Action And Motor Cognition

Action and motor cognition is the study of how minds and brains produce movement—and how movement, in turn, shapes thought.

14 frameworks, from 1890 to today · Cognitive Science · Social Sciences

still a working view ended · each leads to its entry below

Overview

Action and motor cognition is the study of how minds and brains produce movement—and how movement, in turn, shapes thought1. It asks how a body with many joints and muscles, receiving noisy signals from its senses, manages to reach for a cup, walk a tightrope, or return a tennis serve; how movements are planned, learned, and corrected; and how we understand the actions of others21. The stakes are practical as well as theoretical. A theory of action must explain how a stroke patient relearns to grasp, why a child with cerebral palsy moves differently, and how a robot might act flexibly in an uncertain world. Historically, the field has been a meeting point for physiology, control engineering, psychology, and neuroscience, and its central debates—whether action is organized by internal programs or emerges from the body's interaction with its environment—remain live today.

Reflex Physiology and the Localization of Movement

The scientific study of movement began with the reflex. In the seventeenth century, René Descartes described the body as a machine in which sensory stimulation triggers movement through the nerves, an idea that made involuntary action the paradigm case. In the nineteenth century, the physiologist Charles Scott Sherrington gave the reflex a precise experimental basis. Working on decerebrate animals, he showed that a simple spinal reflex such as the knee-jerk is not an isolated event but part of an integrated system: muscles work in reciprocal pairs, one contracting while its antagonist relaxes, and reflexes are modulated by higher centers. Sherrington's 1906 The Integrative Action of the Nervous System argued that the nervous system coordinates movement through the interplay of excitation and inhibition across reflex arcs.

This work established a picture in which movement is built from the bottom up, out of chained reflexes. It was powerful for spinal and postural control, but it could not explain how an animal initiates a voluntary reach or how a pianist plays a novel sequence. The reflex model left the origin of purposive action unexplained, and that gap set the agenda for the next century.

Degrees of Freedom and the Problem of Coordination

In the 1930s, the Russian physiologist Nikolai Bernstein posed the problem in a form that still organizes the field. A human arm has many joints and muscles—many "degrees of freedom," meaning independent ways it can move—so the same hand trajectory can be produced by countless combinations of muscle activations23. How does the nervous system choose among them23? Bernstein's answer was that it does not choose each one separately3. Instead, it groups muscles and joints into synergies: functional units that are controlled together, reducing the number of variables the brain must manage2. He also observed that skilled movement is not a fixed sequence but a flexible solution to a task, adjusted to context—a blacksmith's hammer stroke varies with the metal, yet the goal is preserved.

Bernstein's ideas, published mainly in Russian and translated into English only in 1967 as The Co-ordination and Regulation of Movements, were slow to circulate but became foundational. They reframed the central question of motor control from "how is each muscle commanded?" to "how is coordination organized?"—a question that later dynamical and ecological approaches would take up directly.

Cybernetics, Servo-Control, and the Feedback Loop

During the Second World War, engineers confronted the problem of aiming guns and steering aircraft, and their solutions entered biology. Norbert Wiener's cybernetics, formalized in his 1948 Cybernetics, described control as a loop: a system compares its actual state with a desired state and uses the error to correct its output. The servomechanism—a device that continuously adjusts to reduce error—became a model for how organisms regulate movement.

The psychologist Kenneth Craik, in The Nature of Explanation (1943), proposed that the brain builds small-scale internal models of the world and uses them to predict the consequences of action, an idea that anticipated later forward models. Feedback control explained how movements are corrected mid-course, as when you adjust your hand as a cup slips2. But it had a famous limitation: feedback takes time, and fast movements—a boxer's punch, a typist's keystroke—are too quick for sensory correction to keep up24. Pure feedback could not explain how rapid, accurate action is possible24.

Efference Copy and the Cancellation of Self

A partial answer came from an older physiological idea. In the nineteenth century, Hermann von Helmholtz had noted that we cannot tickle ourselves, and he proposed that the brain distinguishes self-generated from external sensation by anticipating the sensory consequences of its own commands. In the twentieth century, this idea was developed under several names—efference copy, corollary discharge—most influentially by Erich von Holst and Horst Mittelstaedt in 1950, who studied how flies stabilize their gaze. When the fly's head is passively rotated, its eyes move to compensate; when the fly turns its own head, the eyes do not, because a copy of the motor command cancels the expected visual motion.

The principle is general: to act effectively, a system must predict the sensory effects of its own movements and subtract them, so that it responds to the world rather than to itself5. This idea would later become central to computational accounts of action, but for decades it remained a physiological curiosity.

Motor Programs and Schema Theory

The problem of fast movement pushed mid-century psychologists toward the opposite of reflex theory: the motor program. If feedback is too slow, the argument went, then rapid actions must be planned in advance and run off without continuous correction24. A program is a stored set of commands that, once initiated, produces a movement sequence24. The idea explained how skilled typists or pianists produce fluent sequences, but it raised a new puzzle: if every movement required its own program, the storage demands would be enormous, and novel movements would be impossible.

Richard Schmidt's schema theory, proposed in 1975, addressed this. Rather than storing a program for every specific movement, the system stores generalized motor programs—abstract patterns for a class of movements, such as throwing—along with schemas that specify how to parameterize them for a particular distance or speed. A schema is a rule relating the outcome of past movements to the commands that produced them, so that a novel throw can be generated by generalizing from experience. Schema theory was influential and generated testable predictions about practice and transfer, though later work questioned whether such abstract rules are really what the nervous system stores.

Population Coding in the Motor Cortex

Meanwhile, neurophysiology was opening a window onto how the brain represents movement. In the 1960s and 1970s, Edward Evarts and others recorded from single neurons in the motor cortex of monkeys performing trained arm movements, finding cells whose firing related to force, direction, and joint. A decisive step came with Apostolos Georgopoulos and colleagues in the 1980s, who recorded from populations of motor-cortex neurons while monkeys moved a handle in various directions. Individual neurons fired most for a preferred direction, but the population of activity pointed reliably toward the actual movement—a "population vector" that could be decoded to predict the direction of the reach6.

This showed that movement direction is represented not by single command neurons but by the distributed activity of many cells, each broadly tuned6. Population coding became a general framework for relating neural activity to behavior, and it fed directly into the development of brain-computer interfaces, in which decoded population activity drives a cursor or prosthetic limb76.

Forward Models and Predictive Control

The efference-copy idea and the engineering notion of internal models converged in the 1980s and 1990s into a computational framework. A forward model is an internal simulation that predicts the sensory consequences of a motor command; an inverse model computes the command needed to achieve a desired outcome28. Working in part on the cerebellum and on oculomotor control, researchers such as Mitsuo Kawato and Daniel Wolpert argued that the brain uses such models to solve the degrees-of-freedom problem, to compensate for delays, and to learn new skills. A concrete demonstration: when you lift an object whose weight you have learned, your grip force is set in advance based on a prediction; when the weight is secretly changed, the first lift is misjudged and only later trials correct it—evidence that the motor system predicts rather than merely reacts.

Forward models also explained sensory cancellation: the predicted consequences of self-movement are subtracted from actual sensation, which is why you cannot tickle yourself and why the world stays stable as your eyes move25. By the 2000s, internal-model theory had become a dominant framework, linking motor control to perception, learning, and the sense of agency—the feeling that one is the author of one's actions.

Dynamical Systems and Motor Development

A rival tradition treated movement not as computation but as the behavior of a dynamical system2. Drawing on Bernstein and on the mathematics of complex systems, Esther Thelen and Linda Smith, in the 1980s and 1990s, studied infant motor development—reaching, crawling, walking—and argued that new behaviors emerge from the interaction of many components: body proportions, muscle strength, motivation, and environment. In their famous "stepping reflex" studies, newborns who appear unable to walk when held upright will make stepping movements when their legs are submerged in water, which reduces their effective weight; the behavior was present all along, latent in the system, and emerged when a constraint changed.

On this view, there is no stored program that unfolds; development is self-organization. The approach, sometimes called dynamical systems theory of motor development, emphasized variability, the coupling of perception and action, and the body's physical dynamics as causes of behavior rather than mere constraints on it. It aligned naturally with ecological psychology, James Gibson's tradition holding that perception is directly of affordances—opportunities for action—rather than of neutral sensations to be interpreted2.

Embodied and Enactive Cognition

These threads fed a broader movement. From the 1990s, embodied cognition argued that cognitive processes are shaped by the body's form and sensorimotor capacities, and enactivism, associated with Francisco Varela, Evan Thompson, and Eleanor Rosch, held that cognition is constituted by the ongoing interaction of an organism with its environment rather than by internal representation alone910. For action and motor cognition, this meant taking seriously that movement is not the output of thought but part of thinking: the way a chess player's hand hovers over pieces, or a child's gestures while solving a problem, can be constitutive of the cognitive process rather than mere expression of it1. The claims are contested—critics ask how far embodiment can explain abstract thought—but the movement decisively changed what counts as a phenomenon worth studying.

Mirror Neurons and Simulation

In the early 1990s, Giacomo Rizzolatti and colleagues recording from macaque premotor cortex found neurons that fired both when the monkey performed a grasping action and when it observed another individual performing the same action. These mirror neurons suggested a direct link between action production and action observation11. The finding was extended to humans through neuroimaging, and it inspired simulation accounts of social cognition: we understand others' actions, and perhaps their intentions and emotions, by covertly simulating them with our own motor system.

The interpretation was controversial from the start11. Critics noted that mirror responses in monkeys were studied mainly for grasping, that the human evidence was largely correlational, and that simulation cannot by itself explain how we understand actions we cannot perform. The debate sharpened a broader question: is action understanding a matter of inferring hidden mental states, or of motor resonance1211? The "theory-theory" of action understanding—the view that we explain behavior by attributing beliefs and desires, using a tacit theory—had been the dominant alternative, and the mirror-neuron work forced a direct confrontation between the two.

Predictive Coding and the Bayesian Turn

The most recent synthesis treats action within predictive coding, a framework in which the brain constantly generates predictions about sensory input and updates its models by minimizing prediction error13. On this account, perception and action are two sides of one process: action can reduce prediction error by changing the world to match the prediction, just as perception reduces it by changing the model. The framework, developed by researchers including Karl Friston, unifies forward models, efference copy, and Bayesian inference under a single principle, and it has been applied to motor learning, the sense of agency, and even to psychiatric conditions in which the boundary between self-generated and external events is disturbed.

The Field Today

Contemporary action and motor cognition is plural. Internal-model and predictive-coding approaches dominate computational work, population coding and brain-computer interfaces continue to develop in neuroscience, and dynamical and embodied approaches keep the body and environment in view. The old tensions persist in new form: whether the brain is best seen as a controller running internal simulations or as one component of a body-environment system that organizes itself; how far motor representations explain social understanding; and how the sense of agency arises from the interplay of prediction and feedback. What unites the field is the conviction, hard-won over a century, that action is not a mere output of cognition but one of its central problems.

References

  1. Motor cognition, Wikipedia
  2. Motor control (introduction, Computational issues of motor control, Redundancy, Synergies, Closed loop control, Open loop control, Motor Programs, Forward models, Inverse models, Information based control, Behavioral dynamics, Direct perception), Wikipedia
  3. Degrees of freedom problem, Wikipedia
  4. Motor program, Wikipedia
  5. Efference copy, Wikipedia
  6. Neural coding (Population coding), Wikipedia
  7. Neurometric function, Wikipedia
  8. Internal model (motor control), Wikipedia
  9. Enactivism, Wikipedia
  10. Embodied cognition, Wikipedia
  11. Mirror neuron, Wikipedia
  12. Theory-theory, Wikipedia
  13. Predictive coding, Wikipedia

The frameworks of Action And Motor Cognition

Each framework is a way of working the field has used: when it held, and the question it answers.

  1. 1890-1935

    Reflex

    How does sensory stimulation produce movement?

    Fly to its tower in Nousfarer

  2. 1935-Present

    Degrees of freedom problem

    How does the nervous system coordinate many muscles and joints?

    Fly to its tower in Nousfarer

  3. 1943-1975

    Cybernetics and Servo-Control

    How does a system correct its output using error?

    Fly to its tower in Nousfarer

  4. 1950-Present
  5. 1960-Present

    Ecological Psychology

    How do agents perceive and act in their environment?

    Fly to its tower in Nousfarer

  6. 1960-1985

    Motor program

    How are rapid movement sequences produced without feedback?

    Fly to its tower in Nousfarer

  7. 1975-1995

    Schema Theory (Schmidt)

    How can novel movements be generated from past practice?

    Fly to its tower in Nousfarer

  8. 1980-Present

    Population Coding

    How do populations of neurons represent movement direction?

    Fly to its tower in Nousfarer

  9. 1980-Present
  10. 1985-Present

    Dynamical Systems Theory of Motor Development

    How do movement patterns emerge and change in development?

    Fly to its tower in Nousfarer

  11. 1985-Present

    Internal model

    How does the brain predict the consequences of its commands?

    Fly to its tower in Nousfarer

  12. 1991-Present

    Embodied cognition

    How does bodily action shape thought and perception?

    Fly to its tower in Nousfarer

  13. 1992-Present

    Mirror neuron

    How does observing an action engage the observer's motor system?

    Fly to its tower in Nousfarer

  14. 2005-Present

    Predictive coding

    How does the brain predict sensory input from its models?

    Fly to its tower in Nousfarer

How the ideas connect

Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.

ReflexDegrees of freedom problemCybernetics and Servo-ControlEfference copyEcological PsychologyMotor programSchema Theory (Schmidt)Population CodingTheory-Theory of Action UnderstandingDynamical Systems Theory of Motor Dev…Internal modelEmbodied cognitionMirror neuronPredictive coding

Beyond the field

Where Action And Motor Cognition's ideas came from in other fields, and what they led to there: the strongest 12 of 79 links.

A city in Nousfarer: a red-carpeted stair climbing terraces to a crowned tower on an acropolis, more towers round it, forest and snowy mountains beyond.
A city in Nousfarer, the game played in the browser.

Action And Motor Cognition is a planet in Nousfarer

Its city has a tower for each of these 14 frameworks. Land, walk in, and climb one concept at a time.

Land on Action And Motor Cognition

Related fields

Fields its ideas are linked to

More in Cognitive Science

See where Action And Motor Cognition sits on the Map of Human Knowledge