research has shown that the brain can be repaired months and years after injury if the right intervention is applied. The question is "what are the right interventions?"
  • Brain repair mechanisms share commonalities across disease, injuries and age related declines. For example, active cognitive stimulation can help build new connections after traumatic brain injury, stroke, in normal ageing and even in progressive brain disease such as Alzheimer's.
  • Advances in sophisticated brain imaging technology allows us to view changes in the activation of brain regions that occur at the very moment we acquire new knowledge.

EEG patterns are sometimes referred to as "Brain waves". These waves are the simultaneous activation of thousands of neurons indicating a specific activity taking place in the brain.  Information cascades from neuron to neuron via the production of neuromediators, from one group of neurons to the next until it activates the appropriate reaction through the neurotransmitters.

When we examine classical EEG from a single electrode site we are examining potential from the cortex in that area and we see a time varying, quite chaotic signal which is hard to understand. When applying QEEG methods the first step is often to examine these EEG signals in the frequency domain - we split the original signal into a signal diagram that shows the intensity of energy at particular frequencies. This is based on the Fourier series idea that an periodic signal can be represented as a set of sin or cosine waves of carefully selected magnitude and frequency.  This frquency spectrum contains a lot of information that we have learned over the years to relate to mind body states.

One of the basic principles of how the cortex is organised is the existence of excitatory and inhibitory neurons. In general terms, excitation and inhibition are balanced within the cortex.  Brain waves are associated with certain activities and with triggering specific neurotransmitters that will in turn provoke the secretion of hormones by the organism. Certain neurons when activating will trigger the production of a neurohormone whose role is to regulate the activity of the of neurons that trigger this neurohormone. This in normal cases is a system of self-regulation through the activating and blocking features.

Neurofeedback prompts the autonomic nervous system to regulate the production of the neurotransmitters associated with it - acetylcholine for example as a "blocker" and adrenaine as an "activator"

The role of neurofeedback is to train (retrain) the brain to maintain itself within the thresholds of safety and consequently regulate the reaction of the central nervous system.

In related articles we will explore approaches to the application of neurofeedback


I wrote previously about neuroplasticity in the context of rehaCom brain training software.

Neuroscientist Michael Merzenich's TED talk "Growing evidence of brain plasticity"