Psychology by BogdanovaThe Encyclopedia of Psychology for the Artificial Era

RESEARCH DOSSIER

Psychophysiological Regulation, Biofeedback, and Neurofeedback in Sport

Abstract. Psychophysiological regulation in sport comprises learned modifications of respiratory, autonomic, muscular, and neural processes that can influence emotional readiness, attention, motor execution, and recovery. Biofeedback makes a selected physiological signal available to the athlete; neurofeedback uses neural activity, usually electroencephalography, as the feedback variable. This research monograph critically integrates the physiological foundations, history, intervention protocols, controlled sporting experiments, systematic reviews, and methodological challenges of these approaches. It distinguishes demonstrated change in a training signal, durable acquisition of self-regulation, transfer to standardized sporting performance, and reliable improvement in competition. Evidence supports the trainability of some cardiorespiratory and EEG indices and suggests benefits in certain precision-motor tasks, while protocol specificity, study size, inadequate controls, outcome heterogeneity, and limited follow-up constrain general conclusions. Recent pooled analyses report positive average effects but also explicitly low or very low certainty of evidence. A scientifically useful account therefore connects signal physiology, learning mechanisms, causal study design, contextual sport demands, and transparent interpretation of uncertainty.

Keywords: sport psychology; psychophysiological regulation; biofeedback; neurofeedback; heart rate variability; respiratory sinus arrhythmia; baroreflex; resonance-frequency breathing; electroencephalography; sensorimotor rhythm; Mu rhythm; psychomotor efficiency; motor performance; self-regulation; sham-controlled trials.

Introduction

Sport is a demanding setting for psychophysiological research because mental preparation, bodily regulation, and skilled action occur together. A successful golf putt, basketball free throw, rifle shot, or timed sprint depends on domain-specific technique and practice. The athlete's cardiorespiratory state, attentional allocation, muscular tension, and reaction to competitive pressure can contribute to that action, although their effects differ with task structure and individual physiology. Instrumental feedback offers an additional route through which athletes might learn to recognize and modify some of these processes.

Psychophysiological regulation refers here to changes in the organized relationships between physiological processes and psychological activity that are relevant to preparing, executing, and recovering from skilled performance. Biofeedback is a learning procedure in which physiological measurements are converted into information usable by the individual. Neurofeedback is the subset in which a measurement of neural activity functions as the target feedback signal. These definitions specify training procedures and measured variables. They do not entail that any particular device, signal threshold, or intervention causes superior athletic achievement.

The central scientific problem concerns transfer. A sensor may register a larger oscillation in heart rate when breathing is paced, or an athlete may increase power in a selected EEG frequency band during a training game. Such changes demonstrate a proximal manipulation or adaptation. Their functional meaning depends on whether the athlete can reproduce the response without the feedback display, retain it after training, and improve performance relative to a scientifically credible comparator. The chain from measurement to competitive outcome contains multiple causal links; a convincing study evaluates each link rather than assuming that the last follows automatically from the first.

A useful historical foundation is the cardiovascular resonance work of Vaschillo, E., Lehrer, P., Rishe, N., & Konstantinov, M (2002), complemented by the physiological integration of Lehrer, P. M., & Gevirtz, R (2014). On the neural side, controlled sport experiments such as Cheng, M.-Y., Huang, C.-J., Chang, Y.-K., Koester, D., Schac (2015) and the later work of Wang, K.-P., Frank, C., Hung, T.-M., & Schack, T (2023) show why training direction, task complexity, and control condition are central. The literature includes encouraging results and direct evidence of uncertainty. Both belong in a comprehensive account.

This monograph is a critical integrative synthesis of identifiable scientific studies, consensus documents, and systematic reviews. It is not a new registered systematic review or a newly computed meta-analysis. The source synthesis privileges original experimental reports for specific empirical claims, recognizes the distinct functions of mechanistic studies and reviews, and examines methodological alternatives wherever a favorable pooled effect could be interpreted too broadly.

Research Object and Conceptual Boundaries

The research object is an intervention class: sport-relevant psychophysiological self-regulation facilitated by respiratory practices, physiological feedback, or neural feedback. The objects being manipulated are operational variables such as respiration rate, interbeat interval variability, an electromyographic amplitude, electrodermal conductance, or activity in an EEG frequency band at specified electrode sites. These are observable proxies for physiological systems. The underlying processes are richer than the sensor displays.

It is necessary to distinguish an intervention from a general capacity. An athlete may have excellent autonomic recovery after exercise without ever engaging in biofeedback. Another may complete many biofeedback sessions without acquiring transferable self-regulation. Likewise, a baseline EEG signature correlated with expertise is evidence about brain–performance association, not evidence that rewarding that signature produces expertise. The methodological literature summarized by Ros, T., Enriquez-Geppert, S., Zotev, V., Young, K. D., Wood (2020) directly addresses how to separate the learned signal from psychological and behavioral outcomes.

Adjacent psychological skills interventions, mindfulness training, attentional cueing, guided imagery, and coaching feedback can alter arousal or concentration. Their overlap with the present subject creates a need to specify the instrumental or respiratory component that is claimed to be causal. A multicomponent psychological program may be valuable, but if its outcome is compared only with no intervention, the specific contribution of biofeedback is unidentified. The evidence for breathing alone also deserves its own treatment because a paced-respiration intervention can produce considerable physiological change without a feedback loop.

Sports differ in what physiological state is desirable. Fine aiming and preparation may benefit from reduced muscular interference or stable attention, while an explosive sprint may require rapid sympathetic and motor mobilization. The hypothesis that one universally optimal state enhances all tasks is therefore too coarse. The appropriate scientific unit is the relationship among person, phase of performance, task demand, trained variable, and measured outcome.

The primary explanatory questions are whether regulation is learnable, which pathways mediate its effects, whether improvements exceed comparators, how durable they are, and how far they generalize. The term sport performance is reserved for measured outcomes such as accuracy, time, execution quality, or actual competitive results; self-reported calmness is analyzed as a psychologically meaningful but separate endpoint.

History of Psychophysiological Self-Regulation

The historical trajectory of psychophysiological training joins experimental measurement, feedback-based learning, autonomic physiology, and applied performance psychology. Early behavioral approaches to biofeedback used instruments to return otherwise difficult-to-perceive bodily fluctuations to the individual. The underlying learning principle is that contingent information can help a person discriminate and sometimes regulate physiological activity. Subsequent research increasingly demanded mechanistic specificity: what precisely is being learned, which systems are changing, and how much apparent improvement arises from practice, expectation, or context?

Cardiorespiratory biofeedback developed through investigation of heart-rate oscillations and the cardiovascular baroreflex. In an influential experiment, Vaschillo, E., Lehrer, P., Rishe, N., & Konstantinov, M (2002) studied resonance in the cardiovascular system and the use of heart rate variability feedback to assess baroreflex functioning. This work helped establish why breathing near an individually responsive frequency can amplify rhythmic heart-rate changes. Later theoretical integration by Lehrer, P. M., & Gevirtz, R (2014) considered the baroreflex, respiratory–cardiovascular coupling, and possible afferent pathways linking bodily activity and central regulation.

Sport-oriented trials then asked a more demanding question: does training a cardiorespiratory variable change performance on a sport-specific task? Paul, M., & Garg, K (2012) tested HRV biofeedback in basketball players, while subsequent reviews examined the small and varied evidence base. The Jiménez Morgan, S., & Molina Mora, J. A (2017) found seven eligible reports and emphasized their limited sample sizes. This historical stage is important because early indications of practical benefit preceded the large, homogeneous, adequately controlled literature that would be needed for general sport-wide recommendations.

Neurofeedback developed through related ideas about operant modification of neural activity measured at the scalp. Sports applications became especially prominent where a clear, repeatable precision task could be performed before and after intervention. Cheng, M.-Y., Huang, C.-J., Chang, Y.-K., Koester, D., Schac (2015) investigated sensorimotor-rhythm training among preelite golfers in a randomized experiment. Later experiments manipulated Mu rhythm in novice golfers and investigated short-session sensorimotor-rhythm feedback in professional golfers, expanding the range of neural targets and athlete populations.

A new phase of research has been characterized by systematic attempts to appraise accumulated findings. Cheng, M.-Y., Yu, C.-L., An, X., Wang, L., Tsai, C.-L., Qi (2024) critically reviewed randomized EEG neurofeedback studies in sports. Yu, C.-L., Cheng, M.-Y., An, X., Chueh, T.-Y., Wu, J.-H., Wa (2025) synthesized sport-performance outcomes, while Zhang, X., Chang, Z., Zhao, S., & Ning, Z (2025) pooled biofeedback and neurofeedback across performance, cognition, and mental-health endpoints. Li, Z., Liu, H., Liu, C., Liu, F., & Liu, B (2026) focused on golf putting and included other training interventions alongside EEG neurofeedback. These are not interchangeable source populations, and their reported pooled statistics cannot be treated as independent confirmations of identical causal effects.

A Multilevel Model of Regulation and Performance

A coherent interpretation of this literature requires several analytically distinct outcomes. The first is sensor acquisition: obtaining a physiologically interpretable measurement under conditions in which the signal reflects the intended process. Electrodes, sensors, timing, and preprocessing determine whether an apparent change is a meaningful observation.

The second is online regulation: the individual changes the displayed signal while feedback is present. Real-time visual reward, auditory feedback, or a paced-breathing display can generate immediate changes through learning, deliberate strategies, changes in posture, or simple entrainment. The experiment must identify which of these mechanisms is responsible.

The third is retention: regulation can be reproduced after the display disappears and after time has elapsed. Immediate success during a visually guided task may be strongly dependent on external pacing. A follow-up without feedback can therefore provide better evidence of acquired internal control.

The fourth is near transfer: a trained response improves a standardized, untrained sport task compared with an appropriate control. For example, golfers might improve on a putting test or a basket of shots that differs from feedback training itself. The fifth is far transfer: improvements persist across variable situations, competitive pressure, fatigue, unfamiliar demands, and actual contests. A performance improvement measured immediately inside the laboratory is evidence for a narrower claim than lasting competitive enhancement.

The sixth level concerns functional value and cost. A small change in a physiological metric can matter in a precision sport if it reliably alters accuracy, whereas a substantial EEG change may be practically irrelevant if it consumes time, distracts the athlete, or fails to transfer. These levels form a research framework for interpreting studies, not a claim that every intervention progresses through them in a fixed sequence.

Researchers can design measurements explicitly around these levels. A pre-registered trial can include the physiological acquisition metric, a no-feedback regulation probe, a sport-specific primary endpoint, a retention assessment, and a competition follow-up. Such an architecture prevents researchers from replacing an unfavorable performance outcome with a favorable physiological proxy after data collection.

Autonomic Physiology in Athletic Contexts

The autonomic nervous system coordinates cardiovascular, respiratory, thermoregulatory, and other functions relevant to exercise, arousal, and recovery. Sympathetic and parasympathetic influences affect cardiac activity, but their consequences depend on context. During vigorous exercise, heart rate reflects metabolic demand, movement, hydration, environmental stress, and other factors. At rest, respiratory cycles and vagally mediated variations contribute more visibly to beat-to-beat differences. This physiology makes direct interpretation of any isolated index risky.

Autonomic functioning in athletes is shaped by training load, physical fitness, sleep, illness, medication, menstrual and hormonal context, nutrition, altitude, and psychological stress. A measured change after an intervention may reflect these conditions as well as training. Repeated measurements require comparable posture, time of day, respiratory conditions, recent exercise, and equipment quality. These are methodological prerequisites when the goal is to attribute an effect to psychological training.

Different sporting moments also call for different regulatory strategies. An athlete preparing for a stationary precision task can use a brief preperformance breathing routine. A distance runner near maximal effort cannot simply maintain laboratory-paced breathing at an arbitrary slow rate while sustaining the same physiological workload. Coaches and researchers should define the temporal window in which autonomic regulation is studied, including resting preparation, task execution, recovery, and overnight adaptation.

Autonomic metrics can help describe the system's response to a challenge, but they do not offer a direct numerical measure of courage, concentration, resilience, or sporting merit. The psychologist must connect bodily measurement to well-defined psychological constructs and behavioral outcomes rather than treating a device's simplified wellness score as an independently validated construct.

Heart Rate Variability and Its Interpretation

Heart rate variability, or HRV, describes variation in the time intervals between successive heartbeats. In well-controlled recordings, analysis commonly uses normal-to-normal intervals derived from identified sinus beats. The terms HRV, cardiac vagal regulation, and stress resilience refer to overlapping but distinct constructs. A particular HRV value depends on recording length, respiratory activity, heart rate, physiology, and analytic choices. The practical guide by Shaffer, F., & Ginsberg, J. P (2017) establishes the range of commonly used metrics and emphasizes their differing meanings.

Time-domain indices include the root mean square of successive differences, often called RMSSD, and the standard deviation of normal-to-normal intervals, SDNN. RMSSD is frequently interpreted as reflecting vagally mediated beat-to-beat variation under suitable recording conditions. SDNN summarizes variability across the measurement period and depends strongly on its duration. Frequency-domain measures separate power into bands whose physiological interpretations vary with measurement conditions. Low-frequency power during paced slow breathing cannot be reduced to a simple measure of sympathetic activity.

A major complication for performance studies is that slow breathing can shift spectral power substantially, especially around the resonance region. An intervention may produce an impressive frequency-domain increase even when that increase primarily reflects the timing of breathing. It is therefore essential to record respiration and avoid interpreting frequency-domain change alone as independent evidence of improved psychological control.

The review and recommendations of Laborde, S., Mosley, E., & Thayer, J. F (2017) underline the importance of transparent HRV acquisition and analysis. Signal processing choices, artifact removal, and context should be reported. Wearable optical sensors may perform adequately for some resting applications but can be more vulnerable to motion than appropriate electrocardiographic acquisition; the intended analysis determines whether a particular recording method is fit for purpose.

An athlete's goal in biofeedback is not necessarily to maximize HRV at all times. Physiological states must be interpreted in relation to task timing, training history, and functional outcome. A healthy athlete may show reduced vagally mediated HRV during intense exercise without any evidence that the intervention has failed. What matters is the specific operational hypothesis and its controlled evaluation.

Respiration, Baroreflex Resonance, and Paced Breathing

Respiration modifies cardiac timing through several interdependent pathways. Respiratory sinus arrhythmia describes the tendency for heart rate to vary with the breathing cycle, while the arterial baroreflex responds to blood pressure changes and participates in cardiovascular stabilization. Slow breathing can coordinate these oscillatory processes. The experiments of Vaschillo, E., Lehrer, P., Rishe, N., & Konstantinov, M (2002) and the integrative model of Lehrer, P. M., & Gevirtz, R (2014) provide physiological foundations for training near cardiovascular resonance.

In many adults, the relevant oscillatory region is near 0.1 Hz, corresponding roughly to six cycles per minute. This approximate frequency is a useful starting point, whereas an individually identified resonance rate is more precise. Breathing depth, comfort, blood pressure oscillations, and respiratory timing may modify the response. A slower rate also has costs: excessive depth can alter carbon dioxide balance, generate dizziness or discomfort, and change physiological indices for reasons other than improved emotion regulation.

A resonance-frequency protocol typically identifies the respiratory pattern at which heart-rate oscillations and relevant cardiovascular responses are suitably enhanced, then trains that pattern with instrumented information. The feedback may reveal the alignment of respiratory and cardiac waves, the amplitude of oscillations, or other selected features. Protocol details matter: a displayed coherence score computed by a proprietary algorithm cannot be assumed equivalent to a separately validated physiological variable.

For sport, the appropriate test is whether training such cardiorespiratory coordination produces a useful response during the relevant performance phase. A short period of paced breathing before a free throw might influence subjective readiness and attention. Its application during a maximal sprint would require a different hypothesis. Mechanistic plausibility does not establish which competition settings benefit.

The systematic review of slow-breathing research by Zaccaro, A., Piarulli, A., Laurino, M., Garbella, E., Menicu (2018) links respiratory control to diverse psychophysiological responses in healthy individuals, while also documenting questions about the mechanisms. General physiological findings from healthy participants should be regarded as foundational evidence, not as direct proof that an elite athlete's score will increase.

The Mechanisms of Cardiac Biofeedback

Heart-rate variability biofeedback operates through a combined informational and physiological pathway. The athlete receives an interpretable representation of cardiac timing, frequently alongside a breathing guide. Respiratory pacing helps generate a reproducible signal; feedback can reinforce the timing, amplitude, or consistency of that response. Repeated sessions may produce better discrimination of bodily activity, more efficient performance of the breathing procedure, and increased confidence in one's ability to regulate the response.

At least three mechanisms must be differentiated. Entrainment is an immediate alignment of physiological oscillations to an externally supplied rhythm. Learning is an acquired ability to reproduce a target response through repeated experience. Psychologically mediated transfer occurs when the learned response changes attention, emotion, action selection, or motor execution in a separate task. The first two can be measured physiologically, while the third requires behavioral and preferably competition-relevant endpoints.

The baroreflex model provides an important explanation for why slow breathing changes recorded cardiovascular oscillations. Lehrer, P. M., & Gevirtz, R (2014) also discuss possible broader neural pathways, including vagal afferent influences. Such theoretical accounts guide experiment construction; a study that observes improved shooting after breathing practice does not identify one exclusive neural mediator without additional measurements or experimental manipulation.

Expectancy and motivation can contribute to training effects. Receiving attentive coaching and objective-looking physiological information can make a session credible and structured. This can enhance adherence or perceived preparedness independently of the target signal. An active comparator receiving equivalent attention, explanation, training duration, and credible performance expectations is often needed to estimate the incremental benefit of the contingent biofeedback signal.

An especially revealing comparison uses paced breathing with genuine physiological feedback versus the same paced breathing procedure without contingent physiological information. If both groups improve to similar degrees, the training may be useful while the feedback display adds little under the tested conditions. The causal contribution of instrumentation must be evaluated instead of presumed.

Respiratory Training Without Instrumental Feedback

Voluntary regulation of respiratory pace is older and more accessible than contemporary wearable biofeedback. Respiratory routines can be learned using counting, auditory pacing, visual guides, or coaching. In controlled circumstances, such techniques change respiratory timing and associated cardiac dynamics even in the absence of physiological feedback.

The scientific issue is one of component efficacy. A biofeedback package often contains breathing instruction, focused attention, practice, education, self-monitoring, and feedback. Random assignment to that package versus usual practice estimates a bundled intervention effect. It does not determine whether access to heart-rate information was necessary for the observed result.

A factorial design can disentangle these components: for example, breathing training with and without contingent HRV feedback, and a comparison with an attention-matched intervention. The analysis should examine primary performance outcomes and whether the manipulation actually produced distinct physiological changes. This approach also allows researchers to investigate whether baseline interoceptive ability or experience modifies the added value of feedback.

Safety and practicality matter. Training slow breathing while seated under supervision differs from imposing a rigid rate during heavy exercise. Research should document comfort, respiratory symptoms, actual adherence, and reasons for discontinuation. A plausible general benefit during preparation should not be transformed into an instruction to suppress normal exercise-related ventilation.

Electromyographic and Electrodermal Biofeedback

Electromyographic biofeedback uses electrical activity recorded from muscles to make aspects of muscular activation available to the performer. In precision sports, a researcher might examine whether an athlete can reduce unwanted co-contraction or learn a reproducible pre-action muscular sequence. The relevant measure depends on electrode location, normalization, movement phase, skin preparation, and motion-related contamination. Lower recorded activation is not universally desirable: many sporting tasks require precisely timed muscular force.

Surface electromyography is particularly useful for investigating the distinction between unnecessary tension and task-essential activation. An apparent reduction in amplitude might reflect improved movement economy in one context and loss of necessary recruitment in another. To evaluate performance consequences, investigators need synchronized movement data and an objective or appropriately blinded task score.

Electrodermal biofeedback focuses on skin conductance, which is related to sympathetic sudomotor activity. It can provide an index of physiological arousal during anticipation, preparation, or response to a performance demand. Skin conductance is affected by temperature, sensor placement, contact quality, sweating, movement, and time-dependent adaptation; the resulting reading is not a direct measure of anxiety or emotional valence.

Arousal regulation may be especially relevant to tasks with a long preparatory interval, such as aiming or penalty execution. Yet very low arousal is not the objective for all sports. Investigators should test task-specific hypotheses and distinguish training-time arousal reduction from stable improvements in skill execution or competition performance.

The sport literature has also examined multimodal programs that combine HRV, muscle activity, and skin conductance. These are useful applied systems but introduce a causal identification problem: when a combined program produces improvement, the effective component and its interactions are unknown unless the study design separates them. The investigation of biofeedback in sport is therefore partly an exercise in experimental decomposition.

Thermal and Multimodal Physiological Feedback

Peripheral temperature biofeedback has historically been used as an indirect index of changes in peripheral vascular regulation. It is influenced by room temperature, local circulation, movement, and other factors, making careful controls necessary. A change in skin temperature during training may represent a real physiological response while remaining insufficient to demonstrate improved sports performance.

Multimodal feedback systems combine several data streams, sometimes including heart rate, breathing, electromyography, electrodermal activity, movement, and subjective ratings. The appeal is that psychophysiological preparation involves interacting systems. However, the more variables a platform combines, the greater the risk that a proprietary composite score will hide the very mechanisms the experiment needs to test.

An adequate multimodal protocol identifies the primary signal, feedback contingency, timing of each component, individualized calibration, pre-specified outcomes, and the handling of conflicting signals. A high level of respiratory stability may coexist with elevated muscular activity; a useful model does not automatically collapse these states into a single universally desirable number.

Multimodal systems should be evaluated against simple alternatives. If a complex feedback suite provides the same outcome as a brief and well-taught breathing routine, complexity requires justification by usability, adherence, subgroup benefit, or a clearly demonstrated performance advantage. Instrumentation is a means of creating scientifically interpretable interventions, not itself an indicator of training quality.

EEG Neurofeedback and Its Learning Architecture

Electroencephalographic neurofeedback records electrical potentials from scalp electrodes and converts a selected feature into contingent information. The target might be a relative or absolute power measure in a frequency band at specified locations, a ratio between bands, or an explicitly defined connectivity measure. Participants may learn to influence the feature through attentional strategies, bodily relaxation, imagery, or procedural adaptation.

The method combines signal measurement, learning theory, and experimental neuroscience. The feedback display rewards an operational feature of EEG, not the brain's entire functional state. A change in band power can have multiple causes; the research design must assess whether it arises from the intended neural source, changes in eye position or blinking, muscle tension, drowsiness, reference configuration, or signal-processing choices.

The Ros, T., Enriquez-Geppert, S., Zotev, V., Young, K. D., Wood (2020) recommends transparent reporting of the targeted neural variable, control conditions, blinding, learning success, behavioral endpoints, and other critical design features. In sporting research, these guidelines are particularly important because athletes may receive persuasive explanations of purported peak-performance brain states.

Learning itself may be gradual or inconsistent across individuals. A participant who increases the rewarded signal during a session may do so with a deliberate cognitive strategy that alters motor preparation differently from the same apparent signal change achieved through implicit learning. Reports of individual learning curves, not only mean pre/post values, are needed to understand response diversity.

The central question is functional specificity. If rewarding a sensorimotor rhythm leads to better putting, one plausible causal model is that the trained state facilitates action preparation. Alternative explanations include altered expectancy, general relaxation, additional practice, or changing attentional strategy. Separating these mechanisms requires both valid controls and measurements of the proposed mediators.

Alpha, Sensorimotor Rhythm, Mu, and Frontal Theta

EEG frequency labels describe spectral ranges, but their functions depend on electrode location, task, physiological state, and measurement choices. Alpha activity, commonly discussed in the approximately 8–13 Hz region, has been linked to many processes including sensory processing and attentional modulation. Its topography and task-related change determine interpretation. A global instruction to increase alpha as a universal performance strategy lacks adequate specificity.

Sensorimotor rhythm is often operationalized in a narrower band over sensorimotor scalp regions. Sports studies have rewarded increases in this activity during preparatory phases, particularly for skilled precision performance. The randomized golf investigation by Cheng, M.-Y., Huang, C.-J., Chang, Y.-K., Koester, D., Schac (2015) found increased sensorimotor-rhythm power and improved putting accuracy after eight training sessions relative to a control group. The study offers a targeted causal hypothesis, tempered by its small sample of sixteen preelite golfers.

The Mu rhythm, typically studied around central regions in an overlapping alpha-range band, has its own functional literature concerning action and motor processing. Wang, K.-P., Frank, C., Hung, T.-M., & Schack, T (2023) randomized thirty novice golfers to an increase-Mu, decrease-Mu, or sham condition and reported improved putting performance when training decreased Mu activity. This result is especially informative because it challenges the expectation that increasing power in a broadly related frequency region is always beneficial.

Frontal midline theta is frequently discussed in relation to cognitive control and sustained task engagement. In sport neurofeedback, however, assigning a simple meaning to theta power without specifying timing, location, and artifact control can produce misleading conclusions. Different tasks may require different modes of processing; the desired direction of change should be defended with task-specific evidence.

One lesson from these experiments is that the same superficial label, such as alpha training or brainwave optimization, can encompass distinct interventions. Comparisons and meta-analyses must preserve electrode location, spectral definition, reward direction, feedback method, dose, participant skill, and skill outcome. Without this information, an average effect merges physiological interventions that may operate in different directions.

Motor Automaticity and Psychomotor Efficiency

Psychomotor efficiency is a theoretical family of claims about the relationship between neural processing, practiced skill, and economy of action. Highly skilled performers may exhibit different patterns of cortical engagement than novices during preparation and execution. Such observations can motivate neurofeedback experiments, but an association between expertise and a particular brain signal does not prove that deliberately increasing the signal will create expertise.

Automaticity refers to skilled action that can be executed with reduced conscious monitoring of low-level components. It arises through extensive task-specific practice and depends on the organization of perception, decision, and motor control. An intervention might support an existing automatic routine by reducing interfering attention or unproductive muscular recruitment. It might also undermine learning if it encourages the performer to suppress processing needed for an unfamiliar task.

This distinction is relevant to golf putting and similar precision tasks because both practice and attentional strategies can alter execution. A study that includes a new neurofeedback protocol without equating deliberate skill practice can confound improved putting with ordinary motor learning. Sport-specific evidence therefore requires a comparator that receives enough analogous task exposure to separate neural training from practice effects.

The direction-specific Mu findings of Wang, K.-P., Frank, C., Hung, T.-M., & Schack, T (2023) further suggest that task complexity may modify the relationship between spectral power and performance. A signal that correlates with preparation for a relatively simple aiming task may have a different meaning in a complex coordinated action. Neural efficiency should therefore be operationalized as a testable mechanism rather than a general marketing label.

Studies can evaluate proposed mediators by recording EEG during the actual sport task, analyzing time windows before and during movement, and testing whether intervention-induced signal change statistically and experimentally explains subsequent performance change. Mediation analysis alone remains vulnerable to unmeasured confounding; stronger designs manipulate the putative mechanism and compare credible alternatives.

Attention, Anxiety, Arousal, and Preperformance States

Athletic preparation includes the regulation of attention, emotion, expected difficulty, and bodily readiness. A precision athlete may benefit from directing attention toward task-relevant information, while a team-sport athlete may need rapid scanning and flexible switching. Psychophysiological training can be hypothesized to influence these processes, but the relevant constructs need independent measurement.

Self-reported anxiety, stress, or calmness can provide meaningful psychological outcomes. They should be measured with instruments appropriate to the athlete population, timing, and research question. A reduction in self-reported tension can be valuable even when objective performance remains unchanged; the two results should be reported separately. Changes in a heart-rate metric should not be silently treated as psychological improvement.

Arousal is a multidimensional construct involving physiological activation, emotional appraisal, motivation, and task-directed action. A desirable state may include elevated cardiovascular activity and efficient attention. The simplistic expectation that more relaxation always improves performance conflicts with the functional demands of many athletic tasks.

Preperformance routines offer an ecologically plausible site for regulation training. Athletes can rehearse a breathing or attentional procedure before a defined action, and researchers can test its effects on repeatable outcomes under low and high pressure. Yet the athlete's usual routine may already be effective; adding a device or an extra conscious step may create interference. Retention and transfer experiments should accordingly include familiar routines as comparators.

The measurement of attentional effects also requires care. Laboratory reaction times, switching tasks, and subjective concentration ratings do not measure the same outcome as team communication, anticipation, or tactical decision making in competition. Where a neural intervention improves a cognitive test but not actual football performance, the result supports a narrower cognitive claim.

Sport-Specific Experimental Evidence for HRV Biofeedback

A notable early experiment is Paul, M., & Garg, K (2012) of basketball players. The investigation evaluated psychological and basketball-related performance measures following HRV biofeedback training. Its results contributed to the hypothesis that rehearsed physiological control might influence sport-specific skill, although its small sample and task context restrict generalization across sports and levels of competition.

The Jiménez Morgan, S., & Molina Mora, J. A (2017) identified seven eligible studies published through October 2016. Study sample sizes ranged from one to thirty participants; six reports described improvements in psychophysiological variables alongside sport-related improvement. The ratio six of seven is a count of favorable reports, not a pooled causal effect or evidence that nearly all athletes improve. Small studies are more vulnerable to imprecision and selective reporting, and the included sports and outcomes were heterogeneous.

A separate Pagaduan, J. C., Chen, Y.-S., Fell, J. W., & Wu, S. S. X (2020) examined six randomized studies involving 187 athletes and performance-related endpoints. Its synthesis described contrasting effects across tasks. Improvements in some basketball measures were not accompanied by uniform superiority in dance outcomes or running performance. Some comparisons used regular training, whereas others included placebo-like activities or alternative feedback conditions. These variations illustrate why the type of comparator changes the scientific meaning of a reported benefit.

The central limitation of this early evidence is not merely sample size. The interventions varied in schedule, breathing practice, feedback content, population, and simultaneous skill training. Sporting outcomes included measures of skill precision, reaction time, performance ratings, and timed exertion. A response on one endpoint need not predict another. The evidence supports research into specific applications more clearly than it supports an unqualified claim that HRV biofeedback enhances athletic performance generally.

A more persuasive modern experiment would predefine a sport-specific primary endpoint, recruit sufficient athletes, include a credible and time-matched comparator, report both intention-to-treat and adherence-sensitive analyses where appropriate, and repeat the assessment without feedback after a meaningful interval. Physiological acquisition data would demonstrate whether the intended training occurred, while objective performance data would address its functional value.

Sport-Specific Experimental Evidence for Neurofeedback

The randomized experiment by Cheng, M.-Y., Huang, C.-J., Chang, Y.-K., Koester, D., Schac (2015) involved sixteen preelite golfers assigned to sensorimotor-rhythm neurofeedback or a control condition. After eight intervention sessions, the trained group showed greater sensorimotor-rhythm power and better putting accuracy. Its importance lies in linking an explicitly trained EEG variable with a repeatable sport endpoint, while its small sample limits precision and prevents broad claims about other sports.

Wang, K.-P., Frank, C., Hung, T.-M., & Schack, T (2023) advanced the question by randomizing thirty novice golfers to increased-Mu, decreased-Mu, and sham feedback. The decreased-Mu condition was associated with improved motor control and putting performance. This comparison tests protocol direction rather than assuming that all rewarded sensorimotor-region changes are advantageous. It also draws attention to motor-skill complexity and the participant's stage of learning.

In another experiment, Wu, J.-H., Chueh, T.-Y., Yu, C.-L., Wang, K.-P., Kao, S.-C (2024) investigated a single sensorimotor-rhythm neurofeedback session with forty-four professional golfers using a crossover design. Participants performed sixty putts from ten feet, with neurofeedback and a no-training condition assessed in different orders. The report describes increased sensorimotor-rhythm power and improved putting outcomes under the intervention. A no-training comparator gives less protection against expectancy and structured attention than a credible sham session.

The study of Wu, J.-H., Tu, Y.-C., Chang, C.-Y., Chueh, T.-Y., Gentili, R (2024) included seventeen professional female golfers in a crossover design. It tested a short intervention in connection with a longer-range swing task and reported improved accuracy on at least one specified metric relative to no training. This broadens the range of athletic movements investigated, but the small sample, short-term design, and specialized sport context still require careful interpretation.

Together, these studies indicate that EEG neurofeedback can be experimentally associated with sport-skill improvement in certain specific conditions. They also identify what remains unresolved: whether the effect depends on the feedback's veridical contingency, whether learned control is retained, how outcomes change with pre-existing athlete routines, and how laboratory precision tasks predict real-world competitive decisions.

Evidence From Soccer, Golf, and Other Disciplines

Golf has become a frequent test bed for neurofeedback because putting and other defined swings allow repeated measurement of accuracy. This has a methodological advantage: investigators can register many trials and compare pre/post distributions. The disadvantage is that a performance gain in a repeatable laboratory drill may partly reflect repeated exposure and may have limited applicability to competition settings with crowd pressure, changing terrain, fatigue, and strategic tradeoffs.

In football or soccer, cognition and performance involve richer interactions among vision, attention, anticipation, movement, and team coordination. van Boxtel, G. J. M., Denissen, A. J. J. M., de Groot, J. A (2024) studied group-based alpha neurofeedback among elite soccer players, measuring cognitive variables including task switching and mental rotation. These are interesting outcomes within a professional athlete group. Improvement on such measures should not be translated into a demonstrated increase in match outcomes without a direct competition measure.

The study also illustrates the importance of financial and implementation context. Several investigators were affiliated with research or commercial organizations involved in neurofeedback technology. Relevant affiliations should be documented when assessing the independence and potential interests surrounding device research, without using those affiliations as a substitute for evaluating the design and results.

A controlled study by Rusciano, A., Corradini, G., & Stoianov, I (2017) evaluated a multimodal autonomic biofeedback program in twenty elite Italian soccer players, ten assigned to the intervention and ten to a control procedure. The authors reported improvements in selected physiological adaptation, visual attention, and injury-related measures. The small single-setting sample and the combination of several biofeedback components make it difficult to infer which specific mechanism was responsible and how robust any injury-prevention benefit would be in independent teams.

The diversity of sports limits simple transportability. Precision-based tasks, endurance events, combat sports, team tactical performance, and artistic judged sport differ in performance time scale, movement complexity, oxygen demand, perceptual load, and scoring. A biofeedback effect on a stationary aiming task has a narrower inference domain than an effect on multi-hour endurance performance or collective team success.

The question for each discipline should be explicit: which physiological variable is targeted, at what time relative to movement, to improve which objectively measured task? The answer must be tested within the discipline rather than borrowed from a superficially similar laboratory procedure.

What Systematic Reviews and Meta-Analyses Establish

Systematic reviews synthesize heterogeneous scientific evidence, but their conclusions depend on eligibility criteria, study quality, comparison conditions, and selected outcomes. The Cheng, M.-Y., Yu, C.-L., An, X., Wang, L., Tsai, C.-L., Qi (2024) evaluated randomized EEG neurofeedback research and emphasized inconsistent target selection and incomplete evidence that protocols reliably altered the relevant brain activity. This methodological finding matters as much as the direction of reported performance differences.

The Yu, C.-L., Cheng, M.-Y., An, X., Chueh, T.-Y., Wu, J.-H., Wa (2025) assembled a broader sport-neurofeedback literature and reported a favorable average effect on sport performance. Such a synthesis helps identify patterns across studies, but it combines interventions, sports, experience levels, and outcome instruments. Its overall estimate cannot describe the expected change for a particular professional athlete performing a different skill with a different EEG protocol.

The Bayesian review by Zhang, X., Chang, Z., Zhao, S., & Ning, Z (2025) included forty-one randomized studies across biofeedback and neurofeedback. It reported a pooled standardized effect for athletic performance of 0.88, with a 95% credible interval from 0.69 to 1.05. Mental-health and cognitive-performance outcomes also had favorable pooled estimates. Crucially, the authors' GRADE assessment rated certainty for athletic and cognitive performance as very low and for mental-health outcomes as low. This is a central interpretive constraint: a statistically positive pooled estimate is compatible with considerable uncertainty about the true causal effect across settings.

The study also reported meaningful differences by intervention, outcomes, and populations. The comparatively strong combined average should not be taken as a validated prescription for session frequency or duration, because subgroup estimates are often based on uneven and overlapping evidence. Different reviews may share many of the same primary experiments; adding their estimates together would amount to double counting.

The Li, Z., Liu, H., Liu, C., Liu, F., & Liu, B (2026) included twelve randomized controlled trials and four hundred twenty-four participants concerning golf putting. Its overall standardized mean difference was 1.33, with a 95% confidence interval from 0.36 to 2.29, but statistical heterogeneity was extremely high (I² = 94%). The eligibility criteria covered EEG neurofeedback and other training interventions. The overall result therefore cannot be attributed exclusively to EEG neurofeedback. The same analysis found protocol-specific differences, including adverse directions for some neural targets, reinforcing the need for specificity.

The combined state of the evidence is promising but heterogeneous. Some well-defined sport tasks respond favorably in controlled experiments, while important portions of the literature remain small, dependent on limited comparators, or focused on near transfer. The strongest defensible conclusion concerns targeted hypotheses and the need for better causal evaluation rather than universal performance enhancement.

Control Conditions, Expectancy, and Placebo Mechanisms

Any feedback study must establish what its comparison controls. A no-training control compares a full training experience with its absence and can reveal whether the package provides benefit. It does not isolate the effect of contingent physiological information. An attention-matched active control may account for instructor time and engagement, while a credible sham-feedback condition can test whether feedback linked to the athlete's own physiological signal matters beyond the experience of training.

Sham feedback has technical complications. It may be generated from prerecorded signals, simulated data, or another physiological variable. Each choice affects believability and potential learning. A completely flat or implausible display may reveal the condition and undermine blinding. A sham protocol should be credible enough to balance expectation while avoiding a hidden active intervention that also changes the targeted process.

For EEG neurofeedback, participants and staff may infer condition from subjective success, interface behavior, or signal availability. The Ros, T., Enriquez-Geppert, S., Zotev, V., Young, K. D., Wood (2020) makes the control condition an essential component of research reporting and advocates measures of credibility, learning, and relevant outcomes. Blinding integrity should be checked empirically when feasible rather than merely asserted.

Repeated sport trials create further complications. Participants improve through practice, become more comfortable with apparatus, and adjust motor strategies after feedback. A crossover study must consider sequence effects and carryover; a randomized parallel study should equalize practice exposure. If the primary outcome is a score that fluctuates with wind, course conditions, or fatigue, controlled repetition and transparent scoring are necessary.

Placebo-related processes and general psychological benefits are real candidate components of intervention effects. The scientific problem is to estimate their contribution and distinguish them from the effect of the specific trained physiological variable. This distinction enables better interventions: when breathing instruction alone works as well as an expensive device, the simpler intervention may be more accessible.

Signal Validity, Artifact, and Measurement Error

Physiological feedback depends on measurement fidelity. An electrocardiogram can provide beat timing from electrical cardiac events, while photoplethysmography estimates pulse timing through optical sensing. These signals are related but differ under movement and vascular changes. Researchers should state which sensor produced the analyzed variable and how artifacts were detected and corrected.

HRV analysis requires transparent handling of ectopic beats, missing intervals, motion artifacts, filtering, and recording duration. An apparent intervention effect can arise if one group has cleaner recordings or if software uses different correction rules across sessions. Respiration should be measured where a respiratory mechanism is central, and the analyst should avoid equating a respiration-driven spectral increase with comprehensive autonomic improvement.

EEG introduces its own challenges. Eye movement, blinking, facial muscle activity, neck tension, electrode impedance, reference scheme, and body motion can all influence apparent band power. Sport-specific movement often makes artifact management especially difficult. A neurofeedback protocol must explain whether it rewards valid neural activity or a proxy contaminated by behavior that the athlete learns to manipulate.

Preprocessing procedures should be determined before examining group outcomes. Reporting includes sampling rate, montage, reference, filter boundaries, epoch selection, artifact rejection, spectral estimation, training thresholds, and percentage of valid feedback time. Without these details, independent reproduction and meaningful synthesis become difficult.

Measurement validity also extends to performance. A proprietary golf simulator score, a coach's unblinded rating, a reaction-time task, and an official competition result measure different constructs. The most persuasive study uses reliable endpoints that correspond to the stated research question and explains their relationship to actual athletic success.

Experimental Design and Causal Identification

Randomized assignment strengthens causal inference by balancing measured and unmeasured pre-intervention characteristics on average, but small randomized samples can still differ appreciably. A trial's validity depends on allocation concealment, comparator credibility, adherence, attrition, outcome selection, and analytic transparency. Researchers should specify the primary outcome and expected direction before data collection.

Sport trials often recruit convenience samples of athletes from one club, academy, or training cohort. This improves practical feasibility but can create clustering and limited generalization. If teammates receive the same intervention or instruction from the same coach, observations may not be independent. Cluster-randomized or hierarchical analysis may be required, especially in team-based programs.

Power and precision deserve particular attention. Several frequently cited neurofeedback experiments involve fewer than fifty participants, and the early HRV evidence contains even smaller studies. A statistically significant result from a small trial can coexist with a wide interval of plausible effects and heightened selection risk. Investigators should plan sample sizes using realistic effect assumptions, identify minimally meaningful sport differences, and report uncertainty intervals.

For repeated trials within athletes, the analysis must distinguish the number of performances from the number of independently randomized people. Forty putts per athlete produce useful information about that athlete's scoring distribution, but they do not turn a sample of seventeen golfers into hundreds of independent participants. Multilevel models can account for nested outcomes, repeated time points, and athlete-level heterogeneity.

Causal identification can be strengthened further through pre-specified mediator assessment and experimental comparisons that isolate feedback contingency. A well-designed study may have separate endpoints for neurophysiological learning, self-regulation without feedback, task-specific improvement, and retained transfer. Negative outcomes at any step refine the mechanistic model rather than rendering the study uninformative.

Retention, Generalization, and Competition Transfer

Retention requires performance or physiological regulation to persist after training stops. A same-day improvement might reflect temporary attentional orientation, familiarity with the apparatus, or immediate expectation. Follow-up tests after days, weeks, or an appropriate competitive interval are needed to establish durable learning. These tests should report whether the feedback device is still present.

Near transfer occurs when a trained state benefits a related task under changed conditions. A golfer who practices putting with neurofeedback could be tested on unfamiliar distances, slopes, and environmental conditions without the feedback. Far transfer would require evidence that the intervention changes competitive performance, strategic behavior, or enduring performance quality outside the training setting.

Actual sporting outcomes have practical advantages and inferential complications. Competitive scores reflect opposition, environment, tactics, teammate coordination, injury, and chance. An improvement after training may have many causes. Researchers can improve identification using repeated competition observations, comparison athletes, pre-specified windows, and sport-appropriate statistical models.

Retention and transfer may vary with athlete expertise. A novice can improve rapidly through ordinary instruction, making neurofeedback's incremental contribution harder to separate. An expert may have less room to improve on a familiar task, but a small reliable change could be consequential. These different baseline conditions require pre-specified subgroup hypotheses and careful estimation rather than post hoc claims.

A credible performance intervention should eventually demonstrate value under the constraints in which it is meant to be used. Until such data exist, a laboratory result should be described as laboratory performance improvement or near transfer, with the distance to competition made explicit.

Individual Differences and Athlete Populations

Athletes vary in baseline physiology, regulation strategies, attentional control, motor learning history, and familiarity with performance routines. Individual differences can modify both the ease of feedback learning and its practical effects. The same EEG target may be favorable for one task or expertise level and unhelpful for another.

Physiological calibration is particularly relevant for HRV biofeedback. Respiratory resonance and comfort can differ between individuals, and a fixed pacing rate may be inappropriate for some participants. At the same time, highly individualized protocols can reduce replicability unless the calibration procedure is completely reported.

Age, disability, sport classification, training status, sex-related physiology, fatigue, and medication may influence both recording quality and response. Studies should include diverse athlete populations or explicitly limit their claims to the participants tested. When samples are predominantly university athletes, results should not automatically be generalized to children, para athletes, older competitors, or elite professional teams.

There is also an important distinction between selection and treatment. Athletes with certain baseline physiological characteristics may perform better, but using those characteristics to identify likely responders does not establish the effect of biofeedback. Prediction models need independent validation and comparisons with simple baseline predictors before being used to allocate training resources.

Research should report individual trajectories, nonresponse, withdrawal, and adverse experiences. An average favorable effect can conceal a group for whom the intervention creates extra cognitive burden or disrupts an established action routine. Personalized performance psychology is justified through reliable evidence of variation, not merely through a device's capacity to present individualized scores.

Safety, Ethical Governance, and Data Protection

Respiratory and feedback-based interventions are often presented as low risk, but good scientific practice includes active adverse-event monitoring. Excessively deep or uncomfortable breathing can produce dizziness, breathlessness, tingling, or distress. Feedback procedures may amplify performance anxiety, encourage compulsive monitoring, or distract from the sporting task. Individual physiology and medical conditions can alter the safety profile of training.

EEG and wearable measurements raise further issues of data quality, privacy, and interpretation. Raw neural and cardiac signals can be sensitive personal information. Coaches and commercial providers should specify who has access, why the data are collected, how long they are retained, and whether derived scores will be used in team selection, monitoring, or discipline. Consent must be meaningful in settings where athletes are dependent on coaches or institutions.

An athlete may experience pressure to disclose a stress score or submit to a device-driven readiness evaluation. Such uses require an independent ethical analysis because their consequences extend beyond the intervention's efficacy. A physiological score should not be treated as definitive evidence of emotional stability, effort, honesty, or future performance.

Device developers should document analytic methods and the provenance of any training recommendations. Researchers should report commercial affiliations and sponsorship when relevant. The existence of a conflict does not determine a result's validity, but readers need information to assess incentives, analytic independence, and transparency.

Applications with potential medical implications require professional oversight appropriate to their purpose. A sport psychologist studying performance regulation should avoid claiming diagnosis or treatment benefits from an apparatus unless the relevant clinical evidence and scope of practice support that claim. Health-related responses and performance optimization remain distinct outcome categories.

Wearables, Adaptive Feedback, and the Artificial Era

The growth of wearable sensors and algorithmic coaching expands access to psychophysiological data outside the laboratory. Heartbeat estimates, respiratory measures, electrodermal signals, movement features, and in some cases portable EEG can be processed continuously or during repeated training sessions. Such systems create opportunities for ecologically richer longitudinal observation.

An adaptive algorithm may personalize breathing pace, feedback threshold, session duration, or display based on incoming signals. This can improve usability and generate new experimental hypotheses. It also complicates causal inference: when the feedback rule changes during training, the intervention is a dynamic policy rather than a fixed stimulus. Researchers must record the decision rules and their version history to determine what was actually administered.

Artificial systems can analyze patterns beyond the immediate capacity of a human coach, but the interpretability and reliability of those patterns remain empirical questions. An algorithmic readiness score trained on historical athlete data may reproduce selection bias, measurement artifacts, or spurious correlations. Calibration against meaningful outcomes and independent athlete populations is required before claiming that the score measures a stable psychological property.

The Artificial Era provides a research horizon for examining human–artificial performance partnerships. In this context, an instrumented training environment can be studied as a coupled system of athlete, sensor, computational policy, and feedback interface. Its relevant psychological consequences include trust, perceived control, attention allocation, adherence, cognitive offloading, and the distribution of decisions between person and system. These are proposals for operational research, not empirical conclusions about the intrinsic capacities of a particular algorithm.

Future interventions may combine transparent rule-based feedback with machine-learning adaptations, but comparison against simple validated practices remains essential. Scientific value depends on reproducible improvement in human outcomes and accountable handling of physiological information, not on the complexity of the computational platform.

Critical Scientific Discussion

Three findings emerge from the evidence. First, physiological variables can be altered by intentional practice and contingent feedback under defined conditions. Respiratory pacing influences cardiorespiratory oscillations, while some EEG neurofeedback experiments report changes in the trained spectral variable. These effects provide plausible foundations for self-regulation training.

Second, certain controlled sport studies report improvements in selected motor outcomes, notably golf putting and particular basketball measures. Their relevance is strongest for the studied participants, tasks, timing, and protocols. The evidence is less developed for complex team performance, competitive endurance, untrained sporting situations, and long-term outcomes.

Third, the strength of the evidence depends heavily on comparator and measurement quality. The Zhang, X., Chang, Z., Zhao, S., & Ning, Z (2025) reports positive pooled effects while grading the certainty of the athletic-performance evidence as very low. This combination should be reported plainly because statistical significance and evidentiary certainty answer different questions.

Several recurring interpretive errors can be avoided. A rise in HRV during slow breathing is not by itself proof of reduced psychological distress. A favorable EEG change is not direct proof of enhanced cortical efficiency. An improvement in a repeated precision drill is not direct proof of better competition results. A review combining different feedback and nonfeedback interventions cannot establish the efficacy of any single neural target. A favorable short-term estimate from a no-training comparison cannot isolate signal-specific learning.

The literature therefore supports a conditional conception of psychophysiological regulation. Effects arise within identifiable relationships among physiological target, training procedure, learned strategy, task demand, prior expertise, and feedback credibility. Stronger claims require direct, preregistered testing across these relationships and transparent publication of null and adverse outcomes.

This formulation is scientifically useful because it preserves the real promise of the interventions while preventing expansion beyond their evidence. The ultimate standard is not whether a feedback trace looks improved but whether athletes acquire a robust, comprehensible, safe, and transferable capability that measurably contributes to their sporting goals.

A Research Agenda for Reliable Sports Evidence

The next generation of sport neurofeedback trials should start with explicit mechanisms. Each intervention must identify the anatomical or physiological target, justify the direction of training, specify whether the targeted process is proposed to mediate performance, and report the method for verifying target engagement. Training thresholds and adaptation rules need to be documented with sufficient detail for independent reproduction.

Control groups should match expectations, session contact, device presentation, and sport-practice exposure wherever feasible. For HRV interventions, paced breathing without contingent HRV feedback is a particularly important comparator. For EEG interventions, credible sham feedback and blinded outcome assessment can help isolate neural-contingency effects. Where sham feedback is impractical, researchers must describe the remaining causal limitations.

Performance measures should reflect the intended domain. Standardized sport tests need reliability estimates, prespecified scoring, and consideration of meaningful performance thresholds. Researchers should report individual and group effects with confidence or credible intervals, avoid conflating trials with independent participants, and account for repeated observations.

Retention tests should occur without real-time feedback. Transfer studies should vary distance, pressure, fatigue, tactical context, and other sport-specific demands. Longer follow-up can establish whether athletes maintain the intervention strategy and whether its effect survives the complexity of competitive life.

Pre-registration, adequate power, transparent statistical code, accessible de-identified outcome summaries where ethically permissible, and publication of null findings would improve cumulative inference. Meta-analysts should distinguish EEG protocol targets, separate sham-controlled estimates from no-treatment comparisons, and guard against overlapping studies being counted repeatedly across reviews.

Research on wearable and algorithmic systems should add evaluation of privacy protections, model drift, device malfunction, athlete autonomy, and fairness across different populations. Its strongest contribution will be to show when physiological feedback adds meaningful, reproducible value over simpler procedures, and when it does not.

Conclusion

Psychophysiological regulation in sport is a scientifically coherent intervention domain joining autonomic physiology, respiratory control, contingent feedback, neural measurement, learning, and skilled motor performance. Its central promise lies in helping athletes gain access to aspects of bodily and neural activity that are otherwise difficult to observe, potentially establishing regulation procedures that contribute to defined performance outcomes.

The cardiovascular literature explains why controlled breathing and baroreflex resonance alter recorded heart-rate dynamics. Sport-specific HRV biofeedback trials provide selected favorable findings, but their early evidence base is small and heterogeneous. EEG neurofeedback research offers more direct demonstrations of target-related changes and improvements in certain precision-motor tasks, notably golf. The direction and utility of trained neural activity depend on the protocol and the task; results cannot be generalized from an undifferentiated concept of brainwave optimization.

Modern systematic reviews report encouraging average effects while recognizing substantial variation in interventions and outcomes. The low or very low certainty assigned to important pooled outcomes establishes the present boundary of inference. A valid science of performance enhancement must demonstrate measurement fidelity, trainability, retained self-regulation, behavioral transfer, and competition relevance as separate empirical achievements.

The most productive future research will compare biofeedback with credible alternatives, explicitly test hypothesized mechanisms, document individual learning and nonresponse, and preserve athlete well-being and control over physiological data. Under these conditions, psychophysiological training can develop into a more reliable and accountable component of sport psychology.

References

  • Cheng, M.-Y., Huang, C.-J., Chang, Y.-K., Koester, D., Schack, T., & Hung, T.-M. (2015). Sensorimotor rhythm neurofeedback enhances golf putting performance. Journal of Sport and Exercise Psychology, 37(6), 626–636. https://doi.org/10.1123/jsep.2015-0166
  • Cheng, M.-Y., Yu, C.-L., An, X., Wang, L., Tsai, C.-L., Qi, F., & Wang, K.-P. (2024). Evaluating EEG neurofeedback in sport psychology: A systematic review of RCT studies for insights into mechanisms and performance improvement. Frontiers in Psychology, 15, 1331997. https://doi.org/10.3389/fpsyg.2024.1331997
  • Jiménez Morgan, S., & Molina Mora, J. A. (2017). Effect of heart rate variability biofeedback on sport performance, a systematic review. Applied Psychophysiology and Biofeedback, 42(3), 235–245. https://doi.org/10.1007/s10484-017-9364-2
  • Laborde, S., Mosley, E., & Thayer, J. F. (2017). Heart rate variability and cardiac vagal tone in psychophysiological research—Recommendations for experiment planning, data analysis, and data reporting. Frontiers in Psychology, 8, 213. https://doi.org/10.3389/fpsyg.2017.00213
  • Lehrer, P. M., & Gevirtz, R. (2014). Heart rate variability biofeedback: How and why does it work? Frontiers in Psychology, 5, 756. https://doi.org/10.3389/fpsyg.2014.00756
  • Li, Z., Liu, H., Liu, C., Liu, F., & Liu, B. (2026). Effects of EEG neurofeedback and training interventions on golf putting performance: A systematic review and meta-analysis. Frontiers in Psychology, 17, 1736851. https://doi.org/10.3389/fpsyg.2026.1736851
  • Pagaduan, J. C., Chen, Y.-S., Fell, J. W., & Wu, S. S. X. (2020). Can heart rate variability biofeedback improve athletic performance? A systematic review. Journal of Human Kinetics, 73, 103–114. https://doi.org/10.2478/hukin-2020-0004
  • Paul, M., & Garg, K. (2012). The effect of heart rate variability biofeedback on performance psychology of basketball players. Applied Psychophysiology and Biofeedback, 37(2), 131–144. https://doi.org/10.1007/s10484-012-9185-2
  • Ros, T., Enriquez-Geppert, S., Zotev, V., Young, K. D., Wood, G., Whitfield-Gabrieli, S., Wan, F., Vuilleumier, P., Vialatte, F., Van De Ville, D., Todder, D., Surmeli, T., Sulzer, J. S., Strehl, U., Sterman, M. B., Steiner, N. J., Sorger, B., Soekadar, S., Sitaram, R.,. Thibault, R. T. (2020). Consensus on the reporting and experimental design of clinical and cognitive-behavioural neurofeedback studies (CRED-nf checklist). Brain, 143(6), 1674–1685. https://doi.org/10.1093/brain/awaa009
  • Rusciano, A., Corradini, G., & Stoianov, I. (2017). Neuroplus biofeedback improves attention, resilience, and injury prevention in elite soccer players. Psychophysiology, 54(6), 916–926. https://doi.org/10.1111/psyp.12847
  • Shaffer, F., & Ginsberg, J. P. (2017). An overview of heart rate variability metrics and norms. Frontiers in Public Health, 5, 258. https://doi.org/10.3389/fpubh.2017.00258
  • van Boxtel, G. J. M., Denissen, A. J. J. M., de Groot, J. A., Neleman, M. S., Vellema, J., & Hart de Ruijter, E. M. (2024). Alpha neurofeedback training in elite soccer players trained in groups. Applied Psychophysiology and Biofeedback, 49(4), 589–602. https://doi.org/10.1007/s10484-024-09654-1
  • Vaschillo, E., Lehrer, P., Rishe, N., & Konstantinov, M. (2002). Heart rate variability biofeedback as a method for assessing baroreflex function: A preliminary study of resonance in the cardiovascular system. Applied Psychophysiology and Biofeedback, 27(1), 1–27. https://doi.org/10.1023/a:1014587304314
  • Wang, K.-P., Frank, C., Hung, T.-M., & Schack, T. (2023). Neurofeedback training: Decreases in Mu rhythm lead to improved motor performance in complex visuomotor skills. Current Psychology, 42, 20860–20871. https://doi.org/10.1007/s12144-022-03190-z
  • Wu, J.-H., Chueh, T.-Y., Yu, C.-L., Wang, K.-P., Kao, S.-C., Gentili, R. J., Hatfield, B. D., & Hung, T.-M. (2024). Effect of a single session of sensorimotor rhythm neurofeedback training on the putting performance of professional golfers. Scandinavian Journal of Medicine & Science in Sports, 34(1), e14540. https://doi.org/10.1111/sms.14540
  • Wu, J.-H., Tu, Y.-C., Chang, C.-Y., Chueh, T.-Y., Gentili, R. J., Hatfield, B. D., & Hung, T.-M. (2024). A single session of sensorimotor rhythm neurofeedback enhances long-game performance in professional golfers. Biological Psychology, 192, 108844. https://doi.org/10.1016/j.biopsycho.2024.108844
  • Yu, C.-L., Cheng, M.-Y., An, X., Chueh, T.-Y., Wu, J.-H., Wang, K.-P., & Hung, T.-M. (2025). The effect of EEG neurofeedback training on sport performance: A systematic review and meta-analysis. Scandinavian Journal of Medicine & Science in Sports, 35(5), e70055. https://doi.org/10.1111/sms.70055
  • Zaccaro, A., Piarulli, A., Laurino, M., Garbella, E., Menicucci, D., Neri, B., & Gemignani, A. (2018). How breath-control can change your life: A systematic review on psycho-physiological correlates of slow breathing. Frontiers in Human Neuroscience, 12, 353. https://doi.org/10.3389/fnhum.2018.00353
  • Zhang, X., Chang, Z., Zhao, S., & Ning, Z. (2025). The effects of biofeedback training on athletes' mental health and performance: A systematic review and Bayesian meta-analysis. Frontiers in Psychology, 16, 1662868. https://doi.org/10.3389/fpsyg.2025.1662868