Whether it’s rolling out of bed in the morning, or rolling a double play, movement is something most of us take for granted. But, in my opinion, improving our understanding of how movement is generated is absolutely vital for improving our abilities as coaches (or athletes). This article is by no means the definitive article on pitching mechanics — it is simply a small step towards improving motor control literacy in the pitching community.
A Brief History of Motor Control: How Does Movement Come to Be?
Traditional theorists in the study of motor control adopted a “top-down” model for the coordination and control of human movement. This type of hierarchical regulation of movement implies that there is some kind of central commander residing in the brain, organizing, initiating, and executing the appropriate actions for a given task; a puppeteer, if you will, pulling the strings of human movement.
To better understand the this top-down approach to motor control it is helpful to think movement systems like business organizations: the CEO (the motor cortex) directs the behaviour of his employees (muscle activity) who, ultimately, produce the desired output of the company (movement).
In 1968, motor behaviourist Steven Keele dubbed this abstract cortical representation of movement the “motor program”, drawing on the analogy of the human brain operating much like a sophisticated computer: commands are prepared in advance, stored, and run when they are needed, just like a computer program.
The existence of motor programs would suggest that movement is “pre-programmed” by the central nervous system, without the need for external guidance. As such, movement was thought to rely heavily on two things: (1) information processing and (2) memory. While on the surface this may seem like an elegant solution, exploring the idea of pre-programmed movement brings up some significant limitations.
The Case Against Pre-Programmed Movement
One problem that quickly arises with this “top down” approach to motor control is that of storage capacity. The brain is an incredible organ, but even it only has so much space. And when you think of all of the different kinds of movements (and permutations of those movements) you’re required to do in everyday life — let alone in a sport such as baseball — the number of motor programs needed to successfully navigate and interact with our environment start to add up very, very quickly.
Consider, for a moment, all of the different plays a shortstop is required to make over the course of a single baseball game, or, better yet, over the course of a season. Even plays involving a single “category” of movement skills (i.e. fielding, throwing, etc.) would require different movement strategies under different conditions, and thus would each require different motor programs.
For instance, fielding a routine ground ball and throwing the runner out at first requires a much different fielding and throwing pattern than getting that same runner out on slow roller that barely squeaks by the pitcher. Likewise, rolling an easy double play in practice is a lot different than when you’ve got a runner taking a hard slide into second.
In all four cases outlined above, the shortstop’s “job” is to receive the ball (either from the ground or from his second baseman) and make an accurate throw to first — but each of the motor programs that would be needed to get the job done are all slightly unique. Now multiply that over every possible scenario that has occurred and will occur over that shortstop’s career, and, well, that’s a heckuva lot of motor programs! The brain only has so much room.
Furthermore, early motor program theories fail to account for how novel (i.e. new and different) movement comes to be. For instance, we frequently see athletes (in all sports) successfully make plays that they have probably never been exposed to:
You think Bartolo Colon has a motor program stored away for that play?
Having a motor program stored away for every movement we’ve never performed, but may someday need, presents an even bigger storage problem for the brain. Moreover, it would suggest that all movements are somehow genetically wired in, ignoring the fact that many movements are learned rather than innate.
This is essentially the major argument against early motor program theories:
If there is (quite literally) an infinite number of ways to get the job done, how can the brain can possibly account for all of that information?
To account for these issues, Richard Schmidt developed the idea of a “generalized motor program” in 1975. Unlike their predecessors, these generalized motor programs were proposed to represent general movement patterns (e.g., running, throwing, etc.) that could be modified slightly both prior to movement (via information from the environment) and during movement (via reflexes and/or sensory feedback) to meet a variety of environmental demands.
Variations on Schmidt’s Generalized Motor Program (GMP) theory still pervade to this day, and, admittedly, there are a lot of things I like about it. But, there are also some important limitations to this theory, many of which are outside the scope of this article.
One important (and relevant) limitation of GMP theory, is that it does not adequately explain how athletes are able to successfully adapt to rapid and unexpected changes in their environment. This ability is a hallmark of a great athlete. Just take a look at Ozzie Smith throwing Mike Scioscia out at first while jumping over Lenny Harris barreling into second base:
There are hundreds of examples from the sporting world of elite athletes who achieve stable performance outcomes (e.g., Ozzie nailing Scioscia at first) despite considerable external perturbations (e.g., Harris’ take-out slide). Sorry, but pre-programmed movement is not the stuff highlight reels are made of!
Furthermore, Schmidt’s GMP theory fails to account for the dynamic interactions between subcomponents (e.g., muscle, nervous, and connective tissue) within the movement system (i.e. the athlete) itself during complex movement tasks involving multiple biomechanical degrees of freedom (e.g., joints).
Let me explain.
Because skilled motor performance (such as that which is displayed by pitchers at the highest levels of baseball) is often characterized by low variability in outcomes measures (e.g., exceptional command, a “consistent” release point) it is assumed by advocates of GMP theory that skilled motor performance must be the result of highly consistent (so called “invariant”) movement patterns (e.g., a “repeatable” delivery).
And, this is an assumption that predominates in sport — baseball included — such as when we talk about a pitcher “repeating” his delivery. Rarely, is this assumption questioned. That is, do pitchers actually repeat their delivery?
The Repeatability Fallacy and “Repetition Without Repetition”
Although there’s no direct evidence in skilled overhead throwing athletes, studies of skilled marksmen provide some indication that the skills requiring a high degree of consistency (like throwing strikes) may not be as repeatable as we think they are.
Arutyunyan et al. (1968, 1969) found that during a pistol aiming task, skilled marksmen were — to no surprise — very good at aiming the pistol. What was remarkable is how they were able to do it: The skilled marksmen didn’t recreate the same shoulder, elbow, and wrist positions from trial-to-trial. Instead, they exhibited a considerable amount of variation in the position of their joints between trials. But, rather than lead to inconsistent outcomes, these variations actually compensated for each other, and allowed the marksmen to achieve a consistent and stable aiming position — a phenomenon Russian neurophysiologist, Nikolai Bernstein, dubbed, “repetition without repetition”.
This is an important distinction:
No two movements — even those movements that have been practiced tens of thousands of times — will likely ever be identical, no matter how identical they may look at a macroscopic level.
Whether this idea of “repetition without repetition” applies to the high-level throwing patterns of baseball’s elite remains to be seen, but it certainly calls into question the notion of a generalized motor program, calling the shots from above.
Which brings up an interesting question:
If movement isn’t programmed, what’s the alternative?
Is Movement “Self-Organized”?
As an alternative to programmed movement, self-organization refers to the idea that complex, coordinated movement skills (e.g. throwing a baseball) emerge as a result of local coordination between smaller subcomponents (e.g. joints, muscles) of an initially disordered system (e.g. pitcher), rather than being prescribed by some sort of central command centre in the brain.
Let’s try and break this thing down.
An important feature of self-organized movement is that it is adaptive (i.e. it has a degree of built-in “flexibility”, so to speak). This “flexibility” allows the movement system (i.e. the pitcher) to reorganize its component parts in response to perturbations in the environment, in a manner that still meets the movement goals of the task (i.e. throw strikes).
On the other hand, a programmed movement is, by its very nature, “rigid”, and that rigidity compromises the stability of the movement system when it experiences any external perturbations.
This is a key distinction: Whereas advocates of programmed movement see movement variability as a source of error that needs to be reduced or eliminated for improved performance, those who believe that movement self-organizes feel that movement variability reflects the capacity of the sensorimotor system to adapt a movement pattern to specific environmental or task demands. (Remember those skilled marksmen?)
A simple (and potentially oversimplified) analogy for the flexibility that is inherent to self-organized movement is the suspension system on a bicycle. Suspension allows you to navigate a variable — and often unpredictable — terrain without damaging the chassis or falling off your bike. Generally, the more variable the terrain, the more suspension you probably need. For example, if you tried to take a road bike (a rigid system) up a mountain bike path (a variable environment) you probably wouldn’t get very far. The same can be said of human movement systems: a certain degree of variability is needed to cope with external perturbations.
Importantly, though, these perturbations aren’t always physical changes in the environment — they can arise from a variety of factors both intrinsic and extrinsic to the athlete. Instead of a change in terrain, maybe it’s the onset of fatigue? Or a rapid shift of attention caused by a coach yelling at us from the bench?
Collectively these factors, popularly referred to as “constraints”, guide the process of self-organization in the human movement system.
The “Control” of Self-Organized Movement
In 1986, Karl Newell developed his Model of Constraints, in which he proposed that movement patterns emerge from the interactions between the internal and external constraints acting on the movement system.
Importantly, as Kugler et al. (1980) pointed out constraints do not cause movement, but rather exclude movements. Speaking broadly, constraints are boundaries that impose limitations on the number of possible configurations that the involved degrees of freedom at various levels of the movement system can form.
Newell classified these constraints into three distinct categories:
Organismic constraints are those that reside within the individual athlete. They can be further divided into structural and functional constraints:
Structural constraints are the physical and morphological characteristics of the athlete such as their height, weight, genetic make-up, torso and limb length, muscle cross-sectional area, connective tissue strength, joint range of motion, and so on. These constraints are relatively stable over time, although they certainly can (and do) change (e.g., athletes often gain strength and muscle cross-sectional area in the off-season).
Functional constraints, on the other hand, tend to vary considerably. They are the physiological and psychological characteristics of the athlete at any given time: things such as heart rate, hormonal levels, anxiety, motivation, and so on.
An example of organismic constraints in pitching:
Let’s say we ask two different pitchers to throw at a target low and away. One of them is 6’10”, and the other 5’11”. By nature of their individual anthropometrics (i.e. differences in torso and limb length, etc.), there is going to be a huge difference in the way in which they organize their body to get to a release point that is conducive to hitting the target. (Most coaches know this intuitively.)
The same two athletes could also exhibit very different psychological profiles — an example of a functional constraint. Maybe the guy who’s 6’10” doesn’t work well under pressure and as a result is dealing with a ton of anxiety while we watch him try to complete this task. Then, we could walk off the field, and that anxiety goes away.
Environmental constraints refer to the the global and physical layout of the surrounding physical and sociocultural environment in which the athlete is competing. These include variables such as ambient light, temperature, and wind, but can also include peer pressure, family support, and cultural norms.
And example of environmental constraints in pitching:
Consider the difference between throwing on a perfectly-sloped mound compared to your average dirt mound with that huge landing hole the opposing pitcher dug out of it last inning. Moreover, take that perfect mound and place it indoors in a 70 degree facility, with poor lighting and no wind. Now compare that to that previously-used mound outdoors, in 90 degree heat, glare from the sun coming from behind home plate, with a 20 MPH back wind. How would pitching under these two different environmental circumstances affect the movement strategies that emerge from a pitcher trying to throw strikes?
Task constraints are related to the goal of the task and the rules rules, regulations, and instructions that govern the task. Newell (1986) proposed that task constraints “are not physical, rather they are implied constraints or requirements which must be met within some tolerance range”.
An example of task constraints in pitching:
If we stick with the act of throwing strikes, the task constraint in this example would be just that: “throw a strike” or “hit your target”. Alternatively, we can shift the task constraint and instead instruct our player to “throw the ball as hard as you can”. The task has now changed to task where the goal is velocity and therefore the way in which the body organizes a motor solution would be separate and distinct from a task where the goal is accuracy.
With a few exceptions, traditional approaches to coaching often involve manipulating very few of the many possible constraints we could manipulate as coaches or athletes.
One of the most common constraints we manipulate as coaches and athletes comes from that first category of constraints — the organismic constraints — and it is the intentions of the performer. The intentions of the performer (or what has been commonly referred to in the pitching world as “intent”) is perhaps, as Kelso (1995) pointed out, the most influential organismic constraint for shaping coordinated movement patterns.
And, to demonstrate the role of “intent” in governing the self-organization of movement, I’m going to (briefly) discuss why a lot of pitching coaches are likely manipulating the intentions of their pitchers in sub-optimal — even detrimental — ways.
And it all comes down to the language we use with our athletes.
Your Words Matter: The Science of Internal vs. External Cues
As coaches, our role is to help direct the athlete’s attention in an effort to execute a desired movement outcome. And the principle way we do this is through the use of verbal cues (i.e.talking to our athlete). The science behind verbal cueing and it’s relation to attention and performance has been largely pioneered by Dr. Gabriele Wulf.
Verbal cues can be classified into two broad categories:
- Externally-directed and …
To simplify, a cue that produces internally-directed attention is one that places emphasis on thinking about one’s own body parts. For example, if we instruct our athletes (or ourselves, for you pitchers out there) to “brace your leg on landing!”, that’s an internally-directed cue — it gets us thinking about what our leg is doing.
An externally-directed cue, on the other hand, directs an athlete’s attention to the effect of the movement. For example, instead of saying “brace your leg on landing”, we could say “land firm”.
The superiority of external cueing has been studied quite extensively in the field of motor performance, and in a variety of different tasks such as speed and force production (Marchant & David, 2015) and even throwing accuracy (Emmanuel et al., 2008).
The proposed mechanism behind the superiority of externally-directed attention was postulated by Wulf in her Constrained Action Hypothesis in which she states:
“Consciously focusing on the movements of a motor action disrupts automatic motor control processes that regulate coordinated movements. When athletes actively focus and consciously control their movements, they interrupt automatic nonconscious motor behavior processes that normally control movements in an efficient manner. In contrast, directing attention externally to the movement effects allows the motor control system to naturally regulate and organize motor actions. As a result, movements are unconscious, fast, and reflexive.” (Wulf et al. 2011).
Interestingly, the superiority of externally-directed attention fits in well with the concept of self-organized movement. That is, external cues seem to allow the system to self-organize, a finding corroborated by a plethora of literature on attention and coordination by Keith Lohse (2010, 2011). In an ever-fluctuating environment, we want the motor system to be as robust and adaptable as possible.
Alternatively, if we as coaches promote an internally-directed focus, we may be promoting rigidity, which can have adverse consequences for motor performance. Indeed, one of the major causes of “choking” in sports is self-focused attention.
This is particularly true when an athlete is in a high-stress situation, such as when your pitcher is trying to throw to strikes in a close ballgame. If he’s focusing on keeping his front hip closed, creating good separation between his hips and shoulder, and pronating his wrist through deceleration — or even just one of the many internally-directed cues we give our pitchers — he’s going to walk the freakin’ house.
Rob Gray, professor of Human Systems Engineering at Arizona State University, has studied this concept in simulated batting tasks, and found that when players were under pressure or anxiety, their attention focused inwards, making them more rigid and subsequently displayed a significant drop in batting performance (Gray, 2011).
The benefits of external cues over internal cues for improving motor learning and performance could be an article in and of itself. (Hmmm…) However, I’m simply trying to demonstrate how constraints — no matter how insignificant they may seem — can have a massive impact on the organization of the coordinative structures involved in goal-directed motor skills such as pitching.
But this is just one example (albeit an important one) of the many, many variables that can affect the self-organization of movement, both positively and negatively.
Working Towards Improved Player Development
Self-organization is an intriguing lens through which we can view movement, and has many important implications for how we go about coaching pitchers, and developing our athletes in general.
But, perhaps most importantly, in our opinion, is the significance of self-organization for injury prevention, a topic that the Baseball Performance Group is excited to be presenting next week through our free webinar.
About the Author: Tavis Bruce is the Director of Research and Education at the Baseball Performance Group. To learn more about self-organization and motor variability in baseball sign up for our free upcoming webinar on Dynamic Systems Theory here.