RESEARCH DOSSIER
Mental Fatigue, Pacing, and Cognitive Regulation in Endurance Performance
Abstract. Endurance performance emerges from ongoing regulation of work rate under physiological constraints, changing perceptions of effort, task expectations, and decisions about how much effort to invest at each moment. This fundamental research monograph examines mental fatigue induced by sustained cognitive activity, its possible effects on endurance, the perceptual and decisional mechanisms of pacing, and evidence for cognitive and behavioral regulation. It reconstructs the historical development of psychophysics, anticipatory pacing, psychobiological endurance models, cognitive-control theories, and experimental mental-fatigue protocols. Primary randomized crossover experiments, preregistered replications, quantitative reviews, time-trial analyses, and intervention research are evaluated with attention to outcome definition, physiological confounding, construct validity, blinding, task familiarity, publication bias, and the distinction between perceived effort and physical capacity. The evidence demonstrates that cognitive demands can alter subjective fatigue and may compromise selected endurance outcomes, while rigorous null findings and bias-sensitive syntheses preclude a uniform impairment claim. Pacing is best understood as temporally extended behavioral regulation informed by bodily feedback, prospective knowledge, environmental conditions, and learned decision policies. Current brain endurance training evidence is promising for performance under mentally fatigued testing conditions but methodologically heterogeneous. The synthesis identifies falsifiable research priorities for ecological measurement, mechanisms, individual differences, and applied interventions. This is a critical integrative narrative synthesis of identifiable published studies through November 2026, not an original experiment, registered systematic review, or meta-analysis.
Keywords: mental fatigue; cognitive fatigue; endurance performance; pacing; self-paced exercise; perceived exertion; effort perception; cognitive control; response inhibition; psychobiological model; anticipatory regulation; time trial; fatigue induction; attentional focus; self-talk; brain endurance training; decision making; psychophysiology; replication; sport psychology; exercise regulation.
The Scientific Problem of Endurance Regulation
Endurance performance is a temporally extended problem of allocating effort. A runner, cyclist, rower, swimmer, or other endurance athlete must repeatedly determine a sustainable work rate while interpreting bodily sensations, the anticipated duration or distance remaining, environmental conditions, competitive information, and the consequences of changing pace. The selected intensity affects subsequent physiological state and thus changes the options available to the athlete later in the effort. Performance therefore arises through an ongoing reciprocal relation between physical capacities and behavioral decisions.
Mental fatigue adds a distinct question. Can preceding cognitively demanding activity change the pace an athlete selects, the effort they perceive, their willingness to maintain that effort, or the time they can tolerate a fixed workload? A study demonstrating a change in felt fatigue need not demonstrate reduced physical performance; similarly, a change in completion time may reflect the interaction of motivation, expectancy, sensory information, physiological limitation, and pacing decisions. Marcora, S. M., Staiano, W., & Manning, V (2009) established an influential experimental result, while Holgado, D., Troya, E., Perales, J. C., Vadillo, M. A., & Sa (2021) provided a preregistered replication with a different outcome. These findings jointly define the central empirical problem.
This research object belongs primarily to Sport and Performance Psychology and intersects biological psychology, health psychology, behavioral neuroscience, and human factors. Its boundary is athletic endurance and its cognitive regulation. Clinical fatigue syndromes, chronic illness pacing, and workload management in industrial systems require distinct measurement and intervention literatures. The guiding questions are whether fatigue induction changes actual endurance behavior, which mechanisms can account for observed changes, how athletes choose pace, when interventions work, and which findings withstand independent replication.
Definitions, Constructs, and Units of Analysis
Mental fatigue in exercise research is commonly operationalized as a transient state emerging after sustained demanding cognitive activity. It may include subjective mental tiredness, increased aversion to continued cognitive work, changes in attention or task performance, and an altered valuation of further effort. No single questionnaire item establishes the complete state. Researchers frequently use prolonged incongruent Stroop tasks, response inhibition, vigilance, or sustained attention tasks as manipulations, accompanied by subjective ratings and sometimes behavioral or neurophysiological indices. Goodman, S. P. J., Collins, B., Shorter, K., Moreland, A. T (2025) reviewed the methodological challenge of determining which indices reflect fatigue induction.
Pacing is the time-varying selection and adjustment of speed or mechanical power relative to a task goal. The relevant unit is often an entire self-paced trial, but the mechanism must be analyzed at finer temporal resolution: an acceleration, a delay in responding to discomfort, or a change in target intensity. Endurance performance itself should be operationalized separately, such as finishing time for fixed distance, distance covered in fixed time, total work during a prescribed test, or time to exhaustion at constant intensity. These outcomes are related but not interchangeable. A pacing decision directly affects a time-trial trajectory, whereas time-to-exhaustion removes much of the choice over moment-to-moment workload.
Perception of effort denotes an individual's conscious evaluation of how hard the activity feels. It must be distinguished analytically from muscle pain, thermal discomfort, dyspnea, emotional displeasure, subjective mental fatigue, and global session exertion. These sensations may covary while indexing different inputs. The psychophysical approach developed by Borg, G. A. V (1982) supplies tools for scaled effort reports, but each scale needs standardized instructions and a clearly specified referent. Cognitive regulation refers here to attentional allocation, goal maintenance, monitoring, anticipation, appraisal, motivational self-instruction, and other operations involved in sustaining or changing endurance behavior.
Historical Foundations of Effort Psychophysics
Scientific interest in exercise regulation preceded the contemporary cognitive-fatigue literature. Early exercise science investigated oxygen uptake, metabolic pathways, cardiorespiratory limits, and muscular function. These analyses established important constraints on possible performance. The psychophysical tradition added a method for studying how those constraints are experienced by the acting person and how reported intensity relates to work rate and duration. Borg, G. A. V (1982) clarified the use of category and ratio-like methods in perceived exertion research. In endurance experiments this is an important transition: perceived effort becomes a measurable outcome that can be compared across conditions.
Later pacing research shifted the unit of explanation from the final point of exhaustion toward a series of adjustments preceding it. Tucker, R (2009) developed an anticipatory, perception-based account in which expectations about task duration and accumulated strain contribute to intensity regulation. de Koning, J. J., Foster, C., Bakkum, A., Kloppenburg, S., T (2011) examined pacing decisions during simulated competitions using momentary perceived exertion and remaining distance. Their hazard-score formulation is a testable way to describe when pace changes become likely; it is a model of decision-relevant information, not a universal physiological law.
Concurrently, experimental psychology investigated sustained attention, response inhibition, mental effort, and sequential-task effects. These traditions converged when researchers began examining how a prior cognitive task altered later exercise. The 2009 experiment by Marcora and colleagues became a pivotal demonstration; subsequent studies extended the question to self-paced cycling, skilled athletes, heat, and training. The history is best understood as an accumulation of distinct explanatory traditions—exercise physiology, psychophysics, decision science, and cognitive control—whose constructs require careful alignment.
Exercise Physiology as the Constraint Structure
Work rate during prolonged exercise affects oxygen demand, substrate utilization, metabolite production, body temperature, ventilation, and recruitment of muscular systems. The physiological feasibility of maintaining a particular intensity changes over time. Hydration, nutrition, altitude, environmental heat, event duration, movement economy, training status, and illness can modify the relationship between selected pace and physiological strain. Any psychological model must remain compatible with these constraints.
An athlete who reduces pace during a race may be making a prospective decision to prevent an unsustainable accumulation of strain; the same observed deceleration might also follow actual loss of force-generating capacity, inadequate fuel availability, thermal stress, or a combination of these. Conversely, a late acceleration indicates that work rate has been regulated rather than maintained at a single maximal instantaneous output. None of these observations alone establishes that cognitive regulation is the sole cause of the pacing profile.
Mental-fatigue experiments can isolate part of the problem by manipulating cognition while monitoring heart rate, oxygen consumption, lactate, or neuromuscular outcomes. Yet unchanged peripheral measures in two conditions do not prove the absence of all physiological pathways, because monitoring is selective and measurement sensitivity varies. Pageaux, B., Marcora, S. M., & Lepers, R (2013) directly investigated whether prolonged mental exertion altered knee-extensor neuromuscular function, an important test of simple muscular-depletion explanations.
The explanatory aim is therefore conditional: identify how cognitive state alters willingness or ability to sustain a workload within the same physiological task, while estimating alternative physical mechanisms. A credible account integrates physiology and psychology at the level of observed work-rate regulation without assuming that physiological measures are merely background variables.
Perceived Exertion and the Experience of Effort
Perceived exertion is a central construct because athletes act in response to the demands they experience, not only to externally measured power. Borg, G. A. V (1982) provided psychophysical measurement foundations; later endurance work incorporated repeated ratings at standardized time or workload points. These ratings may predict continued willingness to exercise and can also reflect how close the athlete believes they are to finishing.
A key methodological distinction is whether perceived effort is measured at the same absolute workload, the same elapsed time, the same physiological state, or the same proportion of the eventual trial. If a mentally fatigued participant completes an exercise test sooner, ratings collected near exhaustion in each condition may correspond to different durations and accumulated work. If the participant self-selects a lower power output, equal ratings may coexist with poorer performance. These facts complicate a common inference that a higher reported effort must necessarily precede every fatigue-related reduction in performance.
The psychobiological interpretation emphasizes that the sensation of effort and potential motivation jointly inform the decision to continue or stop. An integrative account also considers afferent signals, perceived bodily threat, environmental incentives, prior experience, and reward. McCormick, A., Meijen, C., & Marcora, S (2015) situated perceived exertion and psychological strategies within endurance-specific experimental research. The construct's value lies in its capacity to connect subjective evaluation and objective action; its limitation is that a subjective scale does not identify a unique neural or physiological generator.
Anticipatory Pacing and Knowledge of the Endpoint
Self-paced exercise involves forecasting. At each moment, the athlete must estimate whether present expenditure is compatible with the remaining duration, route, and anticipated changes in difficulty. An identical sensation of strain may invite acceleration near the finish and conservation early in the event. Tucker, R (2009) treated perceived exertion as information involved in this forward-looking adjustment rather than only as a report of accumulated bodily cost.
The model becomes especially informative when the endpoint is known. A cyclist completing a fixed-distance trial can compute remaining distance, compare their progress with expected performance, and revise the estimated sustainable output. If endpoint knowledge is inaccurate or concealed, the decision problem changes. Competitor behavior, environmental cues, and available feedback can further shift the anticipated value of maintaining a given intensity.
de Koning, J. J., Foster, C., Bakkum, A., Kloppenburg, S., T (2011) proposed a hazard score combining momentary rating of perceived exertion with the fraction of the competition remaining. Their analysis found this construct informative for changes in simulated race velocity. The inference is that pacing is sensitive to the interaction of current sensations and prospective distance. The specific mathematical relationship should be evaluated across events, intensities, and individual athletes before being generalized.
This anticipatory framework offers a direct bridge to mental fatigue. If prior cognitive work changes the interpretation of effort, tolerance for demands, or confidence in a future plan, it may alter the selected pace even when physiological capacity remains comparable. The bridge remains a hypothesis until the intermediate variables and temporal sequence are directly measured.
Pacing as Decision Making Under Uncertainty
A pacing choice is a sequential decision with delayed consequences. Exceeding an appropriate intensity can improve immediate position while reducing later options; maintaining a cautious pace can preserve resources while risking a poorer finishing time. Renfree, A., Martin, L., Micklewright, D., & St Clair Gibson (2014) described the application of decision-making theory to work-rate regulation in competitive endurance activity. Smits, B. L. M., Pepping, G.-J., & Hettinga, F. J (2014) extended this discussion through the roles of perception and action.
The relevant information varies with the task. Athletes can know their objective velocity and power yet remain uncertain about future discomfort, terrain, competitor actions, or remaining reserves. They learn the meaning of a particular bodily sensation through training and repeated exposure to comparable efforts. Prediction errors can lead to immediate revision; deliberate restraint can reflect a longer-term strategy. The psychological problem is therefore partly one of allocating limited evidence, resolving uncertain future outcomes, and controlling behavior in light of a target.
Mental fatigue could influence this process through attention, valuation, working-memory demands, reduced confidence, or a preference for less effortful actions. These pathways predict different effects. A valuation shift should change willingness to accept the same effort; degraded monitoring should change the precision of adjustments; altered attention should influence which cues guide pace. If all are labeled 'cognitive fatigue,' mechanisms become experimentally indistinguishable. Future trials should manipulate or measure these pathways separately.
Psychobiological Models of Endurance Performance
The psychobiological model associated with Marcora and colleagues emphasizes conscious perception of effort and potential motivation in the regulation of endurance termination and performance. Under this account, exercise can stop when required effort reaches the maximum effort a person is willing to invest under prevailing incentives and conditions. It predicts that psychological manipulations may alter performance through changes in effort perception or motivational value even without a measurable decrement in muscle capacity.
The theory gained prominence following Marcora, S. M., Staiano, W., & Manning, V (2009), where prior cognitive exertion was followed by poorer subsequent cycling endurance and increased perceived effort in the mentally fatigued condition. Blanchfield, A. W., Hardy, J., de Morree, H. M., Staiano, W (2014) later studied motivational self-talk, a behavioral intervention directly relevant to effort perception. The value of a mechanistic model is its ability to generate contrasts between effort-mediated effects and peripheral changes.
The model also faces important tests. A successful cognitive fatigue manipulation may produce no subsequent endurance decrement, as demonstrated by the preregistered Holgado, D., Troya, E., Perales, J. C., Vadillo, M. A., & Sa (2021). Such a result tests the robustness and boundary conditions of the proposed link rather than denying that perceived effort matters. The most informative experiments measure potential motivation, effort, work rate, endurance outcome, and physiological response while preregistering causal predictions.
Several rival explanations can accommodate some of the same observations. Participants may alter behavior because the preceding task changes arousal, boredom, expectancy, mood, attentional deployment, or understanding of experimenter demands. Distinguishing such accounts requires designs more targeted than a single broad cognitive-fatigue manipulation followed by one performance outcome.
Cognitive-Control and Resource Accounts
Sequential-task theories ask whether performance deteriorates because effortful control leaves the individual less able or less willing to exert further control. The classic limited-resource formulation was advanced in broader self-control research and summarized by Hagger, M. S., Wood, C., Stiff, C., & Chatzisarantis, N. L (2010). Its claim about a depletable common resource attracted substantial experimental attention. Yet Carter, E. C., Kofler, L. M., Forster, D. E., & McCullough (2015) found weak evidence in bias-sensitive meta-analytic tests, and the preregistered multilaboratory work of Hagger, M. S., Chatzisarantis, N. L. D., Alberts, H., Anggon (2016) challenged the generality of the large, robust depletion effects originally proposed.
Mental fatigue and ego depletion are related research traditions but cannot be treated as identical constructs. A long vigilance or inhibition task may produce subjective fatigue without the particular sequential self-control effect assumed by the strength model. Conversely, a short cognitive task can alter motivation or attention in the absence of prolonged fatigue. A researcher must specify whether the theory concerns depleted capacity, changed cost-benefit appraisal, lower task engagement, reduced control precision, or a learned avoidance response.
The endurance context presents a particularly stringent test. A lower work rate following cognitive exertion could result from changed appraisal at intact physiological capacity; a general resource model predicts different relations to other executive tasks. Disentangling these possibilities requires manipulation checks, mediator measurements, and task batteries rather than a simple inference from sequential performance differences.
The Original Mental-Fatigue Endurance Experiment
Marcora, S. M., Staiano, W., & Manning, V (2009) used a randomized crossover design with sixteen participants. Before a constant-load cycling test to exhaustion at eighty percent of peak power output, participants completed ninety minutes of a demanding cognitive task or watched emotionally neutral documentaries for the same duration. This design controlled part of the passage of time and introduced a direct experimental contrast between demanding and comparatively passive preparatory activities.
The authors reported greater subjective mental fatigue and reduced cycling endurance after the cognitive condition. Their measurements also supported the interpretation that the performance change did not simply mirror a clear difference in the conventional physiological variables they assessed. The influential explanatory proposal was that cognitive exertion raised the perceptual cost of maintaining a given exercise intensity, thereby shortening tolerance of the fixed workload.
The study's strengths include within-person comparison, explicit manipulation, and an objective physical outcome. Its important limitations include modest sample size, imperfect blinding of conspicuously different cognitive tasks, and a single primary endurance paradigm. Time to exhaustion can exhibit substantial variability and does not map transparently onto the pacing decisions of a fixed-distance race. The experiment supports a causal contrast under its tested conditions; it does not establish a universal effect across all endurance tasks, athlete populations, and fatigue manipulations.
These distinctions matter because the original finding became widely cited as evidence about what 'the brain' does to physical capacity. What it directly tested was how one preparatory cognitive manipulation changed subsequent performance under a particular controlled exercise demand. Later studies and replications must determine how broadly that result generalizes.
Response Inhibition and Self-Paced Performance
Pageaux, B., Lepers, R., Dietz, K. C., & Marcora, S. M (2014) specifically examined whether prior response inhibition impaired subsequent self-paced endurance performance. Twelve physically active participants completed five-kilometer treadmill time trials following thirty-minute incongruent or congruent Stroop tasks. Running performance was worse after response inhibition, although neither task produced a subjective increase in mental fatigue. This distinction supports a role for preceding response-control demands in pacing without establishing subjective mental fatigue as a mediator. This extension was important because self-paced tests provide opportunities to adjust work rate dynamically, bringing the experimental outcome closer to the decision structure of an endurance event.
The causal question differs from constant-load tolerance. Participants may finish more slowly because they work at lower intensity throughout, because they make conservative decisions in particular segments, or because they cannot sustain a previously chosen pace. Repeated measures of power, speed, and effort along the trial can discriminate these patterns. A total finishing time alone cannot resolve the temporal mechanism.
Response inhibition is also a compound manipulation. An incongruent Stroop task recruits monitoring, conflict resolution, response selection, sustained engagement, and task motivation. The magnitude of conflict and perceived difficulty can change through practice. Calling a subsequent performance change evidence for the exhaustion of inhibition is stronger than the experiment itself warrants without additional measures of selective executive control.
This study provides a meaningful bridge between cognitive exertion and pacing but also illustrates a general challenge: laboratory manipulations designed to produce subjective fatigue are often broad, while the explanation is framed as a precise psychological mechanism. Robust mechanistic evidence requires multiple tasks that converge on the same construct and discrimination between cognitive fatigue, boredom, frustration, and altered incentives.
Neuromuscular Tests and Alternative Physical Mechanisms
The hypothesis that cognitively demanding work exhausts muscles or directly diminishes maximal muscle activation makes distinct predictions from an effort-valuation account. Pageaux, B., Marcora, S. M., & Lepers, R (2013) examined the effects of prolonged mental exertion on knee-extensor neuromuscular function. The study belongs to a line of work testing whether mental-fatigue-induced performance changes can be attributed to measured peripheral or central neuromuscular impairment.
A negative finding on a particular neuromuscular measure is informative but bounded. Measures of voluntary activation, maximal force, peripheral twitch response, and fatigue induced by subsequent exercise reflect different systems. A failure to detect a change in one does not eliminate effects on autonomic regulation, motor coordination, movement economy, or perception. Equally, a subjective report of heavy effort does not establish a structural decline in muscular capacity.
This analysis clarifies why objective performance and subjective experience must be studied together. A participant may be physically capable of generating a certain output but select a lower one; another participant may attempt to maintain output and encounter an actual physiological limitation. Both processes can occur within a single trial and may vary by time, context, and training history.
A careful research design records baseline and postmanipulation physiological status, experimental environment, medication or caffeine exposure when relevant, familiarization, and the timing of neuromuscular measurements. It avoids claiming that mental fatigue has an exclusive 'central' or 'peripheral' location merely from unchanged heart rate or lactate during one task.
Trained Athletes and Resistance to Fatigue
An endurance mechanism inferred from recreationally active participants requires validation among trained athletes. Professional competitors may have more stable pacing expectations, greater familiarity with discomfort, developed task-specific attention, and different performance incentives. Differences in cognitive-task proficiency may also change the degree of fatigue induced by a fixed protocol.
Martin, K., Staiano, W., Menaspà, P., Hennessey, T., Marcora (2016) compared eleven professional and nine recreational road cyclists. Each completed a modified demanding Stroop condition or easier cognitive control, followed by a twenty-minute cycling time trial. The professional group displayed stronger inhibitory-task performance and greater resistance to the negative performance effects observed among recreational cyclists. The result is a valuable demonstration of possible population moderation, although observational group comparison cannot distinguish training effects from selection, habitual cognitive characteristics, or other differences between the groups.
The study also underscores the necessity of individualizing cognitive load. A task capable of inducing substantial fatigue in one population may be relatively easy in another. A fixed-duration manipulation can produce different subjective costs even when it uses the same stimuli. Equal exposure should therefore not be mistaken for equal experimental dose.
Longitudinal work is needed to determine whether resistance to mental fatigue develops through endurance training, independent cognitive practice, occupational demands, or other processes. Cross-sectional contrasts can motivate this program but cannot establish its developmental mechanism.
Preregistered Replication and the Limits of Generalization
A decisive development was the preregistered replication by Holgado, D., Troya, E., Perales, J. C., Vadillo, M. A., & Sa (2021). Thirty recreationally active adults completed a within-subject comparison involving a ninety-minute cognitive task or documentary viewing before a cycling time-to-exhaustion test at eighty percent of maximal oxygen uptake. The manipulation produced convincing subjective mental fatigue, but the study did not find a reduction in average endurance time.
The paper reported a Bayes factor favoring the null hypothesis for time to exhaustion, BF01 = 9.762. This numerical result concerns the authors' defined model comparison and data; it is evidence against the tested performance effect rather than proof that mental fatigue never affects endurance. The replication also reported weaker or inconclusive evidence concerning other measured responses. Most importantly, the manipulation and the endurance outcome were separated empirically: increased felt mental fatigue did not inevitably result in poorer exercise performance.
This finding must inform all subsequent synthesis. A literature that contains positive small experiments and a rigorous null replication warrants examination of task demands, participant characteristics, test reliability, outcome choice, expectancy, and selective publication. Treating either single study as the final adjudication would discard useful evidence.
Preregistration improves the interpretability of a test by documenting planned contrasts. It does not guarantee adequate measurement or universal external validity. The central conclusion is conditional: the causal effect of acute prior cognitive exertion on objective endurance is less stable than would be expected from a universal impairment theory.
Systematic Reviews and Bias-Sensitive Synthesis
Van Cutsem, J., Marcora, S., De Pauw, K., Bailey, S., Meeuse (2017) reviewed the effects of mental fatigue on physical performance and helped consolidate the early experimental literature. Brown, D. M. Y., Graham, J. D., Innes, K. I., Harris, S., Fl (2020) quantitatively synthesized research on prior cognitive exertion and physical performance. Reviews are valuable because individual experiment sizes are often modest and outcomes differ substantially across studies.
A central methodological advance was the bias-sensitive analysis by Holgado, D., Sanabria, D., Perales, J. C., & Vadillo, M. A (2020). Their systematic review and meta-analysis questioned the magnitude and security of the frequently claimed impairment by explicitly considering bias in the available published record. A subsequent Holgado, D., Sanabria, D., Perales, J. C., & Vadillo, M. A (2021) repaired miscoded effect-size signs and added omitted perceived-exertion measurements. The authors reported that the correction did not materially change the original conclusions. The contrast between early general conclusions and later critical analyses is part of the research result itself. Evidence synthesis must assess small-study effects, publication selection, and uncertainty rather than count how many published experiments report statistical significance.
Meta-analyses in this area sometimes combine muscular endurance, whole-body aerobic endurance, strength, and sport-specific skill. A single standardized effect drawn from heterogeneous endpoints can obscure differences in underlying mechanisms and reliability. Time-to-exhaustion effects have different statistical and practical properties from time-trial outcomes; measures of accuracy or reaction time require still another interpretation.
For this dossier, the robust position is that cognitive-fatigue manipulations reliably produce subjective effects under some protocols, while the magnitude, prevalence, and contextual conditions of objective endurance impairment remain contested. An evidence-based synthesis must give positive randomized findings, rigorous replications, and bias-sensitive analyses explicit and proportionate attention.
Mental Fatigue in Heat and Environmental Stress
Environmental conditions test the generalizability of cognitive-fatigue findings. Heat increases thermoregulatory demand, alters thermal sensation, and can influence pacing before exhaustion. An added cognitive task might plausibly compound those demands, yet the prediction must be evaluated rather than assumed. A physically demanding environment can change baseline strain, attention to discomfort, and the marginal effect of an additional cognitive manipulation.
Van Cutsem, J., De Pauw, K., Buyse, L., Marcora, S., Meeusen (2017) investigated endurance performance in heat. Ten endurance-trained male athletes completed a forty-five-minute demanding cognitive task or comparison condition in a hot environment, followed by fixed-pace cycling and a self-paced time trial. Subjective mental fatigue increased following the cognitive condition, but time-trial performance did not significantly differ. This result contrasts with some experiments at other temperatures and exercise intensities.
The study's inference is limited by its small sample and specific heat protocol; a nonsignificant comparison does not establish equivalence under every hot-weather condition. It does, however, reject any claim that subjective mental fatigue invariably translates into reduced subsequent time-trial performance. The plausible interpretation includes moderation by environmental stress, magnitude of induced fatigue, pacing strategy, and prior heat adaptation.
Future factorial designs should independently vary cognitive demand and heat exposure, prespecify the interaction term, and collect repeated effort and thermal ratings. Such trials could determine whether environmental stress amplifies cognitive effects, replaces or masks them, or alters which information athletes prioritize during pacing. Findings from controlled heat chambers should then be tested in ecologically realistic races.
Time-to-Exhaustion Tests and Time-Trial Outcomes
The distinction between time to exhaustion and time-trial performance is foundational to evidence appraisal. In a constant-load time-to-exhaustion test, the participant usually maintains an externally imposed intensity until stopping. Motivation, tolerance of discomfort, and task engagement can markedly affect how long the participant continues. In a fixed-distance time trial, participants choose a pace trajectory that determines their total completion time. They can compensate for an early conservative choice by accelerating later, subject to physiological constraints.
Time-to-exhaustion outcomes are therefore sensitive to both psychological stopping decisions and the reliability of the protocol, while self-paced outcomes offer greater opportunity to study dynamic regulation. A mental-fatigue manipulation could affect one without affecting the other. The original Marcora, S. M., Staiano, W., & Manning, V (2009), the self-paced study by Pageaux, B., Lepers, R., Dietz, K. C., & Marcora, S. M (2014), and the null replication by Holgado, D., Troya, E., Perales, J. C., Vadillo, M. A., & Sa (2021) consequently belong in one discussion but should not be pooled conceptually as identical tests.
Performance should be reported as distance, time, power, work, variability, and response to the initial manipulation. Researchers should also clarify whether the endpoint was known, feedback was standardized, and participants were familiarized to stable performance. For a pacing mechanism, temporal series often have greater explanatory value than a single final number.
A further caution concerns percentage interpretations. Equal absolute differences can produce different proportional effects depending on baseline performance. Statistical significance alone says little about practical relevance to an elite race in which small finishing-time differences matter. Applied conclusions require uncertainty intervals and context-sensitive benchmarks.
Temporal Dynamics of Self-Regulated Work Rate
Pacing profiles often contain an opening phase, a period of relative regulation, adjustments to terrain or competitors, and a final increase in effort when the endpoint approaches. These patterns are descriptions of behavior; assigning each phase to an independent control mechanism requires evidence. A fast start can be deliberate positioning, energetic miscalculation, or adaptation to race demands. A late acceleration may reflect conserved capacity, changed motivation, or more accurate endpoint knowledge.
de Koning, J. J., Foster, C., Bakkum, A., Kloppenburg, S., T (2011) offered a practical example of studying the relationship between momentary perceived exertion, remaining distance, and velocity changes. Renfree, A., Martin, L., Micklewright, D., & St Clair Gibson (2014) framed pacing decisions as repeated allocation of muscular work. These approaches encourage the use of repeated measures instead of treating performance as a unitary outcome.
For mental-fatigue research, the central temporal hypothesis is specific. If cognitive load affects early anticipation, a difference should appear in the opening distribution of work; if it increases the continuing subjective price of effort, divergence might accumulate during the middle phase; if it impairs late attentional control, differences could emerge near the finish. Each profile can be tested against preregistered hypotheses.
Analyses should account for serial dependence, participant-specific baseline pacing, and the fact that segments of a finite event constrain one another. Comparing many time points without correcting multiplicity inflates apparent discoveries. Hierarchical time-series models can estimate individual response curves while separating event position from accumulated elapsed time. Such analysis should complement the primary objective outcome and remain transparent about exploratory choices.
Competitive Context, Opponents, and Information
Real endurance competition changes the reward structure of pacing. An athlete may accept a less efficient energy distribution to respond to a rival, remain with a group, exploit drafting, or preserve tactical options. The optimal work rate depends on objectives such as finishing position, record time, qualification, or team strategy. Accordingly, even a technically precise laboratory model of solo time trials may capture only part of competitive regulation.
Decision-oriented accounts by Renfree, A., Martin, L., Micklewright, D., & St Clair Gibson (2014) and Smits, B. L. M., Pepping, G.-J., & Hettinga, F. J (2014) provide an entry point for linking perception and action to competition. Their perspective suggests that available information, affordances, and opponents' actions influence when an athlete invests additional work.
Mental fatigue may plausibly change how this information is sampled. A cyclist who attends less consistently to changes in a group might miss an opportunity or respond later; another might simplify their strategy and maintain a stable pace. Yet these scenarios are mechanistic proposals. Evidence from solo exercise tests cannot be presented as direct proof of impaired tactical decisions in an actual race.
Ecologically valid studies can use simulated opponents, realistic course variability, and standardized incentives while retaining experimental control over the cognitive manipulation. Researchers must distinguish changes in decision quality from differences in risk preference and strategy. A slower finishing time can be rational for an athlete pursuing a different objective; a single performance metric cannot reveal the decision criterion without additional information.
Attentional Focus During Endurance Activity
Attention during endurance is selective and dynamic. Athletes monitor breathing, muscular sensations, task cues, pace, and external conditions while also managing distracting thoughts. Attention may be directed toward bodily sensations, movement technique, target pace, music, environmental features, or competitors. The functional value of a focus depends on the task, current intensity, expertise, and whether information is useful for immediate regulation.
Brick, N. E., MacIntyre, T. E., & Campbell, M. J (2016) synthesized a cognitive account of self-regulation during endurance, emphasizing attentional focus, metacognition, and action. Brick, N. E., Campbell, M. J., Metcalfe, R. S., Mair, J. L (2016) experimentally examined attentional-focus effects during running with attention to pace control and pace regulation. Such research suggests that attention can affect how the athlete monitors and acts upon performance-relevant cues.
It would be misleading to prescribe one type of attention as uniformly optimal. Directing attention toward bodily sensations may improve monitoring and protect against unsustainable intensity, while some forms of external focus can support economy or reduce unproductive rumination. Distraction may be helpful at one intensity and costly when precise tactical or bodily monitoring is essential.
Mental fatigue may influence the flexibility of attention rather than simply its total amount. An important experimental prediction is that cognitively fatigued athletes will have difficulty changing focus when task contingencies shift, even if they can sustain a familiar focus during steady-state exercise. Testing this requires objective probes of attentional transitions, not only posttrial descriptions of preferred strategies.
Cognitive Monitoring and Metacognition
Metacognition concerns how individuals evaluate and regulate their own cognitive activity. During endurance exercise, an athlete estimates their present state, forecasts tolerable future exertion, recognizes errors in pacing, and updates the strategy. These operations involve confidence and uncertainty. The athlete may know the displayed power value precisely but remain uncertain whether that output will be sustainable for the remaining course.
The framework developed by Brick, N. E., MacIntyre, T. E., & Campbell, M. J (2016) connects cognition and action through deliberate monitoring and self-regulation. Metacognitive inquiry asks whether the athlete's judgments are calibrated: do predicted remaining reserves, performance confidence, and planned versus actual finishing times correspond to later outcomes? Calibration can be measured without claiming to access an internal physiological 'reserve' directly.
A sustained cognitive task may alter these estimates through fatigue, reduced engagement, or biased appraisal of effort. More demanding evidence would demonstrate that measured changes in confidence or monitoring precede and statistically explain changes in pace. Cross-sectional association between low confidence and slow performance cannot establish this pathway, because poor performance may itself reduce confidence.
Repeated prospective judgments are valuable here. Before and during the event, participants might estimate completion time, anticipated discomfort, or probability of sustaining a target output. These predictions can be compared with actual performance. The resulting data reveal how cognitive regulation represents future action and can separate poor pacing caused by inaccurate estimation from deliberate selection of a conservative goal.
The Role of Motivation, Goals, and Incentives
Endurance regulation depends on what the athlete is trying to accomplish and how much effort they are willing to invest. A personal record, a controlled training effort, competition placement, or rehabilitation objective creates a distinct decision criterion. Researchers studying mental fatigue must therefore specify the performance incentive and ensure that participants understand it consistently across experimental conditions.
Psychobiological accounts link potential motivation to willingness to continue as perceived effort rises. This can explain why two individuals with similar physiological capacity choose different stopping times. However, motivation is difficult to infer from behavior alone. A participant's early termination of exercise could follow reduced motivation, fear, physical symptoms, or a shift in the evaluation of future costs. Investigators need independent measures or experimental manipulations of incentive.
The systematic work of McCormick, A., Meijen, C., & Marcora, S (2015) reviewed a range of psychological determinants and interventions in whole-body endurance. Their framework encourages the researcher to distinguish strategies designed to alter motivation from strategies intended to change attentional control or the sensation of effort.
Experimental incentives can create strong tests: vary performance-contingent reward while holding workload and cognitive manipulation constant, then examine whether the effect of fatigue changes. Such a design can discriminate a motivational valuation account from a simple fixed-capacity deficit. Interpretation still requires safeguards against expectation effects and demand characteristics, because a salient reward may also increase arousal or change engagement with the preceding cognitive task.
Motivational Self-Talk and Endurance Regulation
Self-talk is an explicit verbal strategy used to guide attention, motivation, or interpretation of effort. An athlete might rehearse an instruction to sustain rhythm, divide the task into manageable segments, or reappraise discomfort as an expected component of exercise. Its effects depend on actual content, frequency, training, and situation rather than the presence of words alone.
Blanchfield, A. W., Hardy, J., de Morree, H. M., Staiano, W (2014) tested motivational self-talk in a randomized between-group pretest–posttest experiment involving twenty-four participants and repeated cycling time-to-exhaustion tests. The study examined whether a short intervention could improve endurance and alter perceived exertion. It supplied original experimental evidence supporting investigation of self-talk as an endurance-specific intervention.
Several questions remain before generalizing to real competition. Participants may differ in belief in the strategy, prior use of self-talk, and adherence. A relatively brief training program may generate expectancy effects; the value of a phrase may differ by event duration and exercise intensity. A time-to-exhaustion result may not translate directly into faster pacing or race placement.
Effective future tests would compare structured motivational phrases with equally credible active controls, track adherence during exercise, and separate task-specific instructions from generalized encouragement. They would also assess adverse or ineffective responses. The appropriate scientific claim is that self-talk has experimentally supported potential in particular endurance contexts, while precise mechanisms and optimal implementation require further validation.
Brain Endurance Training as an Intervention Program
Brain endurance training integrates demanding cognitive exercise with physical training, either sequentially or concurrently, with the aim of improving tolerance of combined cognitive and physical demands. It is conceptually distinct from an acute fatigue induction. A repeated cognitive task may induce short-term fatigue while potentially producing training adaptation across sessions; this hypothesis requires evidence for durable transfer to real endurance outcomes.
Recent quantitative evidence has accelerated. Yu, J., Tan, Z., Huang, Y., Chen, L., & Li, D (2026) synthesized published training studies addressing mental fatigue, cognitive outcomes, and endurance. Wang, N., Chen, Z., & Zhao, K (2026) analyzed ten randomized controlled trial articles representing twelve independent experiments. Under induced mental-fatigue test conditions, their eight-experiment synthesis (n = 275) favored brain endurance training, standardized mean difference 0.82 with 95% confidence interval 0.33 to 1.30, while heterogeneity was substantial (I² = 69%). The non-fatigued-condition estimate was imprecise, with a confidence interval including zero.
The results are promising but require careful assessment of overlapping study samples, risk of bias, training dosage, comparator credibility, and the distinction between improved fatigue resistance and general athletic development. A training intervention that provides additional structured cognitive contact may also change expectations or commitment.
The clearest interpretation is that brain endurance training may benefit tested performance, especially under induced fatigue, with considerable uncertainty about magnitude and generalization. Replications should compare credible active controls, preregister primary endurance outcomes, and document physical training equivalence.
The 2026 Evidence on Brain Endurance Training
The 2026 literature contains several partly overlapping quantitative syntheses; their results should not be treated as independent confirmations. Yu, J., Tan, Z., Huang, Y., Chen, L., & Li, D (2026) identified nine articles and eleven studies, reporting beneficial patterns in cognitive and endurance outcomes. Wang, N., Chen, Z., & Zhao, K (2026) reexamined endurance performance under fatigued and non-fatigued conditions, with design-specific bias assessment and effect-dependency considerations.
A further three-level meta-analysis by Zheng, X., Sun, Y., Zhao, Z., Wang, C., Shang, Z., Xie, Y (2026) included eleven studies, thirteen trials, and 441 participants. It reported positive pooled associations for cognitive function and physical performance, while finding no statistically significant pooled effect on perceived exertion. The authors rated the certainty for their primary performance outcomes as low. This is especially important for theories that assume that training effects must operate through measurable reductions in effort perception.
Apparent numerical disagreement between meta-analyses can result from differences in inclusion criteria, time windows, correlated outcomes, effect-size orientation, and model structure. Individual trials may appear in more than one review, so citing multiple reviews does not multiply the number of independent randomized experiments. A cumulative evidence assessment should map overlapping trials and appraise their shared limitations.
At present, the mechanistic theory of brain endurance training is less secure than the observation that certain experimental training packages improve selected measured outcomes. A properly powered trial should assess neural and psychological candidate mediators independently while preserving objective task-specific performance as the primary endpoint.
Pacing Strategy Trials and the 2026 Network Meta-Analysis
Pacing instructions can be manipulated experimentally. Athletes may be required to start quickly, maintain an even profile, start slowly, or select their own pace. Each condition changes the distribution of effort and may change what psychological decisions remain available during the event. These comparisons can inform coaching practice but do not necessarily isolate a single mental mechanism.
Ramos-Campo, D. J., Foster, C., Andreu-Caravaca, L., Rubio-A (2026) published a systematic review and network meta-analysis of pacing strategies for distance-based endurance time trials. Their analysis included twenty-four studies, 302 participants, and 765 observations. Across the network, the authors found no consistent performance advantage for imposed fast-start, even, or slow-start strategies relative to self-selected pacing. Longer slow-start phases appeared particularly unfavorable in moderator analyses.
This finding challenges universal prescription of a single pacing profile. Self-pacing can adapt to sport, athlete, intensity, event duration, and context. The review does not prove every chosen pace is optimal, nor does it imply that pacing education or targeted strategy training is ineffective. It establishes that pooled comparisons across heterogeneous experiments do not identify a universally superior imposed profile.
The connection with mental fatigue is direct but still largely inferential: if self-selected pacing normally adapts to individual constraints, fatigue may change either the constraints or the quality of the selection policy. Future factorial experiments should cross pacing strategy with cognitive-fatigue manipulation, enabling examination of whether imposed pacing reduces, increases, or reverses any subsequent performance effect.
Cognitive Fatigue Induction and Construct Validity
A central methodological difficulty is determining whether a manipulation induces the psychological construct under study. A long Stroop task may be demanding because of conflict, sustained engagement, novelty, monotony, error feedback, or motivation. Watching a documentary can vary in interest, emotional arousal, posture, and even cognitive activity. Thus, different conditions may differ along dimensions beyond the intended fatigue construct.
Goodman, S. P. J., Collins, B., Shorter, K., Moreland, A. T (2025) systematically reviewed approaches to inducing mental fatigue and examined neurophysiological indices. Their work illustrates the need to triangulate subjective reports with behavioral or physiological indicators while recognizing that no biomarker is universally definitive.
Manipulation checks should assess fatigue immediately before and after the cognitive task and ideally track changes during it. Repeated perceived-mental-fatigue reports, accuracy, response-time variability, and task engagement can describe different properties of the manipulation. A researcher should not exclude participants post hoc simply because their fatigue report failed to rise, as such selection can break the randomized comparison.
Construct validity improves when different induction methods yield convergent outcomes, when dose-response predictions are preregistered, and when alternative states such as boredom or stress are measured. Standardized instructions, practice trials, stimulus timing, and task difficulty help make studies comparable. An operationally successful manipulation remains a measured experimental state, not a demonstration of a unique exhausted cognitive reservoir.
Statistical Inference, Small Studies, and Publication Bias
Many endurance-fatigue studies have modest samples, often because crossover protocols require multiple long testing sessions. Within-person designs can improve precision, but they also create dependencies that must be modeled correctly. Performance often varies between days, especially when recovery, sleep, nutritional preparation, exercise familiarity, and motivation are incompletely controlled.
Statistical reporting should include the preregistered primary endpoint, mean condition difference, uncertainty interval, paired correlation when relevant, individual trajectories, and an interpretable measure of practical importance. Comparing multiple secondary measures without multiplicity control creates a substantial opportunity for selective significance. The absence of statistical significance does not itself establish equivalence; conversely, a small p value does not validate the proposed mechanism.
Holgado, D., Sanabria, D., Perales, J. C., & Vadillo, M. A (2020) emphasized how publication bias and analytic choices can inflate apparent mental-fatigue effects. Brown, D. M. Y., Graham, J. D., Innes, K. I., Harris, S., Fl (2020) provided quantitative synthesis across prior cognitive exertion studies. These works should be read together because the quality of a pooled estimate depends on the quality and selection of its input experiments.
The stronger evidence standard is independent, sufficiently powered, prospectively registered trials with openly reported negative results. Sensitivity analysis should explore alternative outcome definitions and assumptions about dependence. Researchers should also distinguish effects at the individual and population levels: a practically important response in a subset of athletes may coexist with a small or uncertain group mean.
Experimental Controls, Expectancy, and Blinding
Cognitive-fatigue protocols are difficult to blind. Participants often recognize that a demanding cognitive task differs from a passive control and may infer the experiment's intended hypothesis. Believing that mental fatigue should reduce physical performance can itself influence effort, confidence, and termination decisions. This threat is particularly important when the proposed causal pathway includes motivation or perceived exertion.
A well-designed trial therefore measures expectancy and perceived condition credibility, disguises the precise performance hypothesis where ethically appropriate, and uses an active control matched on time, sensory engagement, posture, and interaction with the experimenter. The cognitive tasks must differ reliably in the intended demand while limiting unrelated variation in emotional salience and incentives.
Crossover order should be randomized and counterbalanced. Adequate washout and standardization of preceding physical activity, sleep, meals, stimulants, and time of day reduce alternative explanations. Familiarization sessions can decrease the variability of time-to-exhaustion and time-trial performance. Experimenters assessing objective outcomes should be blinded to condition when feasible, and all protocol deviations should be recorded.
Holgado, D., Troya, E., Perales, J. C., Vadillo, M. A., & Sa (2021) demonstrates why preregistered replication is valuable: a clear separation between successful fatigue induction and an absent performance effect becomes interpretable. Future investigations can strengthen causal inference further by independently manipulating expectation and cognitive demand, thereby assessing whether subjective beliefs mediate or moderate the performance contrast.
Measurement of Effort, Affect, and Bodily Sensation
A cognitive regulation study should not treat effort ratings as a complete description of the participant's internal state. Perceived exertion, muscular pain, dyspnea, thermal discomfort, affective valence, boredom, and fatigue are conceptually different experiences. They can affect pacing through different pathways even if self-report items are correlated. Measuring them with overlapping or ambiguous wording risks constructing an apparent mechanism from measurement redundancy.
Borg, G. A. V (1982) provides a foundation for effort measurement, but instructions must specify whether participants rate whole-body exertion or a specific sensation. Immediately preceding cognitive-task ratings and exercise-time ratings should be labeled distinctly. Researchers can record mental fatigue before exercise, repeatedly assess effort and bodily sensations during the event, and measure affective or motivational states afterward with clearly differentiated timing.
Repeated prompts may themselves change attention and pace. A rating procedure that interrupts an athlete's natural monitoring introduces another cognitive task and may differentially affect fatigued participants. Investigators can minimize measurement burden, prespecify prompt intervals, or use continuous instruments whose validity has been independently established.
A compelling explanation requires temporal ordering: the proposed mediator should change before the pacing response that it is claimed to explain. Postexercise recall of how difficult the trial felt is vulnerable to final-outcome knowledge and memory bias. Multi-method measurement is useful, but each method requires its own reliability, calibration, and inferential limits.
Individual Differences and Athlete Development
Athletes differ in endurance training, cognitive skill, motivation, attentional habits, sleep, and response to repetitive laboratory tasks. A group-average effect therefore may conceal meaningful heterogeneity. Martin, K., Staiano, W., Menaspà, P., Hennessey, T., Marcora (2016) offered early evidence that elite and recreational road cyclists may differ in both inhibitory control and behavioral response to a cognitive preload.
The source of such differences remains unsettled. Elite selection may favor particular psychological characteristics; sustained practice may change perceived effort or familiarity with high demands; and laboratory cognitive tasks may have different subjective difficulty for highly trained participants. These are competing causal hypotheses. A cross-sectional comparison cannot independently establish each pathway.
Studies should include athletes of different sports, sexes, ages, and performance levels, while avoiding underpowered subgroup claims. Biological and social context can affect recovery opportunities, exposure to cognitive work, and preferred pacing strategies. Researchers should assess whether findings replicate across running, cycling, swimming, rowing, and other endurance disciplines rather than assuming that one ergometer protocol is a general test of human endurance regulation.
Longitudinal designs can separate stable between-athlete characteristics from within-athlete variability. Repeated training and testing may reveal whether mental-fatigue sensitivity changes with season, familiarization, or recovery. Describing the distribution of individual effects is scientifically valuable even when the population mean is small.
Sleep, Recovery, and Everyday Cognitive Load
Acute experimentally induced mental fatigue occurs within a wider recovery system. Sleep loss, chronic stress, travel, occupational workload, psychological distress, heavy physical training, and inadequate recovery may each alter cognition and exercise performance. Their interactions are important but should not be presumed equivalent to a ninety-minute inhibition task.
In real endurance preparation, the relevant exposure may be a day of decision making, prolonged screen work, emotionally demanding events, or disrupted sleep. Such exposures differ in duration, incentive, physiological consequence, and recovery opportunities. A laboratory Stroop manipulation provides experimental control over one kind of demand; it cannot represent every type of everyday mental workload.
A future field study could use repeated daily assessments of cognitive demand, sleep, subjective fatigue, training load, and standardized exercise performance. However, correlations between hard workdays and poorer workouts would be confounded by schedule, nutrition, prior training, mood, and self-selection. Randomized scheduling or within-person crossover designs could improve identification.
Practical implications follow from this methodological distinction. Managing precompetition cognitive load is a plausible intervention target, but its efficacy should be evaluated against other preparation priorities with objective outcomes. Athletes should not infer from the fatigue literature that all mentally demanding activity before exercise necessarily reduces performance. The scientific goal is to specify exposure, timing, task context, and tested outcome.
Neural and Physiological Candidates for Cognitive Regulation
The neural basis of perceived effort, sustained cognitive work, and action regulation is a complex research question. Electrophysiological, imaging, autonomic, and behavioral indices can all contribute, but none alone reveals a single brain mechanism responsible for endurance termination. Neurophysiological differences after a fatigue task can reflect attention, arousal, task engagement, movement, or expectation.
Goodman, S. P. J., Collins, B., Shorter, K., Moreland, A. T (2025) examined neurophysiological indices across fatigue induction methods, underscoring measurement heterogeneity. Pageaux, B., Marcora, S. M., & Lepers, R (2013) addressed neuromuscular function directly. These different research programs show why claims about neural control must be linked to a specific operational measure.
A neural activation pattern can correlate with perceived fatigue without being its causal generator. Mediation claims require temporal precedence, specificity, and preferably targeted intervention. Even when noninvasive measures respond to cognitive exertion, their interpretation must account for physical exertion itself, motion artifacts, and changes in autonomic state.
An integrative model can treat neural, bodily, and behavioral processes as interacting levels of organization. It should specify observable predictions at each level and test where the data support convergence. Claims of a single 'central governor' or a uniquely depleted neural fuel risk substituting terminology for a discriminating mechanism.
Evidence Across Endurance, Strength, and Sport Skills
The broader mental-fatigue literature includes aerobic endurance, maximal force, resistance-exercise volume, technical execution, reaction time, and decision making. These outcomes differ in their cognitive and physiological requirements. A transient change in tactical accuracy cannot be equated with a longer time to exhaustion, and a resistance-exercise decrement does not automatically establish impairment in distance running.
Van Cutsem, J., Marcora, S., De Pauw, K., Bailey, S., Meeuse (2017) examined a spectrum of physical performance outcomes, while Brown, D. M. Y., Graham, J. D., Innes, K. I., Harris, S., Fl (2020) evaluated prior cognitive exertion more broadly. The contemporary literature increasingly distinguishes sport-specific cognitive outcomes from general physical performance. This separation improves mechanism testing by reducing outcome heterogeneity.
The central research object of this dossier is whole-body endurance and pacing. Findings from strength or sport-specific reaction tasks can inform hypotheses about cognitive regulation but should be presented as adjacent evidence. They cannot be used as quantitative estimates of endurance impairment without an appropriate translation model.
Task-specificity offers a falsifiable prediction. If mental fatigue primarily affects regulation under prolonged effort, effects may be more evident when participants must maintain difficult control and monitor an endpoint than during isolated maximal contractions. If effects instead reflect generalized arousal or expectancy, a different pattern is expected. Preregistered, within-person comparisons of multiple performance domains could help resolve this distinction.
Applied Practice, Intervention Fidelity, and Performance Ethics
Applied interventions should specify the outcome they intend to improve: endurance time, consistency of pacing, resistance to cognitive interference, perceived tolerance of effort, or decision quality under competition demands. Each outcome requires different measurement. Strategies with plausible benefit include structured self-talk, task-relevant attention, pacing rehearsal, and carefully evaluated brain endurance training.
The evidence for self-talk includes controlled cycling research by Blanchfield, A. W., Hardy, J., de Morree, H. M., Staiano, W (2014). Evidence for brain endurance training is emerging from randomized comparisons and recent syntheses by Yu, J., Tan, Z., Huang, Y., Chen, L., & Li, D (2026), Wang, N., Chen, Z., & Zhao, K (2026), and Zheng, X., Sun, Y., Zhao, Z., Wang, C., Shang, Z., Xie, Y (2026). These findings justify research and context-sensitive evaluation rather than indiscriminate application.
Intervention fidelity is essential. Investigators should record whether participants actually use the prescribed strategy, how often, at what phase of exercise, and with what task-specific understanding. Athlete preferences and accessibility must be considered; an intervention that creates additional cognitive demands can be counterproductive in some contexts.
Performance ethics includes informed consent, avoidance of coercive overexertion, truthful communication of evidence strength, and recognition that sports training is not a replacement for medical assessment of unexplained fatigue. The boundary between a healthy athlete's transient mental fatigue and clinically meaningful symptoms must remain explicit. This dossier offers scientific analysis of behavioral mechanisms, not personalized medical or training prescriptions.
Cognitive Technologies and the Artificial Era
Modern endurance systems increasingly combine wearable sensors, route prediction, exercise platforms, and automated pacing feedback. In the Artificial Era, these systems can function as external components of the information environment within which athletes regulate effort. They measure or estimate speed, power, heart rate, route characteristics, and historical performance; some also offer adaptive recommendations.
Cognitive offloading can reduce the need to remember split targets or calculate remaining distance. At the same time, continuous feedback can increase monitoring demand, divert attention from bodily cues, or encourage overconfidence in a computational estimate. The psychological effect depends on how information enters the decision process, whether the athlete can override it, and how uncertainty is represented.
Research on human–artificial coordination should distinguish algorithm accuracy from human performance. A pace recommendation might predict an efficient schedule without improving an athlete's decisions if its interface is poorly timed or its rationale is unclear. Conversely, a modest system may help by communicating a small amount of relevant information when it matters.
A defensible research program compares adaptive feedback with static feedback and no-feedback conditions while measuring cognitive load, decision latency, perceived effort, pacing variability, and finishing time. Artificial systems thereby become experimentally tractable elements of endurance regulation, connecting Sport and Performance Psychology to Human Factors and Engineering Psychology. Their role is grounded in observable effects rather than assumptions about the internal status of the system.
Causal Models and the Identification of Mechanisms
A useful causal account must distinguish manipulation, mediator, outcome, and background conditions. Let the manipulation be a randomized cognitive preload, the outcome be objective endurance performance, and a candidate mediator be perception of effort during exercise. The simplest hypothesis is that prior cognitive demand alters effort perception, which then changes pace or stopping behavior. But alternative pathways may run through expectancy, motivation, sleep, arousal, or attentional engagement.
Randomization of cognitive preload improves identification of its total causal effect on performance. It does not, by itself, identify the mechanism. An observed correlation between effort rating and finishing time can result from the fact that harder work itself raises effort. A mediation analysis must address this reciprocal structure, measurement error, unmeasured mediator–outcome confounding, and temporal ordering.
Marcora, S. M., Staiano, W., & Manning, V (2009) and Holgado, D., Troya, E., Perales, J. C., Vadillo, M. A., & Sa (2021) illustrate why total-effect replication is a necessary foundation. If the total effect is not stable across protocols, a proposed universal mediator becomes harder to defend. Nevertheless, heterogeneity may itself help identify moderators.
More discriminating designs could experimentally vary feedback, reward, attentional demands, and pacing autonomy. Their analyses should prespecify primary contrasts and permit simultaneous measurement of behavior and physiology. Rather than merely correlating neural signals with fatigue labels, the next generation of studies should test causal predictions that separate competing mechanisms.
Methodological Framework for Future Research
The strongest next experiment would be an adequately powered, preregistered, randomized crossover investigation with multiple familiarization sessions and matched cognitive-control conditions. It would specify the endurance outcome and clinically or competitively meaningful difference before data collection. Task-induced mental fatigue would be confirmed using multiple measures, while sleep, time of day, training load, temperature, nutrition, and stimulants would be standardized where feasible.
Both fixed-intensity and self-paced outcomes should be assessed when the research question concerns their differences, with sufficient spacing to avoid carryover. A time trial should record work-rate trajectories and endpoint knowledge, with repeated effort and fatigue reports placed carefully to minimize disruption. Physiological measures could include exercise heart rate, oxygen uptake, temperature where relevant, and neuromuscular indices chosen in advance to test concrete hypotheses.
Analysis would estimate total performance effects, individual variability, temporal pacing patterns, and interactions with training status or cognitive-task response. Predefined criteria for missing data and deviations would protect interpretability. Results should include full uncertainty intervals and raw or reproducible processed data, with transparent handling of crossover correlations and multiple outcomes.
Multicenter collaborations could test whether effects generalize across laboratories, endurance modalities, and athlete groups. Such collaborations should include independent null findings and active controls designed to reduce expectancy differences. Mechanistic substudy measures should be justified by their ability to discriminate alternative models, not by novelty alone.
Critical Integration of Competing Scientific Explanations
Several explanatory accounts remain empirically viable. A psychobiological interpretation emphasizes effort appraisal and potential motivation; an anticipatory pacing account emphasizes prospective allocation of work; a cognitive-control interpretation emphasizes monitoring and action regulation; an expectancy account focuses on beliefs about fatigue; and physiological accounts emphasize bodily constraints that determine feasible output.
These accounts can interact, and their predictions must nevertheless be distinguishable. A pure loss-of-capacity model predicts reduced output independent of choice and task incentives. A valuation model predicts sensitivity to rewards and the perceived cost of continued exertion. A monitoring model predicts poor adjustment when new information appears. An expectancy model predicts effects of beliefs and task framing, while environmental stress models predict interactions with heat, terrain, or exercise duration.
The evidence supports the importance of perceived exertion and anticipatory regulation in endurance decisions, documented by Borg, G. A. V (1982), Tucker, R (2009), and de Koning, J. J., Foster, C., Bakkum, A., Kloppenburg, S., T (2011). The causal proposition that prior cognitive effort consistently reduces endurance is less secure, given the contrasting findings of Marcora, S. M., Staiano, W., & Manning, V (2009), Holgado, D., Troya, E., Perales, J. C., Vadillo, M. A., & Sa (2021), and bias-sensitive review by Holgado, D., Sanabria, D., Perales, J. C., & Vadillo, M. A (2020).
A productive synthesis therefore treats mental-fatigue sensitivity as a testable interaction among task demands, prior cognitive state, subjective regulation, athlete characteristics, and environment. It should not dissolve all influences into a single abstract 'mental energy' or claim that all causal questions have already been settled.
Unresolved Questions and Scientific Research Priorities
The first unresolved issue concerns the identity of mental fatigue as an experimental construct. Which combination of subjective, behavioral, and physiological measures most reliably captures the state, and how stable is this combination across different cognitive tasks and populations? Methodological work by Goodman, S. P. J., Collins, B., Shorter, K., Moreland, A. T (2025) makes this question central to future research.
A second issue is the robustness of objective endurance effects. It is necessary to identify when a cognitive preload alters time to exhaustion, time-trial completion, pacing trajectory, or tactical decision quality. Replication needs adequate power and consistent reporting across training levels, event durations, and environments. Investigators should also determine whether an apparent impairment is better predicted by fatigue ratings, task difficulty, motivation, or expectancy.
A third issue concerns intervention transfer. Controlled self-talk and brain endurance training studies report encouraging findings, but how long do gains persist, which active controls are adequate, and do laboratory gains improve competitive results? The 2026 quantitative reviews by Wang, N., Chen, Z., & Zhao, K (2026) and Zheng, X., Sun, Y., Zhao, Z., Wang, C., Shang, Z., Xie, Y (2026) demonstrate the continuing need for larger independent studies with better certainty.
A fourth issue is dynamic causal identification. Temporal tracking of effort, expectancy, motivation, and work rate can show when pace changes, but experimental manipulations must determine why. The development of interpretable computational models and athlete-facing cognitive technologies can support precise testing if their predictions are validated prospectively.
Conclusion
Mental fatigue, pacing, and cognitive regulation form a connected research problem within the science of endurance performance. Endurance is sustained through repeated decisions about work rate under changing physiological, perceptual, motivational, and environmental conditions. The psychological organization of those decisions can be measured through effort ratings, objective performance, time-varying pacing data, attention, and anticipatory judgments.
The evidence provides a clear distinction between reliable subjective effects of demanding cognitive activity and less consistently reproduced impairments in objective exercise performance. Influential positive experiments established the problem; preregistered null findings and bias-sensitive syntheses refined the boundaries of the claim. Pacing theory shows that the same bodily state can support different actions depending on remaining distance, goal, and prospective assessment. Experimental interventions, particularly self-talk and brain endurance training, have scientific promise with important qualifications concerning effect magnitude, method, and transfer.
The resulting synthesis is a program of conditional causal explanation. The next phase requires validated fatigue induction, independent replication, temporally precise measurement, active comparators, and mechanisms that predict distinguishable outcomes. Endurance regulation is most accurately represented as organized behavior arising through the continuous interaction of bodily constraints, perceived effort, learned expectations, and decisions about future action.
References
- Blanchfield, A. W., Hardy, J., de Morree, H. M., Staiano, W., & Marcora, S. M. (2014). Talking yourself out of exhaustion: The effects of self-talk on endurance performance. Medicine & Science in Sports & Exercise, 46(5), 998–1007. https://doi.org/10.1249/MSS.0000000000000184
- Borg, G. A. V. (1982). Psychophysical bases of perceived exertion. Medicine & Science in Sports & Exercise, 14(5), 377–381. https://doi.org/10.1249/00005768-198205000-00012
- Brick, N. E., Campbell, M. J., Metcalfe, R. S., Mair, J. L., & MacIntyre, T. E. (2016). Altering pace control and pace regulation: Attentional focus effects during running. Medicine & Science in Sports & Exercise, 48(5), 879–886. https://doi.org/10.1249/MSS.0000000000000843
- Brick, N. E., MacIntyre, T. E., & Campbell, M. J. (2016). Thinking and action: A cognitive perspective on self-regulation during endurance performance. Frontiers in Physiology, 7, 159. https://doi.org/10.3389/fphys.2016.00159
- Brown, D. M. Y., Graham, J. D., Innes, K. I., Harris, S., Flemington, A., & Bray, S. R. (2020). Effects of prior cognitive exertion on physical performance: A systematic review and meta-analysis. Sports Medicine, 50(3), 497–529. https://doi.org/10.1007/s40279-019-01204-8
- Carter, E. C., Kofler, L. M., Forster, D. E., & McCullough, M. E. (2015). A series of meta-analytic tests of the depletion effect: Self-control does not seem to rely on a limited resource. Journal of Experimental Psychology: General, 144(4), 796–815. https://doi.org/10.1037/xge0000083
- de Koning, J. J., Foster, C., Bakkum, A., Kloppenburg, S., Thiel, C., Joseph, T., Cohen, J., & Porcari, J. P. (2011). Regulation of pacing strategy during athletic competition. PLOS ONE, 6(1), e15863. https://doi.org/10.1371/journal.pone.0015863
- Goodman, S. P. J., Collins, B., Shorter, K., Moreland, A. T., Papic, C., Hamlin, A. S., Kassman, B., & Marino, F. E. (2025). Approaches to inducing mental fatigue: A systematic review and meta-analysis of (neuro)physiologic indices. Behavior Research Methods, 57, Article 102. https://doi.org/10.3758/s13428-025-02620-7
- Hagger, M. S., Wood, C., Stiff, C., & Chatzisarantis, N. L. D. (2010). Ego depletion and the strength model of self-control: A meta-analysis. Psychological Bulletin, 136(4), 495–525. https://doi.org/10.1037/a0019486
- Hagger, M. S., Chatzisarantis, N. L. D., Alberts, H., Anggono, C. O., Batailler, C., Birt, A. R., Brand, R., Brandt, M. J., Brewer, G., Bruyneel, S., Calvillo, D. P., Campbell, W. K., Cannon, P. R., Carlucci, M., Carruth, N. P., Cheung, T., Crowell, A., de Ridder, D. T. D., Dewitte, S.,. Zwienenberg, M. (2016). A multilab preregistered replication of the ego-depletion effect. Perspectives on Psychological Science, 11(4), 546–573. https://doi.org/10.1177/1745691616652873
- Holgado, D., Sanabria, D., Perales, J. C., & Vadillo, M. A. (2020). Mental fatigue might be not so bad for exercise performance after all: A systematic review and bias-sensitive meta-analysis. Journal of Cognition, 3(1), Article 38. https://doi.org/10.5334/joc.126
- Holgado, D., Troya, E., Perales, J. C., Vadillo, M. A., & Sanabria, D. (2021). Does mental fatigue impair physical performance? A replication study. European Journal of Sport Science, 21(5), 762–770. https://doi.org/10.1080/17461391.2020.1781265
- Holgado, D., Sanabria, D., Perales, J. C., & Vadillo, M. A. (2021). Correction: Mental fatigue might be not so bad for exercise performance after all: A systematic review and bias-sensitive meta-analysis. Journal of Cognition, 4(1), Article 32. https://doi.org/10.5334/joc.178
- Marcora, S. M., Staiano, W., & Manning, V. (2009). Mental fatigue impairs physical performance in humans. Journal of Applied Physiology, 106(3), 857–864. https://doi.org/10.1152/japplphysiol.91324.2008
- Martin, K., Staiano, W., Menaspà, P., Hennessey, T., Marcora, S., Keegan, R., Thompson, K. G., Martin, D., Halson, S., & Rattray, B. (2016). Superior inhibitory control and resistance to mental fatigue in professional road cyclists. PLOS ONE, 11(7), e0159907. https://doi.org/10.1371/journal.pone.0159907
- McCormick, A., Meijen, C., & Marcora, S. (2015). Psychological determinants of whole-body endurance performance. Sports Medicine, 45(7), 997–1015. https://doi.org/10.1007/s40279-015-0319-6
- Pageaux, B., Lepers, R., Dietz, K. C., & Marcora, S. M. (2014). Response inhibition impairs subsequent self-paced endurance performance. European Journal of Applied Physiology, 114(5), 1095–1105. https://doi.org/10.1007/s00421-014-2838-5
- Pageaux, B., Marcora, S. M., & Lepers, R. (2013). Prolonged mental exertion does not alter neuromuscular function of the knee extensors. Medicine & Science in Sports & Exercise, 45(12), 2254–2264. https://doi.org/10.1249/MSS.0b013e31829b504a
- Ramos-Campo, D. J., Foster, C., Andreu-Caravaca, L., Rubio-Arias, J. Á., & Rosenblat, M. A. (2026). Comparative effects of pacing strategies on endurance performance: A systematic review and network meta-analysis. Sports Medicine, 56(3), 725–738. https://doi.org/10.1007/s40279-025-02367-3
- Renfree, A., Martin, L., Micklewright, D., & St Clair Gibson, A. (2014). Application of decision-making theory to the regulation of muscular work rate during self-paced competitive endurance activity. Sports Medicine, 44(2), 147–158. https://doi.org/10.1007/s40279-013-0107-0
- Smits, B. L. M., Pepping, G.-J., & Hettinga, F. J. (2014). Pacing and decision making in sport and exercise: The roles of perception and action in the regulation of exercise intensity. Sports Medicine, 44(6), 763–775. https://doi.org/10.1007/s40279-014-0163-0
- Tucker, R. (2009). The anticipatory regulation of performance: The physiological basis for pacing strategies and the development of a perception-based model for exercise performance. British Journal of Sports Medicine, 43(6), 392–400. https://doi.org/10.1136/bjsm.2008.050799
- Van Cutsem, J., De Pauw, K., Buyse, L., Marcora, S., Meeusen, R., & Roelands, B. (2017). Effects of mental fatigue on endurance performance in the heat. Medicine & Science in Sports & Exercise, 49(8), 1677–1687. https://doi.org/10.1249/MSS.0000000000001263
- Van Cutsem, J., Marcora, S., De Pauw, K., Bailey, S., Meeusen, R., & Roelands, B. (2017). The effects of mental fatigue on physical performance: A systematic review. Sports Medicine, 47(8), 1569–1588. https://doi.org/10.1007/s40279-016-0672-0
- Wang, N., Chen, Z., & Zhao, K. (2026). The effects of brain endurance training on physical endurance performance under mental fatigue and non-fatigued conditions: A systematic review and meta-analysis. Frontiers in Psychology, 17, 1858200. https://doi.org/10.3389/fpsyg.2026.1858200
- Yu, J., Tan, Z., Huang, Y., Chen, L., & Li, D. (2026). The effects of brain endurance training on mental fatigue, endurance, and cognitive function: A systematic review and meta-analysis. Frontiers in Physiology, 17, 1746298. https://doi.org/10.3389/fphys.2026.1746298
- Zheng, X., Sun, Y., Zhao, Z., Wang, C., Shang, Z., Xie, Y., Xu, K., Zhou, K., & Girard, O. (2026). Effects of brain endurance training on cognitive function, physical performance and perceived exertion in healthy adults: A systematic review and three-level meta-analysis. Frontiers in Physiology, 17, 1923855. https://doi.org/10.3389/fphys.2026.1923855