Loading field map
What actually limits human performance? Is it the heart's ability to pump oxygen, the muscles' capacity to use that oxygen, the accumulation of metabolic byproducts, or something in the brain that decides when to stop? For over a century, exercise physiologists have built competing explanatory frameworks to answer this question. The field's history is not a steady accumulation of facts but a series of conceptual shifts—from whole-body oxygen delivery to peripheral metabolic limits, then to central regulation and, most recently, to integrative models that try to hold all these levels together.
The earliest systematic investigations of exercise were largely descriptive. Researchers measured heart rate, breathing, and body temperature during physical work, cataloging what happened without a unifying theory of limitation. This era established the basic tools of the trade—treadmills, ergometers, gas collection—and posed the question that would drive the field: what changes in the body when it moves from rest to exertion? The answers were piecemeal, but the methodological foundation was laid for the first major explanatory framework.
A decisive shift came with the work of A. V. Hill and colleagues in the 1920s. They proposed that maximal oxygen uptake (VO2max) set an upper limit on endurance performance. In this view, the cardiovascular system's ability to deliver oxygen to working muscles was the bottleneck; if you could increase oxygen delivery, you could increase performance. This was the first genuinely mechanistic framework in exercise physiology, and it gave the field a central concept that still anchors much of applied sport science.
The Harvard Fatigue Laboratory (1927–1947) served as the institutional engine that scaled up Hill's oxygen-transport framework. Under the direction of D. B. Dill, the lab conducted large-scale studies on environmental stress, nutrition, and fatigue in real-world conditions. It trained a generation of physiologists who carried the oxygen-transport paradigm into university departments and military research centers. When the lab closed in 1947, its legacy was a firmly established view: exercise limitation was primarily a matter of oxygen supply and demand.
By the 1960s, researchers began to notice that the VO2max model could not explain everything. Athletes with identical VO2max values often performed very differently, and the sensation of fatigue during intense exercise seemed to track something other than oxygen consumption. This dissatisfaction produced a cluster of four frameworks that emerged roughly simultaneously, each redirecting attention from whole-body oxygen delivery to what was happening inside the muscle.
The Lactate Threshold Paradigm (1964–present) identified a specific exercise intensity at which blood lactate began to accumulate exponentially. This threshold, it argued, marked the point at which the muscle's demand for energy outstripped its aerobic capacity, forcing reliance on anaerobic glycolysis. The framework gave coaches a practical tool for setting training intensities and offered a new explanation for fatigue: it was not just oxygen lack but the metabolic consequences of exceeding a critical intensity.
At almost the same time, the Critical Power Model (1965–present) proposed a different threshold. It defined a power output that could be sustained indefinitely (the critical power) and a finite work capacity above that level (the W′). This model framed fatigue as the depletion of a finite anaerobic reserve, rather than a simple lactate inflection. The Critical Power Model and the Lactate Threshold Paradigm coexist today as complementary but distinct threshold concepts—one based on lactate kinetics, the other on work-rate asymptotes—and researchers often use both to characterize an athlete's profile.
The Energy Systems Framework (1965–present) provided the metabolic vocabulary that tied these threshold ideas together. It divided energy production into three systems: the phosphagen system (immediate, high-power), glycolysis (short-term, anaerobic), and oxidative phosphorylation (long-term, aerobic). This framework did not directly challenge the VO2max model but rather gave it a finer-grained substrate: the relative contribution of each system depended on exercise intensity and duration, explaining why different events required different training emphases.
Rounding out this cohort, the Fiber-Type Paradigm (1966–present) linked muscle composition to performance. It classified skeletal muscle fibers into slow-twitch (Type I, oxidative) and fast-twitch (Type II, glycolytic) categories, showing that an individual's fiber-type distribution influenced both their metabolic profile and their response to training. This framework absorbed the Energy Systems Framework's logic at the cellular level: Type I fibers relied on oxidative metabolism, Type II fibers on glycolysis. Together, these four frameworks shifted the field's center of gravity from the heart and lungs to the muscle itself.
If the metabolic threshold frameworks pointed to the muscle as the site of limitation, the Peripheral Muscle Fatigue Model (1970–present) made that claim explicit. It argued that fatigue originated within the contracting muscle—through accumulation of inorganic phosphate, hydrogen ions, or other metabolites that interfered with cross-bridge cycling and calcium release. This model directly reacted against the oxygen-transport framework: even if oxygen delivery was adequate, the muscle could still fail from within. The model drew heavily on the emerging tools of Exercise Biochemistry (1967–present), a methodological school that applied biochemical assays to muscle biopsies and blood samples. Exercise Biochemistry provided the empirical evidence for peripheral fatigue mechanisms, measuring changes in ATP, phosphocreatine, lactate, and pH during exercise. It did not replace the VO2max model but narrowed its scope: oxygen delivery mattered, but so did the biochemical environment inside the muscle.
The biochemical approach deepened further with the rise of Molecular Exercise Physiology (1990–present). Instead of measuring metabolites, this framework examined the signaling pathways and gene expression changes triggered by exercise. It asked how a single bout of exercise or a period of training altered the molecular machinery of the muscle cell—upregulating mitochondrial enzymes, shifting fiber-type profiles, and activating transcription factors like PGC-1α. Molecular Exercise Physiology derived directly from Exercise Biochemistry's tools but added a new layer of explanation: the long-term adaptations that underpin training effects. It did not challenge the peripheral fatigue model but explained why repeated exposure to exercise changed the muscle's capacity to resist fatigue.
By the early 2000s, a growing body of evidence suggested that peripheral fatigue models were incomplete. Athletes often stopped exercising before their muscles had reached a true physiological failure point, and the sensation of effort seemed to be regulated in advance. The Central Governor Model (2004–present), proposed by Timothy Noakes, argued that the brain unconsciously anticipated the metabolic demands of the exercise bout and reduced muscle recruitment to prevent catastrophic failure. In this view, fatigue was not a failure of the muscle but a calculated safety margin set by the central nervous system. The model reacted sharply against the Peripheral Muscle Fatigue Model, which it accused of ignoring the brain's role.
The Psychobiological Model of Fatigue (2008–present), developed by Samuele Marcora, agreed that fatigue was centrally regulated but disagreed on the mechanism. Marcora argued that the decision to stop or slow down was not an unconscious calculation but a conscious choice based on perceived effort. When the effort required to continue exceeded the athlete's motivation, they quit. This model derived from the Central Governor Model's emphasis on central regulation but replaced an unconscious governor with a conscious decision-making process. The two frameworks remain in active disagreement: Noakes posits a subconscious regulator that protects the body, while Marcora insists that fatigue is a conscious sensation that can be overridden by motivation. Both, however, have permanently expanded the field's scope beyond the muscle.
The most recent framework, Integrative Biology of Exercise (2014–present), attempts to reconcile the peripheral and central traditions. It treats the exercising human as a system of interacting subsystems—cardiovascular, metabolic, muscular, neural, and psychological—none of which can be understood in isolation. Fatigue, in this view, is an emergent property of the whole system, not a single cause. The framework absorbs the oxygen-transport model's focus on delivery, the metabolic threshold models' attention to muscle energetics, the molecular framework's interest in adaptation, and the central regulation models' insistence on the brain. It does not reject earlier frameworks but positions them as partial accounts that need to be integrated. This is the leading framework today, not because it has settled all debates but because it provides a language for studying how the parts interact.
Exercise physiology today is not a unified field with a single accepted model. The Oxygen Transport and VO2max Model remains essential for understanding endurance capacity and for practical testing in sport. The Lactate Threshold Paradigm and Critical Power Model coexist as alternative threshold concepts, each with its own measurement protocols and training applications. The Energy Systems Framework provides the shared vocabulary for discussing exercise intensity. The Fiber-Type Paradigm explains individual differences in trainability. Exercise Biochemistry and Molecular Exercise Physiology continue to uncover the cellular and molecular mechanisms of adaptation. The Peripheral Muscle Fatigue Model still guides research on muscle function, while the Central Governor and Psychobiological Models have opened up the study of pacing, effort perception, and mental fatigue.
What the leading frameworks agree on is that performance limitation is multifactorial—no single bottleneck explains all cases. They disagree on which factor is most important in a given context and on the level of analysis that should be prioritized. The Integrative Biology of Exercise framework tries to hold all these levels together, but it is more a research orientation than a settled theory. The field remains pluralistic, and that pluralism is a strength: each framework offers a different lens, and the best science comes from knowing when to use which one.