Brief Introduction

‍‍‍‍‍‍‍Dynamic systems theory is a new and influential theory (Miller, 2002) developed in the 20th century by Esther Thelen. According to Miller (2002), dynamic systems theory is the broadest and most encompassing of all the developmental theories‍‍‍‍‍‍. ‍‍‍‍‍‍This theory attempts to encompass all the possible factors that may be in operation at any given developmental moment; it considers development from many levels (from molecular to cultural) and time scales (from milliseconds to years) (Miller, 2002). Development is viewed as constant, fluid, emergent or non-linear, and multidetermined (Spencer & Thelen, 2003). Dynamic systems theory’s greatest impact has been in early sensorimotor development (Thelen & Bates, 2003).‍‍‍‍‍‍‍

"Every act in every moment is the emergent product of context and history, and no component has causal priority" (Thelen, 2005, p. 271).
In other words...."Life is like giving a concert on the violin while learning to play the instrument." -Samuel Butler, 19th century novelist.
Development occurs "in the middle of things." Developmental outcomes are realized in large part by the real-time events that organisms take part in as they live their lives (Lickliter, 2005)

"A theory must be able to handle both the predictable aspects of development and those that surprise us" (Thelen, 2005, p. 260).
Dynamic systems theory accounts for both stability and variability of form and function.

‍‍‍‍‍‍‍Background and Development of Dynamic Systems Theory‍‍‍‍‍‍‍

Dynamic systems theory originated in the fields of physics and mathematics (Miller, 2002). Generally speaking, dynamic systems theory attempts to describe the flow of relationships among the components of some whole phenomenon (e.g., the solar system) (Thomas, 2001).

Dynamic systems theory (as it applies to developmental psychology) was developed by Esther Thelen, Ph.D. at Indiana University-Bloomington. Thelen became interested in developmental psychology through her interest and training in behavioral biology. She wondered if "fixed action patterns," or highly repeatable movements seen in birds and other animals, were also relevant to the control and development of human infants (Thelen & Bates, 2003

Thelen's Early Studies in Infant Behavior and Movement
Esther Thelen's early research in infant motor behavior (particularly stepping, kicking, and reaching) led her to become dissatisfied with existing theories and moved her toward a dynamic systems perspective. Prior views of development conceptualized infants as passive and infants’ motor development as the result of a genetically determined developmental plan. Thelen, in her work, discovered challenges to this view.


Infant Kicking and Stepping

Thelen realized that infants' stepping and kicking movements were remarkably similar. It was commonly known that 3 to 8 month old infants stop making stepping movements when held upright. However, infants continue to make kicking movements throughout this age period when supine, prone, or sitting. The traditional view was that infant stepping was inihibited as cortical centers autonomously matured. Thelen wondered why, then, would such a similar behavior (i.e., kicking) continue when infants were in a different bodily position? Additionally, when Thelen supported infants of this age on a small treadmill, they produced stepping movements and even adjusted their speed to the speed of the treadmill. Thelen concluded that the movement was not determined by the nervous system alone but emerged out of the confluence of circumstances. Movement wasn't "in" the infant in terms of a fixed motor program or set of reflexes, but was contingent on the environment. Also,Thelen explained that 3-8 month old infants stop engaging in stepping behaviors in part because they are developing more subcutaneous fat, making it difficult for them to lift their legs when supported upright, but easier when they are lying down (thus, the continuance of kicking movements). Over time, the development of subcutaneous fat is met with the development of muscle, and infants once again begin to make stepping movements. Some contexts suggest that components of actions are available before behavior actually appears, suggesting that the skill is awaiting many elements and that these elements change asynchronously (Spencer et al., 2006; Thelen, 2005).


See the clip below for an example of how an infant (who according to traditional stage theories of development would not be able to walk at this age) is able to produce walking movements when placed on a treadmill. New behaviors can come about when the proper environmental conditions are present.

‍‍‍‍‍‍‍Thelen and Piaget's A-Not B Task‍‍‍‍‍‍‍
‍‍‍‍‍‍‍Recall Piaget's "A-not B" task.‍‍‍‍‍‍‍ A toy is hidden in the A location several times before it is hidden in the B location. Infants tend to reach toward the A location even though they observed the toy being hidden in the B location. Piaget attributed this to the infant's incomplete schema (Click here for a video of Piaget's "A-not-B" task).
Thelen et al. (2001), however, found that when infants' body posture is changed just before a toy is hidden at the B location, 8-10 month old infants are less likely to make the error of reaching toward A. When infants are standing, they are less likely to make the A-not-B error compared to when they are sitting.

Based on these and other studies, Thelen concluded that infants' body weights and proportions, postures, elastic, and inertial properties of muscle and the nature of the task and environment contribute equally to the motor outcome. Infants can "self-assemble" new motor patterns in novel situations.

It should also be mentioned that because Dynamic systems theory such a new theory, the majority of studies thus far have focused on measuring movement because it is tangible and clearly measurable. For now, the researchers are working with more concrete aspects of development to prove the viability of the theory.

Core Concepts
Development is dynamic. The state of the system at any point depends on its prior states and is the starting point for future states (Thelen, 2005). Behavior emerges in the moment, but the effects of each behavioral decision accumulate over longer time scales, as each change sets the stage for future changes (Spencer et al., 2006).

Development occurs within a system, which is the result of (a) components, (b) patterns of relationships among components, (c) processes that arise from the interaction of components, and (d) outcome (Thomas, ‍2001). Any change in a system impacts other components of the system.

Behavior is both flexible and stable (Miller, 2002). Although behavior is flexible, organisms tend to return to their preferred state of relative stability, which is known as the attractor state (Miller, 2002; Spencer et al., 2006). When patterns self-organize, they settle into preferred states which can be stable or unstable. When patterns are stable, they resist change. Stable patterns are easily elicited and frequently performed.

Behavior is multiply determined and softly assembled. Behavior depends on multiple factors: the person's configuration of skills, his or her intentions or goals, and the nature of the particular task and setting (Miller, 2002; Spencer et al., 2006). Implication: We must examine multiple causal factors within the system and environmental context to understand behavior.

Patterns change when the system becomes unstable due to internal or external elements disrupting its coherence. (‍‍‍‍‍‍‍Examples of internal inputs are changes in a person's chemical balance, growth, and aging. An example of an external input is an infant's discovery that a coffee table facilitates pulling to stand.‍‍‍‍‍‍‍) Then the components reorganize into a new, more stable state. The old behaviors become less stable and less preferred (Thelen, 2005; Thelen & Bates, 2003).

These clips from the movie Apollo 13 demonstrates the concept of soft assembly. The characters are faced with a novel task with no one "right" answer. They must use a) what's in the environment and b) what they know based on past behaviors to softly assemble a solution.

The task:

The solution:


1. Development is nonlinear. Change is not a stage-like progression of new accomplishments but the waxing and waning of patterns (Thelen & Bates, 2003). Implication: Development must be studied over time to be understood.

This epigenetic landscape, depicted below, is an example of non-linear development. Theoretically, the marble could end up anywhere on this landscape, but only under certain conditions (i.e., certain paths are more likely than others).

2. Systems are self-organizing (Miller, 2002; Spencer et al., 2006). Self-organization means that organized patterns arise solely out of interactions of internal processes and contextual influences, with no one component having causal priority over another (Thelen, 2005). Children as systems are viewed as active problem-solvers who are sensitive to changes in their environment and in particular tasks (Thelen, 2005). Implication: The system itself is respnsible for change.

3. Embodiment: Perception, action, and cognition form an integrated system that cannot be partitioned (Spencer et al., 2006). Implication: Change in one aspect of a system impacts other aspects of the system.

4. Individuality: Development happens in individual children solving individual problems in their own unique ways (Spencer et al., 2006). Because each child is different in terms of his or her body, nervous system, and daily experience, the course of development is nearly impossible to predict. There are multiple pathways to development (Thelen, 2005). Implication: There are no "rules" for development.

Watch this clip from the television cooking competition Chopped. What core concepts and assumptions of Dynamic Systems Theory do you see at play here?

View of Human Nature

Dynamic systems is organismic (Miller, 2002; Thelen, 2005), which means that it emphasizes the relations among the parts of living organisms. It embraces a "big-picture" view of human nature in which humans are active, not passive. Dynamic systems theory is not focused on innate tendencies. Rather, dynamic systems theory addresses changes over time in complex holistic systems, especially self-organizing ones. According to dynamic systems theory, we need to consider the interactions among all levels of the developing system, from the molecular to the cultural.

Development as Quantitative or Qualitative

Dynamic systems theory acknowledges both quantitative and qualitative change. According to this theory, gradual quantitative change eventually causes a system-wide change and the emergence of a new, qualitatively different behavior (Miller, 2002).

‍‍‍‍‍‍‍The locomotion of a horse provides such an example.‍‍‍‍‍‍‍ As a horse increases locomotion speed, it shifts spontaneously from a walk to a trot to a gallop. Patterns change only at critical values and do so suddenly without intermediate patterns. Movements are "softly assembled" patterns created and dissolved as tasks and environments change.

Development due to Nature or Nurture

"There is no single element that contains the prior instructions for the behavioral performance . . .the essence of the behavior lies in neither the organism nor the environment alone . . . (and) neither has logical priority in explaining behavior or its changes" (Thelen & Ulrich, 1991, p. 24)

Dynamic systems theory acknowledges the role of neural systems in development ‍‍‍‍‍‍‍but views neural maturation and innate capacities as‍‍‍‍‍‍‍ insufficient to explain the emergence of new behavior patterns and the flexibility of behavior (Spencer et al., 2006). Interestingly, Thelen did not appear to place much value on the nature versus nurture debate. ‍‍‍‍‍She argued that what matters is not whether behavior is hard-wired or whether it is learned; what matters is that patterns can be flexible and shift depending on the situation (Thelen, 2005). Development is not just the result of genetics or the environment, but rather the interweaving of events at a given moment (Thelen, 2005).


What does Develop?

  • Dynamic systems theory is concerned with where new behaviors come from (Miller, 2002). Child development is the process of assembling patterns of behavior to meet the demands of the task in the biological possibilities of the child at that time (Thelen, 2005).
  • Focus is on the process of change: stability and instability
    • Development doesn't consist of a child progressing toward a known conception of mature adulthood. Development does not "know" where it is going. Rather, it is the outcome of continually active, self-organizing processes.
    • A key metric of dynamic systems is their stability, or strength of "attractors" of their patterns. When patterns self-organize, they settle into preferred states, or stable states. For patterns to change, they must lose stability. Internal or external elements must disrupt the components so that they can reorganize into a new and more stable state ‍‍‍‍‍‍‍(see Mechanisms of Development to learn about the different components)‍‍‍‍‍‍‍ (Spencer & Shoner, 2003).
  • Content emphasized
    • In principle, one can examine any sort of content through this theory (Miller, 2002).
    • Most research thus far has been on infant motor behavior, e.g., walking, reaching, searching for objects (Miller, 2002).
    • Other research areas include: cognitive skills (e.g., word learning, object permanence), social development (e.g., play), emotion regulation, personality development, interpersonal communication, social deviance, as well as worldwide problems like poverty and violence (Miller, 2002).
  • Overt behavior, or covert thought and personality?
    • No inferences are made about unobservable mental structures that exist outside of the behavior-in-context. It's not useful to ask a child what s/he really "knows" because behavior that is stable over many tasks can become fragile in different environments.
    • Dynamic systems is focused on measuring overt behavior. It focuses on the sensorimotor origins of cognition (Spencer & Schoner, 2003). Rather than cognition being perceived as "something static sitting in one's head," it is a representational state that is a time-dependent state of the nervous system. Thus, the brain runs simulations of past events during cognitive task

Mechanisms of Development

  • What drives development?
    • Development is driven by interactions between internal processes and contextual influences (Thelen, 2005).
    • Small changes in an initial or early state may result in large changes later (Miller, 2002)
    • Small changes cause changes throughout the system (Miller, 2002)
    • Mechanisms of change include (Thelen & Bates, 2000):
      • Experience
      • External information
      • Biological Constraints
      • Brain Development
      • Embodied Cognition
      • Dynamical systems
      • Formalization
      • Mental representations
      • Computer simulations
      • ‍‍‍‍‍‍‍Note: Social interaction has not YET been studied as a mechanism of change, but most likely will be once the theory has gained more support in the physical realm (i.e., kicking, stepping, etc)
  • ‍‍‍‍Outline of developmental process across lifespan
    • Not yet developed or addressed by this theory.
    • Most research has focused on infant motor development, and the theory has not been applied or explained across the lifespan.
  • Demonstrate 2 connections between early development and later development
    • Not yet developed or addressed by this theory, other than the idea that an infant's prior experience with certain task is one of many factors that contribute to the soft assembly of behavior (Thelen & Bates, 2003).
    • The applications to cognitive, emotional, and social development are in early infancy.‍‍‍‍

‍‍‍‍‍‍‍Methodology for testing presences of various stages‍‍‍‍‍‍‍

Methods must examine moment-to-moment changes over time. The best methods, according to our text (Miller, 2002), are:
  1. Microgenetic methods, which aim to capture cognitive processes over time
  2. Longitudinal designs
  3. Computer-assisted motion analysis (to examine complex patterns of motion). See image below from a study on infant reaching.

While most theories focus on collecting sparse data over wide ranges of time, Dynamic Systems Theory is concerned with collecting vast amounts of data over a very small amount of time. In this way, it is similar to the approach that qualitative researchers take (i.e., very detailed), but it is still highly quantitative because they are recording numeric measurements.

What is the ‍‍‍‍‍‍‍current state of the theory‍‍‍‍‍‍‍ in the field of lifespan development?

Dynamic systems as it relates to human development is a young theory, but shows promise. Unfortunately, it is predominantly utilized to explain infant behavior, and it has not yet been applied across the lifespan. It seems as though the emphasis is on establishing the legitimacy of the theory through the study of infant motor behavior which is easily measurable.

Conceptually, the theory has been applied to several areas, including therapeutic change, identity development,rehabilitation and occupational and physical therapy.

Because the theory is so new, we are unable to critique its "original" and "current" forms, as it has not yet undergone any revisions.

  • Dynamic systems theory has close ties with experimental studies
  • Dynamic systems theory accounts for both stability and flexibility across time
  • Behavior is examined within contexts


  • ‍‍‍Dynamic systems theory might overemphasize the individualism of development and may not adequately address similarities.
  • More research needs to be conducted to examine its utility at other developmental stages in life
  • There appear to be some contradictions: For example, Thelen argued that there are general developmental principles (Spencer et al., 2006) but also that it is almost impossible to predict the course of development (Thelen, 2005).


How Does This Theory Address and Apply to Issues of Diversity?

Because dynamic systems is such a new theory, ‍‍‍‍‍‍‍its‍‍‍‍‍‍‍ applicability to diversity issues has not been addressed. Nonetheless, the theory proves somewhat promising in this area. Because dynamic systems focuses on so many different contextual factors, there is less room for narrowly focusing on only nature or only nurture factors. In addition, the idea that there is no "right" way to develop (i.e., no inner vision of how development should occur or what the end result may be) the theory is resistant to pathologizing those who do not meet the status quo of society. At the same time, however, although this theory purports to address development from many levels (from molecular to ‍‍‍‍‍‍‍cultural), culture's role in development is not overtly addressed.‍‍‍‍‍‍‍

Implications of Dynamic Systems Theory

Dynamic systems theory leads us toward a very complex view of seemingly simple movements and the acknowledgment of situational and contextual factors that contribute to the emergence of behaviors in the moment.

Dynamic systems theory leads us away from parsimonious explanations of behavior, and away from the assumptions that behavior is a) innate and b) linear or stage-like.

‍‍‍‍‍‍‍Mountain Stream: A Useful Metaphor for Understanding Dynamic Systems Theory‍‍‍‍‍‍‍


"I suggest another metaphor for human behavior: a mountain stream. This is an apt comparison to keep in mind, because a stream is moving all the time in continuous flow and continuous change. Development is continuous—whatever has happened in the past influences what happens in the future. But the stream also has patterns. We can see whirlpools, eddies, and waterfalls, places where the water is moving rapidly and places where it is still. Like the stream, development also has recognizable patterns: milestones and plateaus and ages and stages at which behavior is quite predictable. In the mountain stream, there are no programs or instructions constructing those patterns. There is just water and the streambed under it. The patterns arise from the water and natural parts of the stream and the environment, such as the streambed, the rocks, the flow of the water, the current temperature and wind. The patterns reflect not just the immediate conditions of the stream, however; they also reflect the history of the whole system, including the snowfall on the mountain last winter, the conditions on the mountain last summer, and indeed the entire geological history of the region, which determined the incline of the stream and its path through the mountain. In addition, the stream also carves the rocks and the soil and creates its own environment, which then constrains and directs the water. It is not possible to say what directly causes what, because the whole system is so mutually embedded and interdependent.
Development also has these system properties. How a child behaves depends not only on the immediate current situation but also on his or her continuous short- and longer-term history of acting, the social situation, and the biological constraints he or she was born with. Every action has within it the traces of previous behavior. The child’s behavior, in turn, sculpts his or her environment, creating new opportunities and constraints (Thelen, 2005, p. 259-260)."

‍‍‍‍‍‍‍Resources! Articles You May Be Interested In‍‍‍‍‍‍‍

1 Dynamic systems theory and the complexity of change

2 U-shaped changes in behavior: A dynamic systems perspective

3 Moving toward a grand theory of development: In memory of Esther Thelen


Miller, P. (2002). Theories of Developmental Psychology (4th ed.). New York, NY: Worth Publishers.

Spencer, J. P., Clearfield, M., Corbetta, D., Ulrich, B., Buchanan, P., & Schöner, G. (2006). Moving toward a grand theory of development: In memory of Esther Thelen. Child Development, 77, 1521-1538.

Thelen, E. (2005). Dynamic systems theory and the complexity of change. Psychoanalytic Dialogues, 15, 255-283.

Thelen, E., & Bates, E. (2003). Connectionism and dynamic systems: Are they really different? Developmental Science, 6, 378-391.

Thomas, R. M. (2001). Connectionism and dynamic systems. In Recent Theories of Human Development. Thousand Oaks, CA: Sage.