Part 2 The Revival of Movement Studies: Esther Thelen and the Dynamic Systems Approach to Child Development

Esther Thelen (2000) International Journal of Behavioural Development 24: pp. 385-397

Esther Thelen was an expert in the field of developmental psychology. She was President of the Society for Research in Child Development and the International Society for Infant Studies. She was a Fellow of the American Association for the Advancement of Science and American Psychological Society: –

“Movement itself is a form of perception because the proprioceptive and haptic senses are continuously receiving information that is perfectly coupled with information from the external senses such as vision and hearing. Thus movement is an integral part of the ensemble of all our experience”. “After more than two decades of extraordinary theoretical and empirical contributions to our understanding of development, the study of motor systems had declined by 1950 and then lay dormant for nearly 30 years afterwards”. “Piaget’s theory of mental development was entirely grounded in perception-action, later interpreters were more interested in the contents of mind than its sensory-motor origins.”

Linda B. Smith Department of Psychology, Indiana University, Bloomington, IN, USA

Dr. Linda B Smith is Distinguished Professor and Chancellor’s Professor of Psychological and Brain Sciences at the University of Indiana:-

Traditional theories of intelligence and its development concentrated on symbolic reasoning, paying little attention to the body and to the ways intelligence affects and is affected by the physical world. Esther Thelen (1941–2004) was a maverick who argued against that traditional view for the idea that intelligence is both made in and realized through physical actions on the world. This once singular position is now known as the embodiment hypothesis and has become a major organizing theme in contemporary cognitive science, neuroscience, and development (see, for example, Smith and Gasser 2005).

Thelen, a professor of psychology, first at the University of Missouri and then at Indiana University, was never a traditionalist. She received her BS degree in zoology from the University of Wisconsin and her PhD in biological sciences from the University of Missouri, bringing the methods of whole-organism biology to human infants. Her dissertation focused on patterned movements—movements that repeat themselves over and over in a certain rhythmical way.  Her idea, the seed of her later developed dynamics systems theory, was that shifts in the variability of behaviors marked developmental transitions.

Initially, behavior is highly variable and disorganized; as it becomes organized, it is often highly controlled and inappropriately perseverative; truly skilled action is both stable and adaptively flexible.  Building on these ideas, Thelen founded a program of research on motor development that grew to become a major influence in developmental psychology, cognitive science, and physical therapy.  As a consequence of her field-changing discoveries and theories, she received many of the highest honors the field has to offer, held many national leadership posts, and was continuously funded by the National Institutes of Mental Health since 1987.


Thelen, (see Thelen and Smith 1994), envisioned cognition as embedded in, distributed across, and inseparable from the processes of perception and action. Cognition is situated within the same continuous, time-based, and nonlinear processes as those involved in bodily movement, and in the large-scale processes in the nervous system. Action is the Source of Developmental Change. Thelen (1994) asked: How can a learner who does not know what there is to learn manage to learn anyway? This is a more difficult question than might first appear. The issue is whether one needs to pre-specify the learning tasks and the learning goals, whether the baby has to know what needs to be learned in order to learn. Thelen showed us the way out of this quandary by demonstrating how babies can discover both the tasks to be learned and the solution to those tasks through exploration.  Spontaneous movement creates both tasks and opportunities for learning.  Young mammals—including children—spend a lot of time in behavior with no apparent goal.  They move, they jiggle, they run around, they bounce things and throw them, and generally abuse them in ways that seem, to mature minds, to have no good use. Thelen showed us how this sort of behavior is essential to the developmental process. It is our reason for being—to physically interact in a physical world. According to Thelen, the processes that give rise to motor behavior are also the repository of knowledge and the driver of developmental change.

As phenomenon, they also provide the key to the nested dynamics of human development.  The processes that make movement happen over multiple time scales. Neural excitation, for example, happens in milliseconds. Reaction times are of the order of hundreds of milliseconds. People learn skills after hours, days, and months of practice. What we call “developmental change”—transitions from crawling to walking—occurs over weeks, months, and years.

John P. Spencer, Daniela Corbetta, Patricia Buchanan, Melissa Clearfield, Beverly Ulrich and Gregor Schoner: Moving Toward a Grand Theory of Development: In Memory of Esther Thelen.  Child Development, November/December 2006, Volume 77, Number 6, Pages 1521–1538

The authors were colleagues of Esther Thelen who wrote this paper in memory of her after her death at a high point in her career:-

Thelen discovered that actions that were repeated impacted subsequent behavior even if the link between past and present behaviors was not transparent (Thelen & Farish, 1977). This pushed her to focus on process, on how and why these behaviors emerged and changed.  She discovered that these simple, repetitive behaviors not only provide a window for researchers interested in studying motor control, but also pro- vide infants with opportunities to become active participants in their own learning. This theme clearly resonated with researchers studying child development.

From a Classical to a Systems View of Infant Reaching

Before Esther’s work, the development of infant reaching was thought to occur in two phases that all infants traversed in a similar fashion (see Bushnell, 1985, for review). In a first developmental phase beginning at about 3 or 4 months of age, infants’ reaches were characterized by very discontinuous, zigzagging trajectories. In a second phase appearing around 8 months of age, infants began to reach for toys following a more direct path. The classic account of these two developmental phases focused almost entirely on visual control of reaching. During the first phase called the visually guided reaching phase it was assumed that infants’ discontinuous trajectories reflected their continuous effort to monitor and visually control the hand trajectory.

In the second phase called the visually elicited reaching phase infants could look at the target, anticipate an appropriate hand trajectory, and move their hand to the target using a fairly straight path without visual monitoring.  Esther’s work on infant reaching challenged this view of development in two key ways. First, Esther objected to the heavy emphasis on visual control in earlier accounts. By her view, the development of reaching did not reflect changes in a single factor; rather, she saw infant reaching as emergent from the assembly of many components. To reach out and grab an object, infants need to be motivated. They need to be able to localize the object in three-dimensional (3-D) space. They need to understand whether the object is reachable, and they need to transduce the perceived 3-D space into their body space. They need to be able to plan ahead and anticipate how the trajectory will unfold. They need to be able to correct their movements online as their hand approaches the toy. They need to be able to lift and stabilize the arm as they reach while maintaining the stability of the head and the trunk. And, they need to remember what works in context and distinguish this from what does not work.

Although vision is certainly involved in some of these challenges, vision alone cannot account for how infants learn to reach given these many interacting factors. The challenge, then, is to understand how infants manage to assemble all these factors to perform a successful reach.

In addition to showing that infants follow their own unique trajectories, this study of infant reaching revealed that exploration and selection is a key agent of developmental change. To improve over time, each infant experienced and explored a wide range of movements, ultimately leading to the discovery and selection of an optimal way to assemble the many components involved in a reach.

We want to highlight three points about these data. First, goal-oriented reaches were always embedded within a movement context that showed very similar kinematic characteristics. For instance, changes in the velocity of infants’ reaches over the first year paralleled changes in the velocity of infants’ non reaching movements Thus, when infants were more active as reflected in periods of higher average speeds in non reaching movements, reaches were also performed with higher movement speeds.  Likewise, when infants were less active as reflected in periods of lower average speeds, their reaches were also slower.

Second, the graphs illustrate the concepts of exploration and selection. Exploration is reflected in an early phase from reach onset to about 30 – 36 weeks, where all the speed and trajectory parameters fluctuate up and down and show unstable and changing curves. In this phase, infants explored a wide range of movement parameters and movement solutions. They tried fast and slow movements, learning the effect of these varied speeds on their ability to acquire the toy (see Thelen et al., 1996). This exploration generated crucial sensory-motor experience needed to learn to calibrate movements and feel the boundaries of control within the reaching task. This resulted in selection during a second phase from 30 to 36 weeks through the end of the first year. In this phase, movement parameters settled near particular values and showed much more stability over time as infants discovered, for instance, an optimal reaching speed that led to more stable and efficient reaches (Thelen & Corbetta, 1994; Thelen et al., 1996).

The third critical point is that the infants in the study converged on similar movement characteristics at the end of the first year despite their very different starting points at reach onset. For instance, they converged to a similar number of movement units, comparable movement straightness, and similar movement speed when contacting the toy. This reveals that different developmental pathways can lead to similar outcomes (Thelen et al., 1996).

Esther’s work demonstrated that reaches are carved out from the intrinsic dynamics of infants’ self-generated arm movements as they explore a range of movement possibilities and select viable solutions to meet the demands of the task. In this achievement, body and mind come together as infants assemble the many components that make a reach: the bio-mechanics of the body, the details of the specific environment including the perceived location of the toy, the speed and force needed to extend the arms away from the body, the ongoing movement and postural context, and so on.

This integration of body and mind is a fundamental characteristic of all goal- directed actions and creates a bridge to an embodied view of cognition and behavior.  Esther started to think deeply about the connections between her work in motor development and work in other domains of development. She turned naturally to cognition.

Many of the dominant questions in cognitive development stem from Piaget, who asked how children move from the sensorimotor origins of thought to abstract cognition.  Esther questioned the divide between ‘pure’ sensorimotor behavior and cognition. Indeed, in collaboration with Linda Smith, she denied the very existence of this divide. She argued that mental activity is embodied thought is always grounded in perception and action (e.g., Thelen, 2000; Thelen & Smith, 1994). This followed Piaget’s tradition in invoking the importance of the sensorimotor origins of thought. But rather than viewing development as movement toward the abstract and away from perception – action, Esther believed that for infants and adults alike cognition and action are not separate. Instead, cognition is inextricably linked to perception and movement. There is no cognition in the absence of perception and action.

In her Presidential Address to the International Society on Infant Studies in 1998, Esther outlined the importance of these findings: what infants know is always assembled, in the moment, with contributions from memory, attention, and action (Thelen, 2000). Cognition is embodied. Infants’ decision to reach is based on much more than whether or not they have an object concept. Instead, the decision to reach is based on what the infants have just done, their reaching skill, the feel of the body, the salience of the cue, and the perceptual layout of the task. Thus, this reaching task that was assumed to provide a direct window into infants’ abstract concepts is actually a window into the complex interactions among perception, action, and cognition in infancy.

Adam Sheya and Linda B Smith: Development through Sensorimotor Co-ordination: Enactment: Towards a new paradigm for cognitive science 2009 MIT press, Cambridge, MA:

“Piaget (1952) described a pattern of infant activity that he called a secondary circular reaction. A rattle would be placed in a four-month-old infant’s hands. As the infant moved the rattle, it would both come into sight and also make a noise, arousing and agitating the infant and causing more body motions, and thus causing the rattle to move into and out of sight and to make more noise. Infants at this age have very little organized control over hand and eye movement. They cannot yet reach for a rattle and if given one, they do not necessarily shake it. But if the infant accidentally moves it, and sees and hears the consequences, the infant will become captured by the activity—moving and shaking, looking and listening—and incrementally through this repeated action gaining intentional control over the shaking of the rattle. Piaget thought that this pattern of activity—an accidental action that leads to an interesting and arousing outcome and thus more activity and the re-experience of the outcome—to be foundational to development itself. Circular reactions are perception-action loops that create opportunities for learning. In the case of the rattle, the repeated activity teaches how to control one’s body, which actions bring held objects into view, and how sights, sounds and actions correspond.

Edelman (1987) also pointed to the coupling of heterogeneous sensorimotor systems in the creation of cognition. Edelman’s theory starts by recognizing the multimodal nature of the brain at birth; it is—from the start—a complex system made up of many heterogeneous, overlapping, interacting and densely connected subsystems. Like Piaget, Edelman pro-posed that development occurs through activity dependent processes.

Here, we review behavioral evidence from human development, evidence that suggests that transformative change is driven by the sensor-motor co-ordinations of an active agent in a physical world.

In a recent and remarkably inventive demonstration of this approach, Needham, Barrett, and Peterman (2002) fit two to five-month-old infants with Velcro®-covered ‘sticky mittens’. These mittens enabled the infants to grab objects merely by swiping at them, enabling them to precociously coordinate vision and reaching. Infants who were given two weeks of experiences with sticky mittens subsequently showed more sophisticated object exploration even with the mittens off. They looked at objects more and made more visually coordinated swipes at objects than did control infants who had no exploratory experiences with sticky mittens.

Two subsystems—reaching and looking—are coordinated in the sticky-mitten task and in so doing educate each other. But these components are also involved in other co–ordinations, that is, in other tasks that recruit other coalitions of subsystems. Thus, extra experience in the coordination of reaching and looking with sticky mittens ends up not being just about looking and reaching but potentially about other developments, other co-ordinations, generating cascading developmental consequences in other tasks in which some of the same subsystems are involved.

The second example is the development of reaching, Thelen et al.’s (1993) week-by-week study of four infants transition from not-reaching to reaching for visually presented objects. Early in development, the presentation of an enticing toy aroused the infants and elicited all sorts of non-productive actions. These actions were literally all over the place with no clear coherence in form or direction. But by acting, each baby sooner or later made contact with the toy—banging into or brushing against it or swiping it. These moments of contact selected some movements, carving out patterns that are then repeated with increasing frequency. Over weeks, the cycle repeated—arousal, action, and occasional contact. Over cycles, reaches became increasingly stable, more efficient and more effective.

However, the task of reaching is discovered by individual action, and thus, it is specific to the individual. All infants followed the general pattern, but each also had unique subtasks to solve. Some babies at first could hardly lift their arms, but sat placidly watching the world. Other babies were more high-strung and active, flailing and flapping and always moving.

These different babies had to solve very different problems in order to reach out and grasp an object. The flailer needed to become less active and to lower his hands bringing them into midline creating balance. The placid baby needed to be more active, to raise her hands, to lift them up from their usual positions on her side. What is remarkable in the developmental patterns of the children is that each found a solution by following individual action-defined developmental pathways that eventually converged to highly similar movements. Because action defines the task and because action—through the coordination of heterogeneous sensory systems— finds the solution, development is very much an individual and context-dependent matter.

An action in some context creates a task that coordinates multiple sensorimotor systems, and through this coordination, the component systems and their couplings to each other are changed. The next action may form a new consortium of systems, systems that will have been shaped by their participation in previous tasks. Because action creates tasks and transformative change in the components systems, action is a strong organizer of the developmental trajectory itself. Thus, motor development has a strong effect on the ordering of development as a whole.

The developmental course suggests gradual, action-driven discovery. When nine- to ten-month-old infants are given sets of objects containing like kinds, they do not group them. However, they do pick up objects, one in each hand, and bang them together (Forman 1982). By twelve months of age, these manipulations become more systematic and children manipulate like kinds in a like manner (Sugarman 1983). For example, given four cars and four dolls, the child may systematically push each car. Around eighteen months of age, children not only manipulate objects from one category in sequence, but they also systematically manipulate in different ways objects from two different categories, for example, first pushing each car, but patting each doll. This pattern of behavior—called “sequential touching” in the literature—is compelling to adult observers and seems to be, on the part of the child, a comment on the likeness of the individual instances. From these behaviors spatial classification emerges progressively.

Four behavioral tendencies in infancy may be enough to start the developmental progression. The first is that infants reach to objects in which they are interested. The second is that infants have a tendency to repeat just performed motor acts, and in particular to repeat reaches to nearby locations (e.g., Smith et al. 1999). Third, perceptually similar objects may be similarly enticing to infants. Fourth, infants may notice the outcomes of their own actions.”

Click here to view Part 3.

Published by


Walter Solomon is the father of Robert, the autistic boy who is the first subject of the book. He was educated at St Paul’s School, London and has a Masters degree in economics from Cambridge. He is a passionate advocate of the Waldon Approach, works with developmentally delayed children and adults in three different locations in Israel. He lives in Zur Hadassah a small town in a beautiful wine growing area in the Jerusalem hills.