Review
EEG-neurofeedback for optimising performance. I: A review of cognitive and affective outcome in healthy participants

https://doi.org/10.1016/j.neubiorev.2013.09.015Get rights and content

Highlights

  • The first review of validity in EEG-neurofeedback optimal performance field.

  • Over thirty controlled studies provide evidence of post-training benefits.

  • Twenty-three in addition provide evidence of neurofeedback learning.

  • Eight studies provide nine instances in support of feedback mediation in gains.

Abstract

A re-emergence of research on EEG-neurofeedback followed controlled evidence of clinical benefits and validation of cognitive/affective gains in healthy participants including correlations in support of feedback learning mediating outcome. Controlled studies with healthy and elderly participants, which have increased exponentially, are reviewed including protocols from the clinic: sensory-motor rhythm, beta1 and alpha/theta ratios, down-training theta maxima, and from neuroscience: upper-alpha, theta, gamma, alpha desynchronisation. Outcome gains include sustained attention, orienting and executive attention, the P300b, memory, spatial rotation, RT, complex psychomotor skills, implicit procedural memory, recognition memory, perceptual binding, intelligence, mood and well-being. Twenty-three of the controlled studies report neurofeedback learning indices along with beneficial outcomes, of which eight report correlations in support of a meditation link, results which will be supplemented by further creativity and the performing arts evidence in Part II. Validity evidence from optimal performance studies represents an advance for the neurofeedback field demonstrating that cross fertilisation between clinical and optimal performance domains will be fruitful. Theoretical and methodological issues are outlined further in Part III.

Introduction

Since the 1960s (Kamiya, 1968, Kamiya, 1969) the ability has been demonstrated for learned control of brain electrical activity through instantaneous feedback in the form of spectral power, event-related potentials or slow cortical potentials, putatively through operant conditioning or volitional control. This has now been extended to fMRI (e.g., Yoo et al., 2004, Weiskopf et al., 2004, deCharms, 2008, Rota et al., 2009, Hamilton et al., 2010, Zotev et al., 2011, Berman et al., 2012, Ruiz et al., 2013), transcranial doppler sonography (Duschek et al., 2011) and near infrared spectrometry (NIRS; Mihara et al., 2012, Kober et al., 2013). However, in contrast to autonomic and somatic peripheral biofeedback which became an established domain of Health Psychology (Feuerstein et al., 1988, Brannon and Feist, 2009), EEG-biofeedback was sidelined by science in the 1980s largely due to flawed studies and clinical overstatement, combined with the equivocal outcome of theoretically simplistic applications of alpha training for relaxation, when at the time psychophysiological understanding of alpha was at an early stage and was under the sway of unitary arousal models (Lindsley, 1952, Duffy, 1957). From the 1980s university research on EEG-biofeedback, later branded neurofeedback, was confined to a few centres, notably under the direction of Sterman (e.g., Sterman and Friar, 1972, Sterman et al., 1970, Sterman et al., 1974, Sterman, 1996, Sterman, 2000) and Lubar (e.g., Lubar and Shouse, 1976, Lubar and Lubar, 1984, Lubar et al., 1995a, Lubar et al., 1995b) in the USA and Birbaumer in Germany (Elbert et al., 1984, Rockstroh, 1989). Meanwhile a largely dedicated practitioner following arose outside of universities, especially in North America, encouraged by instrumentation companies. While helping keep the field alive and inventing diverse treatment protocols, this following mostly proceeded without scientific validation, sometimes accompanied by pseudo theorising and advertising speak.

Over the last decade or so university research, largely in Europe, has revisited neurofeedback, a revival of interest which the Society of Applied Neuroscience (SAN) has encouraged, and, as this review shows scientific evaluation and an evidence-base for applications is growing exponentially. This has come at a time when enhancing potential by ‘brain training’ is fashionable in contemporary culture (Owen et al., 2010, Rabipour and Raz, 2012), making the cause of scientific validation ever more essential and urgent, for aside from false promises, premature popularisation has led science to discard a field more than once.

While ‘neurofeedback’ has now been applied to a range of brain imaging modalities, here the focus will be on the accumulating evidence of validation particularly this millennium in favour of neurofeedback applications with the EEG in healthy participants, commonly called the ‘optimal’ or ‘peak performance’ field. Validation has taken the form of controlled studies showing differential group outcomes advantaging a neurofeedback protocol, and more importantly has included correlation between the feedback learning and the outcome allowing the mediation of neurofeedback learning to be inferred. The review will be initially structured around neurofeedback protocols in conjunction with psychological processes, followed by applications in sport and with the elderly. A companion review, Part II, will cover applications in the performing arts and creativity (Gruzelier, 2013a) while Part III will be devoted to methodological and theoretical issues in order to help advance this re-emergent field (Gruzelier, in preparation). The review will not concern clinical applications which have been the main concern of the field to date. To give some examples there are reviews on Attention Deficit Hyperactivity Disorder (ADHD; Arns et al., 2009, Lofthouse et al., 2012), epilepsy (Tan et al., 2009), autism spectrum disorder (Coben et al., 2010); controlled studies on addiction (Scott et al., 2005), tinnitus (Hartmann et al., 2013) and insomnia (Cortoos et al., 2010); case studies on stroke (Bearden et al., 2003), anxiety and depression (Hammond, 2005) and cognitive disorganisation with anhedonia in conjunction with drug misuse (Unterrainer et al., 2013). Similarly there are reviews on the burgeoning approach branded Brain Computer Interface (BCI), one typically focussing on the neuro-rehabilitation of patients with tetraplegia from brain or spinal cord injuries enabling patients through EEG feedback to communicate and/or physically interact with their environment, and up until now largely consisting of engineering innovations and a small clinical evidence base (Birbaumer et al., 2008, Mak and Wolpaw, 2009, Silvoni et al., 2011).

As will be seen a diversity of neurofeedback training protocols has been applied for optimising performance. The most popular one has involved training-up the amplitude of the Sensory Motor Rhythm (SMR) 12–15 Hz band while inhibiting outer-lying bands in the EEG spectrum. This had followed evidence that within the 12–15 Hz frequency range recordings from sensory-motor and pre-motor cortices showed a distinctive oscillation which was maximal during periods of quiet wakefulness, with reduced muscle tone, and was absent during goal directed activity and desynchronising with motor intention (Sterman and Wyrwicka, 1967, Wyrwicka and Sterman, 1968, Sterman and Friar, 1972, Sterman et al., 1974). Notably application of the SMR neurofeedback protocol has reduced motor seizure rates in epileptic patients while normalising their sleep patterns and EEG (Sterman, 2000, Sterman and Friar, 1972, Sterman et al., 1970, Sterman et al., 1974). By extrapolating to ADHD the potential of reducing the excitability of the sensorimotor system with concomitant suppression of theta, and following SMR training with training up adjacent low beta activity (16–22 Hz; beta1) an index of EEG desynchronisation, improvements in attention and hyperactivity were first demonstrated in case studies and early controlled trials (e.g., Lubar and Shouse, 1976, Lubar et al., 1995a, Rossiter and LaVaque, 1995, Linden et al., 1996) and now have a substantive evidence base (Monastra et al., 2005, Arns et al., 2009, Lofthouse et al., 2012).

Another pioneering protocol involved raising the theta-alpha ratio with auditory feedback and eyes closed, termed alpha/theta (A/T) training. Following on from the earliest attempts to up-train alpha for relaxation and reduce anxiety (Budzynski and Stoyva, 1972, Hardt and Kamiya, 1978), the focus on theta first grew out of diverse cultural evidence that the deeply relaxing, hypnogic reverie or twilight theta state was conducive to creative insights (see Gruzelier, 2009). From clinical studies the goal evolved of elevating theta over alpha to achieve crossover. Early on this had demonstrable benefits for addiction and PTSD which included enhanced well-being when introduced as a central part of a therapeutic package (e.g., Peniston and Kulkosky, 1989, Peniston and Kulkosky, 1991, Saxby and Peniston, 1995).

Subsequently the SMR and A/T protocols have been contrasted for their optimal performance effects in healthy participants, while new neurofeedback protocols have been developed from contemporary neuroscience including training upper-alpha, peak alpha frequency, gamma, and various theta protocols, the subject of this review. Historically self-regulation of slow cortical potentials (SCPs) had received extensive validation (Elbert et al., 1984, Rockstroh, 1989), but has not attracted interest in the optimal performance field despite promise in trials with ADHD where it has been compared favourably with EEG-spectrum training (Gevensleben et al., 2009).

In what must be a necessarily concise selection of study features to be considered the following issues have been documented.

What evidence is there that feedback learning occurred? Have learning indices been reported or is learning inferred from demonstration that a group receiving neurofeedback out-performed a comparison group? What would evidence of learning consist of: learning curves within sessions, across sessions, the correlation of learning indices with outcome assessments? If the most basic theoretical premise about neurofeedback holds, namely that the brain's rhythms are changed by training, is learning reflected in subsequent training sessions including baselines?

There are issues of specificity. (i) Is there band specificity or independence such that only the trained bands are influenced or is there leakage or even reciprocity within the EEG spectrum? (ii) Is there specificity and independence regarding cognitive/affective outcome such that performance enhancement is specific to some processes leaving other processes unchanged? (iii) Is there topographical specificity such that the EEG outcome is specific to the training site, or is it distributed locally beyond the training site, or is it only distal from the training site?

Are there sufficient sessions to give learning a chance given that a learning process is being built? Can learning even be obtained from one-session and if so what is the validation? What is the interval between sessions given there are considerations for learning of spaced versus distributed practice. Are some people unable to learn to self-regulate their brain rhythms – sometimes termed non-responders? The outsider may be surprised at how few subjects are typically included in most intervention studies. Consider though that with say fourteen subject sessions to include a ten session course of training plus two pre and post-training outcome assessments, then for a study involving three groups of ten this would add up to 420 experimenter sessions to acquire data, and not allowing for dropout replacements which are inevitable when enterprising, mostly student subjects are required to fit fourteen sessions into their busy lives; let alone the logistics of engaging as trainees professionals such as eye surgeons (Ros et al., 2009).

All these issues will be reconsidered in detail in Part III focussing on methodology and theory (Gruzelier, in preparation). In Part I the over-riding objective will be to weigh up the now rapidly accumulating evidence base in the search of validation of neurofeedback, a concern of the author's from an early review (Gruzelier et al., 2006). As will be seen the majority of studies report evidence of neurofeedback learning and in this respect the optimal performance field is far ahead of other neurofeedback domains in reporting evidence of learning.

Section snippets

Sustained attention, selective attention and memory: SMR and beta1 training

SMR and beta1 neurofeedback protocols widely used with ADHD were found to have a positive impact on the sustained attention of healthy subjects in two studies (Egner and Gruzelier, 2001, Egner and Gruzelier, 2004a) which were an adjunct to music performance enhancement (Egner and Gruzelier, 2003; see for review Gruzelier, 2012, Gruzelier, 2013a). In the first music study as a secondary aim SMR (12–15 Hz, C4) and beta1 (15–18 Hz, C3) protocols with theta (4–7 Hz) and high beta (22–30 Hz; beta2)

Alpha/theta training

In establishing what was the first evidence of operant control over the theta/alpha ratio with eyes closed, Egner et al. (2002) had shown in healthy subjects that when comparing a contingent with a non-contingent control A/T protocol, both protocols led to identical ratings on Activation Deactivation self-report scales (Thayer, 1967), whereas only contingent training produced an increase in the theta/alpha ratio. One implication from this was that there was likely to be more to the reported

Sporting skills

Applications of neurofeedback to improve sporting performance and to compare elite with novice performance are at an early stage but will be apposite (Thompson et al., 2008). Little progress has been made since the pioneering studies of Landers et al. (1991) who reported that the feedback of an EEG analysis of slow potentials improved performance in pre-elite archers. They followed evidence that in sportsmen prior to the execution of a skill there was a unilateral slowing of left temporal

Validation

Having reviewed the efficacy of EEG-neurofeedback for optimising function in healthy individuals, before concluding Part I there are two sources of evidence in support of validation to be summarised. The first is where a neurofeedback experimental group is characterised by a uniquely successful cognitive and/or affective outcome when compared with a comparison group and/or a control group; should in the absence of a control group two neurofeedback protocols show the same benefits these are not

References (169)

  • T. Egner et al.

    The effects of neurofeedback training on the spectral topography of the healthy electroencephalogram

    Clinical Neurophysiology

    (2004)
  • S. Enriquez-Geppert et al.

    Boosting brain functions: improving executive functions with behavioral trainings, neurostimulation, and neurofeedback

    International Journal of Psychophysiology

    (2013)
  • M. Grunwald et al.

    Theta power in the EEG of humans during ongoing processing in a haptic object recognition task

    Cognitive Brain Research

    (2001)
  • J.H. Gruzelier et al.

    Validating the efficacy of neurofeedback for optimising performance

  • H.J. Haarmann et al.

    Individual differences in semantic short-term memory capacity and reading comprehension

    Journal of Memory and Language

    (2003)
  • H.J. Haarmann et al.

    Active maintenance of sentence meaning in working memory: evidence from EEG coherences

    International Journal of Psychophysiology

    (2005)
  • S.W. Hughes et al.

    Synchronized oscillations at alpha and theta frequencies in the lateral geniculate nucleus

    Neuron

    (2004)
  • T. Inouye et al.

    Potential flow of frontal midline theta activity during a mental task in the human electroencephalogram

    Neuroscience Letters

    (1994)
  • N. Jaušovec et al.

    Differences in event-related and induced EEG patterns in the theta and alpha frequency bands related to human emotional intelligence

    Neuroscience Letters

    (2001)
  • A.W. Keizer et al.

    The effect of gamma enhancing neurofeedback on the control of feature bindings and intelligence measures

    International Journal of Psychophysiology

    (2010)
  • A.W. Keizer et al.

    Enhancing cognitive control through neurofeeback: a role of gamma-band activity in managing episodic retrieval

    Neuroimage

    (2010)
  • W. Klimesch

    EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis

    Brain Research Reviews

    (1999)
  • W. Klimesch et al.

    EEG alpha oscillations: the inhibition–timing hypothesis

    Brain Research Review

    (2007)
  • Y. Kubota et al.

    Frontal midline theta rhythm is correlated with cardiac autonomic activities during the performance of an attention demanding meditation procedure

    Cognitive Brain Research

    (2001)
  • D.B. Lindsley

    Psychological phenomena and the electroencephalogram

    Electroencephalography and Clinical Neurophysiology

    (1952)
  • K.J. Anderson

    Arousal and the inverted-U hypothesis: a critique of Neiss's “Reconceptualizing arousal”

    Psychological Bulletin

    (1990)
  • Arns, M., Keneman, L., 2013. Neurofeedback in ADHD and insomnia: Vigilance stabilization through sleep spindles and...
  • M. Arns et al.

    Golf performance enhancement and real-life neurofeedback training using personalized event-locked EEG profiles

    Journal of Neurotherapy

    (2008)
  • M. Arns et al.

    Effects of neurofeedback treatment on ADHD: the effect on inattention, impulsivity and hyperactivity: a meta-analysis

    Clinical EEG Neuroscience

    (2009)
  • E. Basar et al.

    Brain alpha activity: new aspects and functional correlates

    International Journal of Psychophysiology

    (1997)
  • O.M. Bazanova et al.

    Interpreting EEG alpha activity

    Neuroscience and Biobehavioural Reviews

    (2013)
  • T.S. Bearden et al.

    Neurofeedback training for a patient with thalamic and cortical infarctions

    Applied Psychophysiology and Biofeedback.

    (2003)
  • J. Becerra et al.

    Neurofeedback in healthy elderly human subjects with electroencephalographic risk for cognitive disorder

    Journal of Alzheimers Disease

    (2012)
  • J. Beatty et al.

    Operant control of occipital theta rhythm affects performance in a radar monitoring task

    Science

    (1974)
  • I. Berner et al.

    The significance of sigma neurofeedback training on sleep spindles and aspects of declarative memory

    Applied Psychophysiology and Biofeedback

    (2006)
  • N. Birbaumer et al.

    Brain-computer interface in paralysis

    Current Opinion in Neurology

    (2008)
  • B.L. Bird et al.

    Biofeedback training of 40-Hz EEG in humans

    Biofeedback and Self-Regulation

    (1978)
  • L. Brannon et al.

    Health Psychology: An Introduction to Behaviour and Health

    (2009)
  • T.H. Budzynski et al.

    Biofeedback techniques in behaviour therapy

  • A.P. Burgess et al.

    Short duration synchronization of human theta rhythm during recognition memory

    Neuroreport

    (1997)
  • J.F. Cavanagh et al.

    Theta lingua franca: a common mid-frontal substrate for action monitoring processes

    Psychophysiology

    (2011)
  • C.R. Clark et al.

    Spontaneous alpha peak frequency predicts working memory performance across the age span

    International Journal of Psychophysiology

    (2004)
  • R. Coben et al.

    Neurofeedback for autistic spectrum disorder: a review of the literature

    Applied Psychophysiology and Biofeedback

    (2010)
  • A. Cortoos et al.

    An exploratory study on the effects of tele-neurofeedback and tele-biofeedback on objective and subjective sleep in patients with primary insomnia

    Applied Psychophysiology and Biofeedback

    (2010)
  • Y.M. Cycowicz et al.

    Remembering the colour of objects: an ERP investigation of source memory

    Cerebral Cortex

    (2001)
  • R.C. deCharms

    Applications of real-time fMRI

    Nature Reviews Neuroscience

    (2008)
  • S. Diekelmann et al.

    The memory function of sleep

    Nature Reviews Neuroscience

    (2010)
  • E. Duffy

    The psychological significance of ‘arousal’ or ‘activation’

    Psychological Review

    (1957)
  • F.H. Duffy et al.

    Age-related differences in brain electrical activity of healthy subjects

    Annals Neurology

    (1984)
  • M. Doppelmayr et al.

    Effects of SMR and theta/beta neurofeedback on reaction time, spatial abilities and creativity

    Journal of Neurotherapy

    (2011)
  • Cited by (354)

    View all citing articles on Scopus
    View full text