July 1, 2021

The Mechanisms of Training Fatigue


Training is simultaneously a stimulus and a stressor to the body. Intelligent training is then a balancing act of providing a sufficient stimulus to elicit an adaptation while not burying yourself for subsequent sessions.

We’re going to explore what that looks like from both a physiological and practical perspective. Before we dive in, it’s important to understand the idea of an allostatic load and the many bio-psycho-social influences that affect your ability to perform on any given day. Our allostatic load encompasses the totality of stressors the body is taking on or exposed to on a consistent basis. None of us are input-output machines. To some degree, we all have some sort of stress from sources outside the gym. External stressors can have just as much (if not more) influence on your session’s performance than your ability to recover from training itself. That’s for another discussion though.


In this article, we are going to examine fatigue (and its mechanisms) from an objective and quantifiable perspective. For the time being, we are only considering fatigue that results from training stress for the sake of simplicity.

To derive a functional definition of fatigue, we must start at the neuromuscular level and cover some basic principles of fiber recruitment and muscle contraction.

The term “neuromuscular” literally refers to nerves and muscles and how they interact. The central nervous system (brain and spinal cord) must fire a signal to a given amount of muscle fibers to coordinate voluntary movement. That signal or “impulse” must travel on a path from the brain, to the spinal cord, out to the peripheral nervous system, and to the muscle. This is largely facilitated by a motor neuron (nerve cell), a structure that innervates a group of muscle fibers. We can call this site (where the two meet) a motor unit. (Note: there can be several motor units per muscle)

These structures are responsible for initiating the process of muscle contraction. When a motor unit fires the impulse, every fiber within that group is activated or “recruited” to produce force. The process of motor unit recruitment can vary the amount of force produced individually (in a single motor unit) and by varying the number of motor units recruited. The ability to modulate force output is important for coordinated movement and performance. Different types of motor units, low-threshold and high-threshold motor units, are designed for this reason.

Within this process of recruitment exists an orderly pattern that is explained by the Size Principle. At first, low-threshold motor units recruit fibers with predominantly high-fatigue, low force capacities (i.e. slow twitch muscle fibers). Depending on force demands and state of fatigue, high-threshold motor units are then cycled in with fibers much greater at producing force (i.e. fast twitch fibers) to accomplish the task. This process happens within a set or bout of effort taken to or close to muscular failure (1).

Acutely, the inability to perform the task or perform at a level previously capable of would be an observation of fatigue. For example, the first few reps within a set will move with a greater velocity than the last few will leading up to muscle failure. Assuming maximal force is put into each rep, we can distinguish fatigue by the involuntary reduction of velocity. This reduction in force producing capacity is how we can objectively define training fatigue.

We can see that fatigue must result from changes within the central nervous system (central mechanisms) and/or changes within muscle itself (peripheral mechanisms).


Peripheral fatigue is observed when reductions in performance result from changes occurring to structures within the actual muscle fiber (2). As we go through repeated muscle contractions, the accumulation of metabolic byproducts (or metabolites) can interfere with biochemical reactions that need to occur to provide energy to the muscle fiber. When peripheral fatigue rises, the central nervous system compensates by recruiting more motor units (as we just discussed). Note: peripheral fatigue can occur in the absence of metabolite accumulation (i.e. decreased release of calcium). This mechanism of fatigue makes its most significant contribution within the training session itself and doesn’t seem to linger around much longer.

With the goal of hypertrophy or strength, this mechanism of fatigue is very important. We need those high-threshold motor units cycled in so that we can train the fibers with the greatest growth potential. Sure, low-threshold fibers will respond to some degree, but because we use them with every movement throughout the day, they already have a higher training status and won’t grow as much. So the question then becomes, should every set be taken to failure? And the answer would be: not necessarily. Reaching task failure isn’t always reflective of full motor recruitment (more on that to come).


Central fatigue arises due to changes occurring within the central nervous system, specifically, a shift in concentration of important neurotransmitters (i.e. norepinephrine, dopamine, acetylcholine) and a reduced ability to recruit high-threshold motor units (2).

Central fatigue rises throughout a training session and can stay around for multiple days. It most notably occurs from high volume workouts and seems to be linked to muscle damage. Sessions that cause a lot of muscle damage (i.e. eccentric contractions) trigger CNS fatigue through “afferent feedback.” As a protective mechanism, when damage is high enough, your neuromuscular system goes into a feedback loop by signaling your brain to stop firing at 100% to prevent further damage.

There are many variables at play that could influence the relative contribution of central fatigue to performance decreases (i.e. the size of the muscle being trained, exercise type, rep ranges, and the session’s total volume). As just mentioned, if central fatigue rises high enough, the ability to reach full motor unit recruitment can be reduced or inhibited.

The nature of central fatigue may provide us with a viable explanation for some of the findings on muscle growth. For example, lighter load training taken to muscular failure causes more central fatigue than heavier loads taken to muscular failure. This may explain why using training loads any lower than ~25-30% 1RM has been demonstrated to be suboptimal, even when taken to failure (3).

Low intensity/higher volume sessions seem to result in greater central fatigue. It seems to be most notable in endurance athletes which would follow the same logic (very high, voluminous sessions with lower loads) (4). If a weekly set volume threshold exists for hypertrophy, perhaps this is the mechanism behind it.


To tie everything together… Within a training session, any bout of effort or any set done causes some level of central fatigue, peripheral fatigue, and some degree of muscle damage. Knowing this is important so that we can properly go about juggling the three.

In the context of training for muscle growth, we absolutely need to reach high levels of motor unit recruitment so that we can adequately train those type II fibers (with the highest growth potential). We can think of those final ~4-5 reps leading up to muscular failure as “effective reps” or reps done under full recruitment (credit: Borge Fagerli and Chris Beardsley). Within a training session and throughout a training week, there is a minimum amount of “effective reps” or “effective training” that needs to take place to stimulate an adaptation.

That means, a certain amount of peripheral fatigue is required within each session in order to be maximally effective. So does that mean we should just take every set to muscular failure? Well, it’s not that simple. Theoretically… yes, that would maximize the number of effective reps, but there is a cost for taking things too far.

The key word(s) here are: stimulus and stressor. The greater the stimulus, the more stressful it is (all else being equal) and the less (of that stimulus) your body can handle. Training to failure has been observed to be substantially more stressful and damaging than training just shy of failure (5,6), and we’ve already discussed the negative implications of muscle damage. At high enough levels of muscle damage, central fatigue may inhibit full motor unit recruitment from being reached even though you hit “task failure” (7). However, this doesn’t mean there isn’t a place for failure training (we will talk much more about that at a later time).

This is where “the balancing act” becomes important. To reiterate the intro: we need to provide a sufficient stimulus that warrants an adaptation. Our goal in training is to trigger a protein synthetic cascade – the first step that eventually leads to muscle growth. However, our bodies can only handle so much stress productively, which means everything must come in moderation.

Think of it like this – within a training session, we have limited resources to spend, and past that point, the work becomes unproductive. If we budget correctly, we can get more out of less and improve a session’s “efficiency.”

Throughout a training session, central fatigue will rise, meaning it becomes much more difficult to maximize a movement the later it is in the training session. From a practical standpoint, we need to place the most demanding exercises (or our priority muscle group) early in the session.

To order types of movements from most centrally demanding to least:

Bi-lateral compounds > Uni-lateral compounds > Bi-lateral single-joint/isolations > Single-limb single-joint/isolations

Free weights seem to be more demanding than machines simply because there are more muscles involved and a greater coordination requirement. Generally, less coordination is required in more stable environments. Stability is a major component in force production. For example, your pushing muscles will be able to generate a lot more force on something like a seated machine shoulder press compared to an exercise like a standing barbell overhead press. In the freeweight option, greater coordination is required to provide a stable base to generate force. That means, the amount of energy or resources devoted to force production will depend greatly on movement and it’s set up. Other various considerations include: a movements range of motion, the size of muscle being trained (8,9), and the load being used (i.e. high rep sets cause more central fatigue than lower rep sets).

In most cases, compounds should come before isolations, and bilateral work should come before unilateral work. Front-loading the most demanding exercises is a good idea for a couple reasons: when we consider the idea of effective reps and motor unit recruitment, we can reach full recruitment through two mechanisms – heavy loading and/or a high enough state of fatigue.

Full recruitment happens much sooner with higher loads (lower rep sets). This means it becomes less important to take a set to failure on a heavy compound lift to get a hypertrophic stimulus… and the literature would support that statement. So if at least one of those criteria (load or fatigue) needs to be reached, it just makes sense to leave the heavy loading for compounds. On the flipside, trying to use really heavy loads on single-joint movements can become injurious and impractical.

Consider this, a heavy compound would be maximally fatiguing if it were to be taken to failure. We can offset this by leaving the heavier loading for earlier in the session, thus not requiring complete muscular failure (although getting very close). Although still fatiguing, it’s not maximally fatiguing. If we were to derive a fraction of how fatiguing that example is in proportion to how fatiguing it could be (whatever that arbitrary value may be), it may be a good idea to maintain that fraction or percentage of fatigue across the entire session. As we know the absolute loading doesn’t need to be super specific to cause growth, we can let the absolute loading fall from exercise to exercise as we move from compound to isolation, but still maintain a similar fatigue cost by increasing the level of effort per set the simpler the movement is and the later we get in the session.

Exercise sequencing seems to be pretty straightforward and will take care of itself. I would say the same about load selection and sequencing too.

The rest of a session’s/week’s organization is much more thought-provoking in my mind. There isn’t any direct evidence we can refer to regarding volume allocation and hypertrophy. However, we do have evidence from Dr. Mike Zourdos’ lab suggesting how we organize training stress throughout the week can have a significant impact on strength performance. I’m going to speculate and say it probably does matter from a growth perspective, and if there are better/worse ways to distribute the training stress throughout the week, the same is probably true on the per-session level.

So many questions then come to mind… If we have a minimum effective dose of X number of sets on a given muscle group, should more sets be distributed to lower demanding exercises or higher demanding exercises? Should we aim to maximize effective reps while minimizing ineffective reps? Should we bias our training towards heavier loading/lower set number/higher proximity to failure? And several more…

I don’t know the answer to any of those questions, but it’s certainly interesting to think about.


  1. Henneman E. Relation between size of neurons and their susceptibility to discharge. Science. 1957 Dec 27;126(3287):1345-7.
  2. Bigland‐Ritchie BW, Woods JJ. Changes in muscle contractile properties and neural control during human muscular fatigue. Muscle & Nerve: Official Journal of the American Association of Electrodiagnostic Medicine. 1984 Dec;7(9):691-9.
  3. Lasevicius T, Ugrinowitsch C, Schoenfeld BJ, Roschel H, Tavares LD, De Souza EO, Laurentino G, Tricoli V. Effects of different intensities of resistance training with equated volume load on muscle strength and hypertrophy. European journal of sport science. 2018 Jul 3;18(6):772-80.
  4. Thomas K, Goodall S, Stone M, Howatson G, Gibson AS, Ansley L. Central and peripheral fatigue in male cyclists after 4-, 20-, and 40-km time trials. Medicine & Science in Sports & Exercise. 2015 Mar 1;47(3):537-46.
  5. Zourdos MC, Klemp A, Dolan C, Quiles JM, Schau KA, Jo E, Helms E, Esgro B, Duncan S, Merino SG, Blanco R. Novel resistance training–specific rating of perceived exertion scale measuring repetitions in reserve. The Journal of Strength & Conditioning Research. 2016 Jan 1;30(1):267-75.
  6. Morán-Navarro R, Pérez CE, Mora-Rodríguez R, de la Cruz-Sánchez E, González-Badillo JJ, Sanchez-Medina L, Pallarés JG. Time course of recovery following resistance training leading or not to failure. European journal of applied physiology. 2017 Dec;117(12):2387-99.
  7. Muddle TW, Colquhoun RJ, Magrini MA, Luera MJ, DeFreitas JM, Jenkins ND. Effects of fatiguing, submaximal high‐versus low‐torque isometric exercise on motor unit recruitment and firing behavior. Physiological reports. 2018 Apr;6(8):e13675.
  8. Rossman MJ, Venturelli M, McDaniel J, Amann M, Richardson RS. Muscle mass and peripheral fatigue: a potential role for afferent feedback?. Acta physiologica. 2012 Dec;206(4):242-50.
  9. Vernillo G, Temesi J, Martin M, Millet GY. Mechanisms of fatigue and recovery in upper versus lower limbs in men. Medicine & Science in Sports & Exercise. 2018 Feb 1;50(2):334-43.
Josh Bradshaw, MS