Synergies
And how to explore them with a four-step template
In the science of motor control, the word "synergy" refers to coordinated activity of muscles, joints, or neural signals that work together to perform a particular task. The term originates (as most good concepts about motor control do!) with the Russian physiologist Nikolai Bernstein’s recognition that coordinated movement of the body cannot possibly arise from the brain sending motor commands that specify the exact actions of each individual muscle. This would be impossible because:
The body has many joints and muscles and a near-infinite number of degrees of freedom.
The body is made of soft, elastic stuff whose movements are very hard to predict.
Nervous system signals always involve noise, which creates errors in proprioception and motor commands.
Errors must be constantly corrected on the fly during movement, but nervous signals from the periphery to the brain travel too slowly for such corrections to occur in time.
Given these challenges, Bernstein proposed that the nervous system simplifies the problem of controlling movement by relying on "synergies" - groups of muscles and joints that work together in coordination. These synergies serve as building blocks that can be combined to create more complex movements. For example, the cross-lateral patterns used in crawling are also used in walking, and the triple extension seen in squatting is also used in jumping and running.
Since Bernstein's work, the word synergy has been used in many different contexts, and often with slightly different meanings. The motor control scientist Mark Latash has written extensively on the subject, including a book I'm reading right now called Synergy. This prompted me to write about the subject, including how I use the idea to develop simple templates for exploring movement (see below for a detailed description.)
The choir analogy
Latash explains synergies through analogy to a choir conductor trying to control the overall volume of a choir. A very bad control strategy would be for the conductor to individually instruct each singer on the exact volume they should sing. This process could theoretically create the correct volume, but would involve a ton of work, and correcting errors would be a huge problem. If one singer failed to show up for practice, the conductor would need to issue a completely new set of instructions to the other singers to compensate. And if during the performance one singer failed to sign at the right volume, it would be impossible to make the necessary corrections on the fly.
The better strategy to control volume would involve a single command to the whole choir that made them focus on a precise goal: each singer should attend carefully to the overall volume, and then sing just a little louder if the volume is low, and a little softer if the volume is high. This ensures that every singer is constantly working to correct potential errors by other singers. Note that this process does not specify any particular contribution from any particular singer, and there are thousands of different combinations of individual volumes that the choir can use to reach a solution. The conductor doesn't care which of these solutions are used, only that the individual singers co-vary their contributions in a way that stabilizes the overall volume. According to Latash, this is a good example of a synergy, which basically means coordinated sharing of a task.
Similarly, in motor control, synergies allow muscles and joints to work together and share the task of stabilizing the performance of a particular movement or posture. Rather than the brain controlling each element independently with specific commands, it can provide a general command related to the goal, and the elements within the synergy can automatically adjust and compensate for each other to achieve that goal.
For example, if you want to pick up a wine glass with five fingers and hold it steady, the fingers need to provide a certain amount of force against the glass. It doesn't matter whether the total force comes from all the fingers pressing equally, or some more than others, or even whether the fingers are changing their relative contributions from moment to moment. All that matters is that the fingers have formed a synergy and are able to co-vary the pressure in some combination that keeps the total amount of pressure constant.
With these examples in mind, Latash provides the following technical definition of a synergy: "a task-specific covariation (or sharing) of elemental variables that stabilizes a particular performance variable." In the context of motor control, the elemental variables are muscle activations, or joint angles, or patterns of neural activity, and the performance variable is the movement task you want to accomplish.
How synergies are created
So how does the body ensure that there is coordinated sharing and error compensation amongst the members of a synergy? Several mechanisms could contribute to the local coordination of elements within a synergy:
Reflexes can provide rapid, automatic adjustments based on sensory feedback. For example, the stretch reflex can help maintain posture and balance by automatically contracting a muscle when it is stretched, while reciprocal inhibition can prevent opposing muscles from working against each other.
Central Pattern Generators (CPGs) are neural circuits within the spinal cord that can generate rhythmic motor patterns without requiring input from higher brain centers. They are thought to play a role in the coordination of synergies in rhythmic movements like walking, running, or swimming.
Modular motor control means the nervous system issues signals that specify pre-structured combinations of muscle activations, rather than controlling each muscle independently. In other words, one neural signal could implement many actions.
The biomechanical constraints of the musculoskeletal system, such as the shape of joint capsules or the arrangement of muscles, can contribute to the coordination of synergies by making certain coordination patterns more likely or efficient. For example, two-joint muscles like the hamstrings and rectus femoris will coordinate the hip with the knee, while the gastroc coordinates the knee and ankle.
Co-contractions can create predictable and orderly patterns of functional movement, based on a simple command that simultaneously contracts all the muscles surrounding a joint. This creates a tug of war, whereby each muscle is trying to pull the joint in its direction. Any muscle that is “winning” the battle will get shorter, and then also weaker, because strength depends on optimal length. This allows the antagonist muscle to pull the joint back until some neutral equilibrium is reached. This neutral position will be highly functional for many activities. Some examples identified by Frans Bosch include: a “hip-lock” position that is seen in the mid-stance of running; a 90° abducted position of the shoulder, which is used in throwing a punch or a ball; and a “long spine” position which is useful for … nearly everything. All of these useful positions are obtained by the computationally simple order of a co-contraction.
The above physiological mechanisms likely work together and interact to enable the coordination of synergies.
Synergies, attractors, primal patterns and muscle chains
The synergy concept aligns well with some other common concepts that are used to understand human movement, including attractors, primal patterns, and muscle chains.
The term “attractor” (as used in dynamic systems theory or by Frans Bosch) refers to patterns of muscular activation or joint movement that stabilize motor control under variable conditions and in the face of perturbation. Thus, attractors are analogous to synergies - both involve patterns of cooperation in elemental variables that help to stabilize a performance variable. Both concepts acknowledge that patterns of sharing may emerge in a bottom-up fashion without the need for top-down central control from the brain.
Primal patterns (also called movement primitives, fundamental movement patterns, etc.) are simply synergies or attractor states that are in some sense innate, and which emerge naturally in children under conditions of normal motor development. They include movements like squats, crawling, walking, running, jumping, climbing, and reaching. These primal patterns are like building blocks that can be flexibly combined to form more complex movements.
The concept of muscle chains or slings (as popularized by Francois Meziere, Paul Chek, and Tom Myers) is an attempt to define muscular synergies that span the whole body, and which coordinate to perform primal movements related to gait, reaching, postural control, and full body flexions, extensions, rotations and side-bending.
How to explore synergies
So how can we apply this information in a practical way?
Here’s a rough template you can use the next time you are working on your movement patterns, or doing exercises that are aimed at improving mobility, stability, or coordination. The template is a way to generate a wide variety of interesting movement explorations that can (1) teach you how you move and (2) help you move better. There are four steps:
pick a fundamental or primal movement pattern;
precisely define a performance variable related to that pattern;
while executing the movement, observe the behavior of the elemental variables that form the synergy executing the movement;
play with constraints that create variations in the way the elemental variables must coordinate to perform the task.
Here’s more detail on each step.
Step 1: pick a fundamental pattern
If you look at exercises that are designed to improve mobility, posture, or general movement patterns, you will see they very often involve some of the synergies discussed above, for example:
primal patterns like squats, lunges, reaches, or quadruped movements;
stretches to global muscle chains like forward bends, back bends, or rotations;
postures that challenge the stability of a muscle chain, or that invite co-contractions in the spine, hips and shoulders, such as planks, side-planks, bridges, bird-dogs, single-leg stance, or standing yoga poses.
Any of these movements (and many more) are great starting points for exploring synergies.
Step 2: define the performance variable with precision
To effectively challenge and improve your synergies, you need to identify a relevant performance variable that you want to stabilize. Ask yourself: what is the real-world purpose of the movement you are doing? What specific outcome are you trying to achieve? Having a well-defined goal will place greater demand on the neuromuscular system. And you will notice that this creates a very different experience from just going through the motions.
For example, when squatting, you could do so with a vague intention to just go down and then back up. This places very little demand on the major joints (ankles, knees, hips, and trunk) to cooperate in a specific pattern. By contrast, you could introduce a precise target and clear intention by placing a stool at a particular height and requiring your butt to touch the stool on each rep. You could demand even more precision by requiring the touch to happen on a specific part of your butt, such as the sit bones, or the hamstrings. Or you could require the touch to be very light. An alternative performance variable might focus on balance - try to squat up and down while keeping the pattern of pressure under your feet uniform, so that you do not feel even the slightest pressure shift forward/back, left/right, or inside/outside.
In movements that are mostly about creating range of motion, such as a forward bend or rotation, a good way to introduce a precise performance variable is to pick a target for the hand or foot to reach. For example, to encourage precise movement in rotation, reach for something behind you with the hand. (You can also imagine “reaching” for targets with the nose, sternum, navel, or pubic bone.)
If the goal of the movement is primarily to create stability, as in a plank, bird-dog, or single-leg balance, you can form specific intentions that increase the challenge, such as: keeping the spine long as long as possible from the tailbone through crown of the head; imagining that you are balancing a wine glass on a part of you that is not supposed to move; or ensuring that the pressure under the feet does not shift forward/back or side to side.
In all of these cases, a clear target encourages better aim. Another useful performance variable for any kind of movement - make sure it is done with as little energy as possible. This will increase the demand for optimal coordination.
Step 3: identify the elemental variables that make up the synergy
As you are doing the movement, notice what joints are moving and which muscles are working. This identifies the elemental variables in the synergy and the members of the “team” that is cooperating to create the movement. It’s a way to learn how the team coordinates and to promote better coordination.
For example, in a standing rotation, where you are reaching to a target behind your body, you can notice that movements are happening throughout the body: in the ankles, hips, every level of the spine, the ribs, the shoulder blade, and the humerus in the shoulder joint. You may feel that these movements are well-distributed and proportional throughout the body. Or you may feel that some segments (perhaps some thoracic vertebrae) are not moving at all, and that others feel like they are moving too much. The simple act of noticing inactive segments will tend to get them moving, which will spontaneously change the organization of the whole synergy. It’s like the choir conductor noticing that one of the singers is not singing, or a soccer coach noticing that one of the players is not on the field.
You can do something similar in attending to the elemental variables forming a muscle synergy. In other words, what muscles feel like they are working hard and which feel lazy? This question might be especially relevant in exercises that are focused more on static stability than movement (eg a plank or bridge.) As with joint movement, you may notice that the sense of muscular effort feels either well-distributed over a large area, or concentrated in one area. For example, in a supine bridge position you might feel that the hamstrings are very active and the glutes or not. Just notice that and form an intention to distribute the work smoothly over as many muscles as possible.
Step 4: introduce variations that change the way sharing occurs in the synergy
This simply means doing a movement that is similar, but in some way slightly different than the first one. Think “repetition without repetition” or “same but different.” Each variation is like an experiment that provides information that helps to make the synergies become more robust and efficient. There are a million ways to vary a movement, and of course some will be more informative than others. But don’t be afraid of making mistakes and be creative and curious in exploring alternatives to see what information they provide. Here are some examples.
For the squat, you could vary the height of the stool, its distance from your heels, or the starting position of your feet. Any of these changes will cause your nervous system to spontaneously rearrange the synergy.
For each variation, use your attention as in step two to notice how the elemental variables of the synergy change. What moves more, and what moves less? Where does the sense of effort shift? You can use this attention to consciously learn about how your body works, or to encourage unconscious processes to focus on the sensory information that it needed to form coordinated synergies.
For example, if you put the stool farther behind you, this will require a squat pattern that is more hip-dominant. If the stool is right underneath you and close to the heels, you need more knee and ankle flexion and an upright torso. You can also investigate the effects of placing the stool to the left or right of center, or using staggered foot positions.
In the rotational movement, you could change the position of the target to require movement on a slightly different vector, either higher or lower. You could introduce constraints that prevent movement in certain segments of the body. If you do the movement in sitting, you eliminate contributions from the lower body, so the upper body needs to move more. Other interesting constraints might be interlocking the hands and placing them behind the head, or lying on your side with the lumbar spine flexed. (For some similar ideas, see my post Differentiation and Integration.)
In a plank, you could play with different support positions for the hands or elbows - further and closer from the feet, or from each other, or in staggered positions. Notice how these changes affect the feeling of muscular action through the trunk. You could remove half or all the weight from one elbow or foot, and see whether the spine position remains stable and long. You could introduce slight pulses of movement that oscillate the body forward/back or side to side, and see how stable the spine can remain in response to perturbations.
With each variation or new constraint, you are challenging the elemental variables to form different kinds of synergies providing you with a larger and more robust movement vocabulary.
Rinse and repeat with different fundamental movement patterns. Go slow and be curious and creative about the process. Over time, you will surely learn a lot about movement and improve your movement as well.
If anyone is interested, I am considering posting regular videos showing 5-10 minute explorations of different movement patterns based on the above template. If you would like to see some content along those lines, let me know in the comments.
Related posts
References and resources
Bizzi E, Cheung VC. The neural origin of muscle synergies. Front Comput Neurosci. 2013 Apr 29;7:51. doi: 10.3389/fncom.2013.00051. PMID: 23641212; PMCID: PMC3638124.
Giszter SF, Hart CB. Motor primitives and synergies in the spinal cord and after injury--the current state of play. Ann N Y Acad Sci. 2013 Mar;1279:114-26. doi: 10.1111/nyas.12065. PMID: 23531009; PMCID: PMC3660223.
Hogan N, Sternad D. Dynamic primitives of motor behavior. Biol Cybern. 2012 Dec;106(11-12):727-39. doi: 10.1007/s00422-012-0527-1. Epub 2012 Nov 3. PMID: 23124919; PMCID: PMC3735361.
Latash ML. Motor synergies and the equilibrium-point hypothesis. Motor Control. 2010 Jul;14(3):294-322. doi: 10.1123/mcj.14.3.294. PMID: 20702893; PMCID: PMC2921643.
Latash ML. Stages in learning motor synergies: a view based on the equilibrium-point hypothesis. Hum Mov Sci. 2010 Oct;29(5):642-54. doi: 10.1016/j.humov.2009.11.002. Epub 2010 Jan 8. PMID: 20060610; PMCID: PMC2891849.
Latash ML. One more time about motor (and non-motor) synergies. Exp Brain Res. 2021 Oct;239(10):2951-2967. doi: 10.1007/s00221-021-06188-4. Epub 2021 Aug 12. PMID: 34383080.
Profeta, Vitor L.S., and Michael T. Turvey. Bernstein’s Levels of Movement Construction: A Contemporary Perspective. Human Movement Science 57 (February 2018): 111–33.
Torres-Oviedo, Macpherson, and Ting. Muscle Synergy Organization Is Robust Across a Variety of Postural Perturbations, , 01 Sep 2006.
Yes, I’m interested in seeing your videos that illustrate the concepts introduced in today’s article.
Todd, this is a tour de force article. Thank you for writing it and condensing complex material into such a clear and elegant package.