What is the principle of neuromechanical matching?
What is the principle of neuromechanical matching? Is it an established phenomena that has been well-described in the exercise science literature? Or did some fitness industry influencers just make it up because it sounded good? Let us take a look at the literature and find out.
What is the principle of neuromechanical matching?
What problem does it solve?
Before looking at any physiological principle related to any given neural phenomenon (like motor control, motor unit recruitment, or antagonist coactivation), it helps to consider what problem the phenomenon is trying to solve. For example, Henneman’s size principle solves the problem of the order in which the central nervous system switches on motor units within a muscle or muscle region. The principle of neuromechanical matching in turn solves the problem of which muscle or muscle region receives the central motor command that the motor cortex generates. In other words, if you discard the principle of neuromechanical matching, you need an alternative explanation for how the brain decides which muscle or muscle region must contribute most in an exercise (and saying “anatomy” doesn’t count because you still need to describe what feature of the anatomy is the factor that matters, like the internal moment arm length).
What is the definition?
The “principle of neuromechanical matching” can be defined as the observation that, during any given movement, the central nervous system allocates central motor command from the motor cortex to motor units (and hence muscles and muscle regions) according to their mechanical advantage. The mechanical advantage can be measured by the internal moment arm lengths of the muscle fibers within the motor units. Using muscle fibers with the most mechanical advantage means that joint torque is maximized for each unit of muscle force, and this clearly makes the movement more efficient. This makes perfect sense in evolutionary terms because animals that did not use this principle would not have been efficient and would therefore have died out quite quickly. In contrast, the animals that used this principle would have been more efficient and hence more successful since they would have used less energy to move around, find food, escape predators, and do anything else they needed to do.
What does the principle of neuromechanical matching imply in practice?
It helps us choose exercises to target a muscle
In practice, the principle of neuromechanical matching shows us how to choose the best strength training exercises for activating any given muscle simply by identifying exactly when those muscles have the best leverage for contributing to a movement. For example, we might observe that the internal moment arm of the biceps brachii is longer when [1] the elbow is extended, and [2] the forearm is supinated. Hence, a biceps curl that is done [1] with peak effort occurring in the extended elbow position, and [2] with the dumbbell held in a supinated grip (as in with the palm facing upwards) is therefore likely to cause the central nervous system to send more of its central motor command to the biceps brachii instead of the brachioradialis or brachialis muscles. This means that more biceps brachii muscle fibers are activated than brachioradialis or brachialis muscle fibers. Nevertheless, this only gives us a connection between the principle of neuromechanical matching and muscle activation. Why do we therefore often say that it allows us to choose the optimal exercises for hypertrophy? To answer that question, we must consider the importance of muscle fiber activation in permitting mechanical tension to be generated.
Muscles fibers only grow if they are activated
Why is muscle fiber activation important? Well, muscle fibers must be activated in order to experience mechanical tension during strength training exercises. You cannot cause a muscle fiber to experience active mechanical tension unless it is activated. And you cannot cause a muscle fiber to experience passive mechanical tension unless it is either [1] activated, or [2] stretched to a really, really long length like we do during static stretching. Simply going to the normal ranges of motion that we use during the vast majority of common strength training exercises is not going to be enough to create passive mechanical tension in a muscle fiber that is not activated. Hence, the principle of neuromechanical matching tells you which muscle is going to grow most as a result of doing a strength training exercise because it tells you which muscle fibers are activated. Of course, having made that point, we now have to address the predictable reactions of the commentators who once read an opinion article saying that muscle activation does not predict hypertrophy.
Muscle activation DOES predict where hypertrophy happens
Commentators who argue that muscle activation does not predict muscle growth are misunderstanding the argument here, knowingly or not. When researchers say that muscle activation does not currently predict muscle growth, they mean that surface electromyography measurements of muscle activation do not correlate with the long-term measurements of muscle size (indeed, they should be clearer about this because there are several other proxy measurements of muscle activation that do not involve surface electromyography). Therefore, the criticism that they are making relates to our technological ability to assess the number of muscle fibers that are activated in a muscle with the electromyography tools currently available to us. And these are very valid criticisms. Yet, they are not saying (or at least I hope they are not saying) that there are ways to make a muscle fiber grow during strength training that happen without needing to activate the muscle to apply active or passive mechanical tension. That certainly is not possible. Indeed, if you think that you can make a muscle fiber grow in a normal strength training exercise (not a static stretch) without activating it, then please explain how this happens.
What is the origin story of the neuromechanical matching principle?
We can identify a number of key events in the development of research into the principle of neuromechanical matching going back more than forty years, as follows:
- 2023 — researchers showed that changes in body rotation can alter the activation patterns of the spinal erectors, potentially in accordance with their leverages. Similar findings were reported for the same muscles in response to trunk flexion. These researchers cited the earlier work in the respiratory muscles, and quoted the principle of neuromechanical matching as an established phenomenon. Yet, they did not measure the leverages of the various muscles to confirm that there was a direct link between mechanical advantage and muscle activation.
- 2019 — the phrase “principle of neuromechanical matching” was first formally set forth in a review paper by Anna Hudson, Simon Gandevia, and Jane Butler. This terminology replaced previous terminologies like “functional differentiation” or “neuromuscular compartmentalization.”
- 2014—a group of researchers found that the psoas major and quadratus lumborum muscles and their various internal regions display activation levels that parallel their functions as indicated by their internal moment arm lengths. While not using the phrase “neuromechanical matching,” they did reference the literature regarding the respiratory muscles and noted that the same phenomenon applied to these larger muscles.
- 2009 — the researchers who had previously worked on the respiratory muscles extended their findings to finger muscles. By moving the thumb into different positions, they experimentally altered the length of the first dorsal interosseous muscle (since it originates on the thumb) and thereby altered its leverage for contributing to finger flexion. The activation of the muscle increased when its leverage increased, showing that the principle of neuromechanical matching was applicable.
- 2007 — the phrase “principle of neuromechanical matching” was first described in a scientific magazine article in reference to the respiratory muscles of humans and other animals.
- 2005 — a group of researchers tested the relationship between the internal moment arm lengths of various regions within the deltoids and their muscle activation levels during shoulder movements. They found a relatively close relationship, which they attributed to the concept of “functional differentiation” between the compartments according to the leverages that each region possesses. Based on the lack of any reference to the respiratory muscle research, it seems likely that this group was not aware of the similar work being done in other muscles at the time of writing up their findings.
- 2003–2011 — a series of studies were done in the human respiratory muscles, showing a direct, strong correlation between the mechanical advantage of various regions of the muscles and the extent to which those regions were activated during breathing. In these papers, the researchers repeatedly noted that motor units are recruited in direct proportion to their mechanical advantages, which is the essence of neuromechanical matching.
- 1995 — 2005 — a series of studies were carried out in canine respiratory muscles, showing a direct, strong correlation between the mechanical advantage of various regions of the muscles and the extent to which those regions were activated during breathing. While reviewing this literature, the researchers proposed that the central nervous system tries to minimize the metabolic cost of any movement by its choice of which motor units to recruit. Since metabolic cost is closely related to the degree of muscle activation, the only way to minimize the metabolic cost of a movement is to preferentially activate the muscle regions with the greatest leverage for that movement.
- 1989 — Uwe Windhorst argued that muscles were likely partioned according to their mechanical functions to achieve greater efficiency during a movement, such that some regions were activated in some movements and other regions in other movements. Greater efficiency would clearly be achieved by using those muscle regions with greater mechanical advantage. In this review, he referred to this phenomenon either as “neuromuscular compartmentalization” or “neuromuscular partitioning” but neither terminology attracted many other researchers.
- 1984 — Researchers working with small primates discovered that a facial muscle was likely partitioned in a task-dependent manner by assessing the muscle activation levels during different facial movements. They suggested that this differentiation may relate to the unique mechanical advantages of each of the various facial muscles.
- 1981 — Researchers investigating the tensor fasciae latae found that anteromedial fibers (which have better hip flexion leverage) were only active in jogging, running, and sprinting, while the posterolateral fibers (which have better hip abduction and internal rotation leverage) were active during walking as well as running. Since running activities require hip flexion to a much greater extent, this data was regarded as an indicator the two muscle regions were activated in line with their mechanical advantages. The researchers described this behavior as “functional differentiation.”
- 1977 — Researchers studying the relationship between integrated surface electromyography amplitudes and muscle force during isometric contractions in humans discovered that the various muscles were not activated identically during elbow flexion movements with differing forearm positions. Specifically, they noted that the biceps brachii activation was reduced while the forearm was pronated relative to when it was supinated, and they ascribed the difference to alterations in mechanical conditions, including the length-tension relationship. However, they did not mention mechanical advantage specifically.
Therefore, anyone claiming that neuromechanical matching is “made up” has not done the reading necessary to understand that this phenomenon has been described in varing levels of detail for over forty years.
What DIRECT evidence do we have for the principle of neuromechanical matching?
INTRODUCTION
Direct evidence for the principle of neuromechanical matching has been recorded in humans (let us ignore the animal studies simply to avoid the article getting excessively long). Direct evidence means that researchers have identified a clear statistical association between [1] the internal leverage of the muscle areas, and [2] the activation of those same areas in a movement, during a single investigation. Indeed, this was the way in which the principle was initially studied during the series of investigations that gave rise to the idea.
#1. Respiratory muscles
The inspiratory muscles include the sternomastoid, the scalenes, the parasternal intercostals, and the external intercostals. In one investigation, researchers showed that the dorsal part of the external intercostal is more strongly activated than the ventral part of the same muscle. By drawing on previously-collected data on the leverages of the regions, they showed that this difference in muscle activation was closely associated with differences in mechanical advantage between the regions. In another investigation, the researchers found that the rostral (superior) parasternal intercostal muscle in the first interspace displayed a higher level of activation than the levels in the caudal (inferior) parasternal intercostal muscles of the third and fifth interspaces. By again drawing on previously-collected data on the leverages of the regions, they showed that this difference in muscle activation was associated with differences in mechanical advantage between the regions. Later research showed that this observation was valid for the parasternal intercostal muscles during both unconscious breathing and conscious breathing tasks. Lastly, in an investigation into the sternomastoid and the scalenes, researchers showed that the scalenes displayed higher activation levels, in accordance with their superior leverage. The relative levels of activation were not affected by lung volumes, indicating that the afferent feedback from the lungs and muscles did not affect the distribution of the central motor command.
#2. Finger muscles
The first dorsal interosseous muscle of the index finger originates on the thumb and contributes to finger flexion, along with other muscles in the hand. Since it originates on the thumb, its shape (and leverage) changes with alterations in thumb position. In a study examining the principle of neuromechanical matching in this muscle, the researchers found that both leverage and muscle activation during constant external force activities were lower when the thumb was elevated compared to when the thumb was level with the fingers. Importantly, since leverage of the muscle was lower with the thumb elevated, the muscle would need to increase its level of activation in order to maintain a constant external force if it were the only muscle acting at the joint. Since other muscles are able to contribute, however, this result tells us that when a muscle has poor leverage, its level of activation is reduced so that the activation of other muscles with better leverage can be increased.
#3. Psoas major and quadratus lumborum
The psoas major and quadratus lumborum muscles and their internal regions all display different internal moment arms for spinal flexion, extension, and rotation that vary with body position. In a study examining the principle of neuromechanical matching, the researchers implemented a standard protocol for challenging the equilibrium of the trunk with rapid voluntary arm movements that causes anticipatory postural adjustments of the trunk and pelvis. Muscle activation measurements found that the psoas major (transverse process origin) behaved mostly as a spinal extensor, but the psoas major (vertebral origin) behaved moreso as a spinal rotator. This was in line with their anatomical locations and therefore their leverages. Similarly, the quadratus lumborum (anterior origin) behaved as both a spinal flexor and spinal rotator, having good leverages for both activities. In contrast, the quadratus lumborum (posterior origin) was indiscriminately activated irrespective of the direction of movement, having generally quite poor leverage for everything.
#4. Deltoids
The deltoids can be subdivided into seven functional regions, each of which has its own internal moment arm length for shoulder flexion, extension, abduction, and adduction. When researchers measured these internal moment arm lengths in cadavers, they found that the segments with the best leverage were the same segments that were activated most strongly in movements performed by living subjects. This investigation extended the principle of neuromechanical matching from respiratory, finger, and trunk muscles into muscles commonly trained in the gym. Other, indirect studies have been performed that link the leverages of the deltoids (that are already well-known from cadaver studies) with muscle activation recorded using surface electromyography.
What INDIRECT evidence do we have for the principle of neuromechanical matching?
INTRODUCTION
Indirect evidence for the principle of neuromechanical matching has also been recorded in humans (let us ignore the animal studies simply to avoid the article getting excessively long). Indirect evidence here means that the identification of [1] the internal leverage of a muscle, and [2] the activation levels of that same muscle were recorded in separate investigations. For example, one paper may have recorded the way in which the leverage of a muscle changes over a joint angle (relative to other muscles), while another investigation may have measured changes in the activation of that muscle over the same joint angles. We can compare these two changes and show that they follow the same basic trends, with the location of peak leverage corresponding to the location of peak muscle activation.
#1. Gastrocnemius and soleus
The creators of the principle of neuromechanical matching evidently considered that we can use indirect evidence since they referred to the gastrocnemius in their review paper. The study they cited only recorded that the gastrocnemius displayed greater muscle activation when it was lengthened compared to when it was shortened, and did not also record the leverages in the same paper. Nevertheless, other data do show that the gastrocnemius increases its leverage with increasing muscle length. Also, it is noteworthy that the soleus accordingly increases its activation while the gastrocnemius reduces its leverage as it shortens. So there is a very clear transition between the gastrocnemius and the soleus as leverage increases in one and decreases in the other.
#2. Gluteus maximus (and hamstrings)
The gluteus maximus has its best leverage when the hip is fully extended (indeed, the leverage increases gradually from peak hip flexion through to full hip extension). The greatest activation levels of the gluteus maximus also occur when the hip is fully extended (and again, activation increases gradually from peak hip flexion through to full hip extension). In contrast, the hamstrings have their best leverage when the hip is flexed to 45 degrees and they are most strongly activated during 45-degree back extensions, at least in comparison with stiff-legged deadlifts.
#4. Biceps brachii
The biceps brachii (long and short head) experience better elbow flexion leverage while the wrist is supinated, but the brachioradialis is advantaged by wrist pronation. This then corresponds to improved biceps brachii voluntary activation levels in maximum elbow flexion efforts when the wrist is placed in a supinated position. Also, the biceps brachii have better elbow flexion leverage while the elbow is extended, but the brachioradialis is advantaged by elbow flexion. This corresponds to comparatively more biceps brachii muscle activation with an extended elbow and relatively greater brachioradialis muscle activation with a flexed elbow.
#5. Triceps brachii medial head (distal region)
The triceps brachii medial head distal region (as well as the triceps brachii lateral head proximal region) have increasingly better leverage than all the other parts of the triceps brachii complex as the elbow is extended. When research has investigated the muscle activation of the medial head distal region, it has shown that it similarly increases towards full elbow extension relative to the activation of several other triceps brachii regions.
#6. Latissimus dorsi
The latissimus dorsi can be subdivided into three major regions: the superior, thoracic region, the middle, lumbar region, and the inferior, pelvic region. The internal moment arm lengths and muscle activation levels of each of the latissimus dorsi regions are well-described in the literature for the frontal and sagittal planes of motion. Therefore, we can compare them to see if they behave in the same way.
- Sagittal plane (extension) — Internal moment arm length data show that the thoracic region is the best shoulder extensor in the sagittal plane. This observation matches with multichannel surface electromyography data assessing muscle activation levels. Moreover, we can see that the latissimus dorsi has its worst leverage for the shoulder extension in the sagittal plane when the arms are overhead. This corresponds closely with the joint angle at which minimum levels of muscle activation occur during sagittal plane pull-down and pull-over exercises. Also, the exact same directional trends are observed. Both the internal moment arm length and the levels of muscle activation increase gradually as the arm is lowered in the sagittal plane.
- Frontal plane (adduction) — Internal moment arm length data similarly show that the lumbar and pelvic regions are the best shoulder adductors in the frontal plane, which also matches with the multichannel surface electromyography data assessing muscle activation levels.
#7. Pectoralis major
The pectoralis major can be subdivided into three major regions: the superior, clavicular head, the middle, sternal head, and the inferior, costal head. The internal moment arm lengths and muscle activation levels of each of the pectoralis major heads are well-described in the literature for the frontal and sagittal planes of motion. Therefore, we can compare them to see if they behave in the same way.
- Sagittal plane (flexion) — Internal moment arm length measurements show that the clavicular head alone is a good shoulder flexor in the sagittal plane. This observation matches with the multichannel surface electromyography data assessing muscle activation levels. The high involvement of the clavicular pectoralis major during shoulder flexion, especially up to 90 degrees of shoulder angle (arm parallel to the floor), can be seen in studies looking at front raises, where the muscle is very clearly involved to a greater extent than the anterior deltoid.
- Sagittal plane (extension) — The costal head of the pectoralis major has its best leverage for the shoulder extension in the sagittal plane when the arms are overhead. This observation matches with the multichannel surface electromyography data assessing muscle activation levels. It also corresponds closely with the joint angle at which maximum levels of muscle activation occur during sagittal plane pull-down and pull-over exercises. Moreover, the same directional trends are observed. Both the internal moment arm length and the levels of muscle activation decrease gradually as the arm is lowered in the sagittal plane.
- Frontal plane (adduction) — Internal moment arm length data show that both the sternal and costal heads are excellent shoulder adductors in the frontal plane, which also matches with the multichannel surface electromyography data assessing muscle activation levels.
#8. Hip flexors (rectus femoris, psoas, and iliacus)
During hip flexion, the leverage of the psoas and iliacus increases progressively as they shorten with increasing hip flexion (while the relative leverage of the rectus femoris correspondingly decreases). Similarly, as regards muscle activation, the activation levels of the psoas and iliacus also increase progressively with increasing hip flexion. Meanwhile, the muscle activation of the rectus femoris is relatively greater when it is lengthened and decreases progressively as it shortens with increasing hip flexion.
What is the conclusion?
The exact phrase “the principle of neuromechanical matching” was first proposed by a research group in 2007 to refer to the observation that the distribution of central motor command during a movement is matched to the mechanical advantage of the motor units involved. Yet, the basic idea that motor units are recruited in accordance with their leverage was suggested many years earlier. Direct evidence has been recorded in four places: [1] respiratory muscles, [2] finger muscles, [3] psoas major and quadratus lumborum, and [4] deltoids. Direct evidence means there is an association between the leverage of the muscle regions and muscle activation in a movement reported in a study. Indirect evidence has been found in eight places: [1] gastrocnemius and soleus, [2] hip extensors (gluteus maximus and hamstrings), [3] the upper gluteus maximus region, [4] biceps brachii, [5] triceps brachii medial head distal region, [6] latissimus dorsi regions, [7] pectoralis major heads, and [8] hip flexors (rectus femoris, psoas, and iliacus). Indirect evidence means that separate studies have reported the leverages and the muscle activations, but they are still located in the same place as each other. There is no better hypothesis (nor a hypothesis with more evidence behind it) for explaining how the central nervous system allocates neural drive from the brain to the various working muscles during an exercise.