Brain-Based Performance

01What the recordings show

The Numbers Behind Quantitative EEG in Athletes

Eight findings on what a brain recording establishes about an athlete, and where the reading stops.

02Speed upstream of muscle

Brain-Based Performance Training Works Where the Movement Is Decided

Every fast movement an athlete makes was first a fast decision in the brain. Vision, prediction and motor planning all run before a muscle ever contracts, and the time spent there is measured in milliseconds the scoreboard never shows. Brain-based performance training is the work of measuring that upstream stage and training the nervous system directly, which is why the highest ceiling in sport is a neurological one.

This is why two athletes with identical strength can perform worlds apart. The faster nervous system gathers cleaner sensory input, integrates it in higher centers, and issues a more precise motor command. Sharper afferent input, faster central processing and more accurate output are skills, and skills respond to specific loading. That pathway is trained as deliberately as a coach trains the squat. The brain is the most trainable organ an athlete owns.

What the cortex does before a practiced movement

The cortex of an expert works less, not harder, and the difference is visible in a recording. Ten elite karate athletes and 12 non-athletes performed brisk voluntary wrist extensions under a 56 channel EEG Del Percio 2010. Alpha desynchronization was lower in the athletes in primary motor cortex and in the lateral and medial premotor areas.

Alpha desynchronization is the drop in 8 to 12 Hz power that marks a region switching on. Less of it for the same movement means the same job done with less cortex recruited. The authors call this neural efficiency, and it appeared during both the preparation and the execution of the movement.

Reaction time is the output measure most programs already collect, and its variability carries more information than its mean. Reaction Time and Motor Control holds that evidence. This page stays on the instrument that reads the brain directly.

03What a QEEG records

Quantitative EEG Turns Cortical Rhythm Into Numbers, One Scalp Site at a Time

Brain-based training begins by measuring the brain, not guessing at it. Quantitative EEG is the instrument that does the measuring. Electrodes sit at named positions on the scalp and the amplifier records the voltage swinging under each one. Software then breaks that signal into frequency bands and reports how much power sits in each band at each site.

The standard clinical array is the international 10-20 system, and its 19 recording positions cover the head at fixed proportional distances. That density is not arbitrary. Sensorimotor mu rhythm clusters, the rhythm most directly tied to movement, were recovered reliably from those 19 channels when the same recordings were also analyzed at 61 channels Tacchino 2015.

The bands are conventions with physiology behind them. Delta and theta are the slow end. Alpha runs roughly 8 to 12 Hz and rises when a region idles. Beta sits above it. A report scores an athlete's power in each band at each site against a reference database of people the same age, which is where the phrase normative comparison comes from. The result is a profile of how this specific nervous system processes the world.

Which numbers hold still enough to trend

Only some quantitative EEG measures are stable enough to follow across a season. Fifteen healthy adults were recorded ten times over two months, and reliability was highest for power spectral measures, next for entropy and complexity measures, and lowest for coherence Gudmundsson 2007.

Montage and epoch length moved the result considerably. The average montage was most reliable, and reliability improved with epoch length up to 40 seconds, with little gain beyond. Coherence reliability depended strongly on which channels and which frequency band were involved.

That is a practical instruction rather than a complaint. A coherence map is the most visually persuasive page of a QEEG report and the least trustworthy line to trend in one athlete. Power measures at fixed sites, recorded the same way each time, are what a serial reading can be built on.

What the scalp cannot reach

From that map come the drivers that matter most for the athlete's sport. Frontal regions govern decision speed and impulse control, and the electrodes sit directly over them. The two other systems that drive athletic timing are not read the same way. The cerebellum tunes timing and coordination and the vestibular system anchors the head and eyes in space, and neither produces a signal a scalp electrode reads directly.

Those systems are measured through their outputs instead: eye movement accuracy, gaze stabilization and postural sway. The Cerebellum and Timing carries the cerebellar timing evidence and Vestibular Function and Balance carries the vestibular and sway metrics. Pairing a cortical recording with those output tests is what makes the profile mean something. What gets measured gets sharpened.

04The boundary of the reading

What a Quantitative EEG Does Not Establish

A quantitative EEG describes cortical rhythm at the moment of the recording. It does not diagnose an injury, grade an athlete, or measure talent. Three limits on the reading are documented well enough to state as fact, and each one has a direct consequence for how an athlete is recorded.

The first is muscle. Two subjects were recorded with and without complete neuromuscular blockade while performing the same cognitive tasks Whitham 2007. The power of frequencies above 20 Hz differed by 10 to 200-fold between the two states. Most of what a scalp electrode picks up above 20 Hz comes from muscle rather than brain.

For an athlete that limit lands hard. A clenched jaw, a braced neck and a held breath all sit inside the recording, and a competitor who arrives tight from a session brings more of them than a rested office worker. Beta and gamma readings taken without artifact control describe the muscles of the head as much as the cortex under them.

Where the boundary matters most

The second limit is diagnostic. A review of EEG and quantitative EEG after mild traumatic brain injury found slowing of the posterior dominant rhythm and increased diffuse theta after injury Nuwer 2005. It also found no features unique to that injury. Late after a head injury the correspondence between the electrical findings and the clinical symptoms is poor.

The sports-specific review reached the same place. A search of 8,567 publications on new concussion technologies produced nine studies with original data and a comparison group, four of them on quantitative EEG Kutcher 2013. The authors record that clinical utility remains to be established. Concussion recognition and clearance run on the clinical protocol described in Concussion in Sport.

When the normal range itself moves

The third limit is the normative comparison. The most studied band ratio in the field pooled 1,253 children with attention deficit hyperactivity disorder and 517 without Arns 2013. The grand-mean effect size was 0.75 for ages 6 to 13 and 0.62 across ages 6 to 18.

Heterogeneity remained significant, so the authors treat those effect sizes as an overestimation. The gap also shrank across the decade, and the reason was that the ratio rose in the children without the diagnosis. The comparison group moved. A ratio that separates two groups on average cannot classify the person in front of you, and the authors conclude it works as a prognostic measure in a subgroup rather than a diagnostic one.

What survives all three limits is real and worth having. A QEEG gives a repeatable description of one athlete's cortical rhythm, built from the measures stable enough to trend, recorded the same way each time. It is a baseline, and its value comes from the comparison with that athlete's own later readings rather than from a verdict printed against a database.

05The expert recording

Elite Performers Show Less Cortical Activation Before Their Best Attempts

The clearest EEG signature of expertise in sport is a quieter cortex. Eighteen elite air pistol shooters and 10 non-athletes each took 120 shots under a 56 channel recording Del Percio 2009. Across the whole scalp, the drop in alpha power that marks a region switching on was smaller in the elite shooters than in the novices.

The within-athlete result is the more useful one. In the elite shooters, alpha synchronization at 10 to 12 Hz was larger for the higher-scoring half of their shots than for the lower-scoring half, over right parietal and left central areas. The same difference between good and poor shots did not appear in the non-athletes.

Two claims come out of that. Skill is visible in the electrical record before the movement happens, and it looks like restraint rather than effort. The finding repeats in a sport with nothing in common with pistol shooting, since elite karate athletes also showed less alpha desynchronization than non-athletes during a simple wrist extension Del Percio 2010.

Reading that result against the muscle problem

The alpha finding survives the artifact question and the beta finding is weaker, which is worth stating plainly. The shooting study confirmed its result with a second index of cortical activation in the beta range around 20 Hz, and 20 Hz is exactly where scalp recordings start carrying muscle Whitham 2007.

Surface Laplacian estimation, which both studies used, sharpens the spatial reading and reduces the contribution of distant sources. It does not remove muscle from a frequency band. The alpha result at 8 to 12 Hz sits well below the contaminated range, so that is the measurement a performance program should build on.

This is also the reason the pre-shot window is the interesting one. It is the same window where quiet eye behavior is measured, and Sport Vision and Eye Movements carries the oculomotor side of the same moment.

06What the neurofeedback trials show

Neurofeedback Moves Some Measures, and the Sham Trials Set the Ceiling

Neurofeedback is the training arm of brain-based performance work, and its athlete trials are small, positive on average, and thin at the point where it matters most. The method is simple to describe. An athlete watches a display driven by their own EEG power in a chosen band, and the display rewards them whenever that band moves the intended way.

The best-known sport result used sensorimotor rhythm, the 12 to 15 Hz band recorded over the sensorimotor cortex. Sixteen preelite golfers were randomly assigned to sensorimotor rhythm training or a control condition for eight sessions Cheng 2015. The trained group putted more accurately afterward and showed greater sensorimotor rhythm power during action preparation.

Pooled across athletes, the reaction time effect is the one with a confidence interval that clears zero. Seven randomized controlled trials covering 173 athletes produced a standardized mean difference of 1.08 in favor of neurofeedback for reaction time, with a 95 percent confidence interval of 0.25 to 1.90 de Brito 2022.

The results that did not follow

The larger analysis in healthy adults is where the shape of this literature becomes visible. Thirty-three studies were reviewed and 16 randomized trials covering 374 subjects were pooled Onagawa 2023. The standardized mean difference for motor performance after the last session was 0.85, with a 95 percent confidence interval of 0.18 to 1.51.

The same authors report publication bias and substantial heterogeneity across those trials. They also state that the effect on each individual motor measure, such as speed, accuracy and hand dexterity, stays inconclusive because the samples are small. Their meta-regression found a dose gradient, with benefit appearing beyond roughly 125 minutes of cumulative training.

One athlete trial shows the gap directly. In an ice hockey shooting study, both the neurofeedback group and the control group improved, and the trained group improved faster Christie 2020. The trained group also raised sensorimotor rhythm power in the laboratory. During actual shooting, their sensorimotor rhythm power did not change.

The band moved where it was trained and not where it was needed. That gap is the central unsolved problem of the method rather than a footnote to it.

What the sham comparisons found

Two controlled literatures set the ceiling. A review of neurofeedback studies in healthy adults with proper control groups found that protocols training fewer bands and using more electrodes produced larger changes in the trained band Rogala 2016. It found no evidence supporting a relationship between that power change and specific behavioral effects.

The sharpest test came from outside sport. Twenty-five patients with primary insomnia spent nine polysomnography nights in the laboratory and received 12 sessions of sensorimotor rhythm neurofeedback and 12 sessions of sham feedback Schabus 2017. Both were equally effective on their sleep complaints, and the improvement did not appear in the objective EEG-derived measures.

The authors attribute the change to nonspecific factors, including trust and the experience of receiving care. That result decides how a serious program talks about the method. Neurofeedback changes the trained band under laboratory conditions. Whether the change carries into the arena, and whether the band is the reason for anything the athlete feels, are separate questions with separate evidence.

The Unified Model of Tone reads the scatter across these trials as the input law rather than as noise in the method. There is no such thing as an input acting on an empty body, so the same protocol meets a different starting organization in every athlete and produces a different result. Group means average away the thing the model says is doing the work.

The rest of brain-based work is sensory rather than electrical, and it runs on the visual, vestibular and cerebellar pathways that feed and steer movement. Visual work loads saccades, smooth pursuit and peripheral awareness, the machinery that picks up targets sooner. Vestibular work loads gaze stabilization during head rotation and impact. Cerebellar drills load timing and rhythm. Each pathway is specific, and each responds to the right dose of stimulation. The Functional-Neurology Workup sets out how those systems get tested together in a single visit.

07State under speed

Composure Is a Measurable State, and One Input Meets Many States

Composure is a measurable nervous-system state, not a personality trait. Whether an athlete stays composed when the game speeds up is a question about that state. Heart rate variability read through RMSSD tells a coach whether an athlete is recovered or still running hot, and Heart Rate Variability carries that measurement in full. A brain governed by an over-revved sympathetic system narrows attention, slows decisions and tightens movement at the exact moment sport demands the opposite.

What a program sets out to train is the central integrative state: the sum of everything arriving at a neuron at once, which decides how easily that neuron fires next. The Neuron and the Central Integrative State gives the cellular account. Breath pacing, autonomic drills and targeted stimulation are inputs aimed at that state, shifting the balance toward poise. The goal is keeping the prefrontal cortex online when the game accelerates, so decisions stay sharp and the hands stay quiet.

Precise chiropractic adjustment enters the same account for the same reason. Mechanoreceptor input from the spine feeds the cerebellum and cortex, which is where motor output gets shaped. Cleaner input builds a faster brain, and the evoked potential is where that arrival can be measured.

The N30 peak, recorded over frontal cortex after median nerve stimulation, indexes how sensory input is being integrated with motor areas. All three studies below report it.

The same thrust, three directions

N30 after spinal manipulation has no fixed direction, and the split is worth stating in full. In 12 people with recurring neck stiffness and no acute symptoms, a single cervical manipulation lowered both the parietal N20 and the frontal N30 Haavik-Taylor 2007. The change lasted about 20 minutes, and a passive head movement control condition produced none of it.

In 17 men with chronic stroke, the same class of input raised the N30 amplitude by 39 percent compared with a control intervention Navid 2020. In 96 adults with recurrent mild neck pain, a thrust delivered to a clinician-selected segment lowered the N30 complex by 16.76 plus or minus 28.32 percent Niazi 2024. A thrust at a predetermined segment produced no change at all.

Three studies, one class of input, three different answers. A model that predicted one fixed effect would have to explain away two of them. The Unified Model of Tone predicts exactly this, because the effect of an event is set by how it meets the organization already present. Different starting states, different directions, one input law.

The resting map missed what the evoked response caught

The stroke study contains the methodological finding this page exists to state. In the same sessions where the N30 moved by 39 percent, the resting power spectra did not change, the delta-alpha ratio did not change, and the brain-symmetry index did not change Navid 2020. Source localization of the resting recording showed nothing either.

Resting quantitative EEG and the evoked response are two windows onto the same cortex with different sensitivities. One is sampling ongoing rhythm with the athlete doing nothing. The other is timing the cortex's answer to a known stimulus, averaged over hundreds of repeats, at millisecond resolution.

The practical rule follows directly. A flat resting map after an intervention is not evidence that nothing reached the cortex, because the instrument that showed the change was the other one. Motor evoked potentials and corticospinal excitability form a third window, and Cortical Drive and Force carries that evidence.

08The retest decides

A Brain-Based Program Is Judged on the Retest, Not the Session

Whatever a brain-based program claims is settled by the retest. Changes in the nervous system are written through neuroplasticity, the lifelong capacity to reorganize around the input that keeps arriving, and Neuroplasticity and Rehabilitation carries that mechanism. QEEG, reaction time and balance are recorded again so that progress is shown, not assumed.

A retest is only as good as the stability of the measure it repeats. Power spectral measures held steady across ten recordings in two months and coherence did not Gudmundsson 2007, so a serial brain reading is built from power at fixed sites. The output tests carry their own thresholds, and The Functional-Neurology Workup sets out how far each has to move before the change belongs to the athlete.

Dose is part of the same accounting. The pooled analysis of motor performance trials found benefit concentrated beyond roughly 125 minutes of cumulative training Onagawa 2023, which is a threshold rather than a guarantee. The brain that practiced the right drill keeps the upgrade, and the recording is how anyone knows whether it did.

This is the work that compounds across a career. The body sets the floor, but the brain sets the ceiling, and the ceiling is where elite sport is decided.

09Corrections to this page

Two Claims and One Quotation Removed

This page previously claimed elite reaction gaps of 30 to 80 milliseconds against an average competitor. No published source carries that comparison at any size, so the number does not appear here. Reaction time and its variability are measured in milliseconds and reported with their numbers in Reaction Time and Motor Control.

The page also credited Suter and colleagues with documenting reduced muscle inhibition after cervical adjustment. The study exists and the site is wrong. Suter and colleagues manipulated the sacroiliac joint in patients with anterior knee pain and measured inhibition in the knee extensors Suter 1999. The attribution has been corrected rather than repeated.

A quotation attributed to Dr. Jason Dulberg about how much of the world the brain actually sees was not drawn from anything he said or wrote, and it has been removed. Every claim made here about quantitative EEG and brain-based performance training is either sourced to the literature or named as the model's.

10The model's claim

One Amplifier, Four Readouts, One Organization

Two layers run through this page and the instrument keeps them apart. Layer one is what the electrodes recorded and the statisticians pooled. The reliability coefficients, the muscle contamination above 20 Hz, the alpha result in the shooters, the pooled neurofeedback effects, and the three directions the N30 took. Each of those belongs to the group that measured it.

Layer two is what this practice claims about the set. The Unified Model of Tone holds that these measurements are readings of one organization rather than of separate faculties that happen to share a skull. It names four readouts. How a rhythm varies internally, how a slow rhythm couples to a fast one, how responsive a reflex is, and how long a return to baseline takes after a demand.

Why the QEEG room is the right place to test it

Each of those four is a window onto one organization, and the instrument on this page can open two of them at once. Cross-frequency coupling is computed from the same recording that produces the power spectra. Reflex responsiveness at the cortex is the evoked potential, recorded from the same electrodes in the same sitting Haavik-Taylor 2007.

The third readout arrives on an electrocardiogram lead running alongside the EEG. The fourth is a clock. So the whole battery the model names can be recorded on one athlete, in one chair, in one afternoon, with equipment a functional neurology practice already owns.

The prediction this page makes

Take one squad and record four things, four times across a season. High-frequency alpha synchronization at 10 to 12 Hz in the second before a self-paced attempt. N30 amplitude to median nerve stimulation. RMSSD from a five minute seated recording. Time for the resting alpha map to return to its pre-exertion values after a standardized submaximal exertion protocol.

Each reading is scored against that athlete's own preseason values rather than a population database, and only the measures stable enough to trend are used Gudmundsson 2007. This is a claim about how performance is organized rather than a claim about what treatment does.

Nobody has recorded the four together in one athlete cohort, so the question of whether they move as one has never been put. The pairwise version already exists in fragments. Alpha synchronization tracks shot quality within the same shooters Del Percio 2009, and a single spinal input moves the N30 while leaving the resting map untouched Navid 2020.

If pre-attempt alpha synchronization, N30 amplitude, RMSSD, and time to return to the pre-exertion resting map are shown to move together within athletes across a season, the unification claim is confirmed.

11The tone reading

What the Electrodes Are Reading When They Read an Athlete

Three signatures of tone appear on this page, each inside a measurement the recording already produces.

The rest of the library carries the same logic through its other foundations. Coupling is the phase relationship between a slow rhythm and a fast one, computed from the same recording that yields the power spectra. Prediction is what the cortex is doing in the quiet second before the target is engaged. Time course decides how long a change lasts, and the manipulation effect on the evoked response ran about 20 minutes. Load arrives in the chair with the athlete, since a hard session leaves muscle tension in the recording. Constraint is what a cortex held too tightly loses, which is the quiet the best attempts run on. Set-point is the value being defended, and a normative database prints a population's rather than this athlete's. The full framework is set out in the Unified Model of Tone.

12Where this sits

How This Page Relates to the Rest of the Library

Seven places this argument continues, each with the claim that earns the link.

13Questions athletes ask

Questions Athletes Ask

14The sources

References

18 primary sources, each linked to its record. Figures quoted on this page were checked against the published abstract.