Printer Friendly

Discovery of motor development: a tribute to Esther Thelen.

In this tribute to Esther Thelen's legacy, it is discussed how she brought concepts of new theoretical perspectives into the domain of motor development. As a consequence of this a rejuvenalisation of motor development took place in the mid 1980's. The study of motor development skills became a testing ground not only for the task at hand, but also for other areas of development.

Key words: Thelen, infant motor development, stereotypy


"The (re) discovery of motor development" is part of a title used by Esther Thelen (1989) for a discussion paper in Developmental Psychology. It reflects why she became famous in the world of developmental psychology during the early 1980s. She put the field of infants' motor development in infants back on the scientific map as an important topic of study both in its own right and as a window into other, less accessible domains of development. Her achievements are relevant for any academic interested in teaching and research in the areas of movement co-ordination, movement control and skill acquisition. Esther Stillman Thelen passed away at the age of 63 on December 29th, 2004 at a hospital in Bloomington, Indiana.

Esther was trained as ethologist and started her career by carefully documenting repetitive movement stereotypy's, first in grooming wasps (masters degree, 1973) and, later in developing infants (doctoral degree, 1977). This early work resulted in a number of well-received publications and the beginning of what would be continuous funding by NSF and NIH. In a comparatively short career, spanning only 25 years, Esther's impact on the field is reflected in three books, an SRCD monograph, and over 120 scientific articles and book chapters. She served on the editorial boards of 15 scientific journals, gave countless invited lectures and was elected president of the International Society on Infant Studies (1996-1998) and president of the Society for Research in Child Development. Esther's influence and significance for scientific work in development will be measurable beyond these outputs for many years to come.

In this tribute to her legacy, we will comment on some of Esther's impact and contribution to the field of developmental science. We will start with her early classic muscle-fat study and end with her A-not-B error experiments, and Dynamic field Theory. In the latter she worked towards an encompassing dynamic theory for the development of action, perception, and cognition.


The experiments conducted by Thelen in the early 1980s (Thelen & Fisher, 1982, Thelen et al., 1983; 1984) can be considered as the starting point of a major intellectual shift in motor development. In these experiments, she challenged the, then-prevailing view on motor development as being determined by neural maturation. Before discussing these experiments, we go back into history.

In the first two years an infant acquires different a range of goal-directed motor behaviors, such as reaching, grasping, sitting, crawling and walking. These so-called milestones are well known since the work of Gesell (1929), McGraw (1935) and Shirley (1933). The description of these milestones resulted in a view of motor development as a rather rigid and gradual unfolding of postures and movements, that was mainly attributed to the general process of maturation of the central nervous system. From a neural-maturation perspective the development of movement co-ordination is regarded as a gradual unfolding of predetermined patterns (from cephalo-to-caudal and central-to-distal sequences) in the central nervous system and an increasing cortical control over lower reflexes. This view was still dominant among scientist way into the 1960s and 1970s, the so called heydays of the information-processing approach. Esther challenged this perspective with her experiments. In one of the key papers, 'The relationship between physical growth and a newborn reflex' published in Infant Behavior and Development, she and her co-workers illustrate instead that the complex interplay between infants' bodies, their environment, and earlier experiences determine the course of developmental changes. They found that in 8-week-old infants held upright, the stepping movements observed at a younger age disappeared. However, when the infants were lying supine, they performed kicking movements that were kinematically similar to the earlier observed stepping movements. Moreover, when the same infants were held upright in water the stepping pattern also reemerged. If the disappearance of the stepping reflex is due to cortical inhibition, as the traditional explanation would have it, why would the cortex inhibit movements in the upright posture but not in the supine posture or in the water? Thelen explained this disappearance of newborn stepping movements as a consequence of the disproportionate growth of leg muscles and fat tissues (Thelen et al., 1984). Specifically during this period of development infants acquire fat at a greater rate than muscle mass, which leads to relatively less muscle force. Thus, the occurrence of stepping movements (task) is a consequence of the interaction between organismic constraints (body proportions) and environmental constraints (orientation to the gravity vector), and not uniquely determined by neuro-maturational constraints. No single factor has priority, a theme that would resurface throughout her work.

In the earlier eighties, ecological psychology, coordinative structure theory and dynamic systems theory were introduced (Gibson, 1987; Kelso, 1995; Kugler, Kelso, & Turvey, 1982). The research described in the article of Thelen et al. (1984) can be considered as one of the first experimental studies to test the theoretical concepts provided by these new avenues of theoretical thinking. Twenty years later, studies of motor development have become a major testing ground for examining the developmental implications of these new theoretical perspectives (some examples from our Amsterdam group: Savelsbergh & Van der Kamp, 1994, Kawai, Savelsbergh & Wimmers, 1999; Van Hof, Van der Kamp & Savelsbergh, 2002). For example, the Van Hof et al. (2002) experiment examined how the development of crossing the midline is interwoven with the development of bimanual reaching. Previously, it was held that the development of midline crossing is uniquely determined by maturation of the hemispheric specialisation (Provine & Westerman, 1979) or the maturation of spinal tracts (Morange & Bloch, 1996). Van Hof et al. observed infants longitudinally at 12, 18 and 26 weeks of age while reaching for two balls (3 and 8 cm in diameter) at three positions (ipsilateral, midline and contralateral). With age, the infants increasingly adapted the number of hands used to the size of the object. The number of reaches crossing the body midline increased with age. Furthermore, the majority of the midline crossings were part of two-handed reaches for the large ball and occurred at, or after onset of bimanual reaching. Together, this strongly suggest that the development of crossing the body midline emerges in the context of bimanual reaching. It is concluded that the need to grasp a large ball positioned contralaterally with two hands induces midline crossing. Hence, the development of midline crossings is not exclusively dependent on organismic constraints (e.g., the maturation of hemispheric connections), but on their interaction with environmental constraints (e.g., object size). Similar to Thelen's work, no single factor had priority.

The experiments conducted by Thelen in the early 1980s can be considered as the starting point of a major intellectual shift in motor development. Although their primary intention in conducting the research in the original paper was to provide empirical evidence to provide an alternative explanation for the results of training on infant stepping reported by Zelazo (1983). The impact of the articles however, is also (and perhaps mainly) due to the change in theoretical thinking about motor control and particularly motor development. Concepts stemming from new perspectives had begun to achieve popularity in the scientific community, but empirical data supportive of these new ideas was lacking, especially in motor development. Thelen et al.'s article provided some of the earliest support for these new emerging theoretical perspectives and became one of the major catalysts responsible for the rejuvenation of the study of motor development.


Inspired by the work of Bernstein (1967), Gibson's (1979, 1988) and Turvey (1990) and dynamic systems theory by Kelso (Kelso 1995; Kugler, Kelso, & Turvey, 1982; Kugler & Turvey, 1987), Esther started to use motor development as a vehicle to provide empirical evidence for (i) developmental processes are nonlinear, (ii) action and perception form an inseparable loop and (iii) variability in development is functional.

For instance, Ester was inspired by Nikolai Bernstein (1967), a Russian physiologist, who formulated one of the central issues in understanding the development of motor co-ordination: the 'degrees of freedom' problem. The degrees of freedom problem refers to the possible movements of all the components (e.g., muscles, tendons, joints etc.) of the motor apparatus of the human body. Bernstein realised that the non-linear nature of the interactions among these different components of the human body makes their separate regulation impossible and inferred that to be able to control all these components, or degrees of freedom, these movements have to be coordinated. Co-ordination, therefore, is the process of mastering the redundant degrees of freedom into a controllable system. It is this idea of non-linearity what comes back in many papers of Esther (e.g. Thelen et al, 1989; Thelen & Smith, 1994).

A couple that had a considerable influence on Esther's thinking is Eleanor and James Gibson's (1979, 1988). The Gibson's ecological psychology approach to perception is also known as the direct perception perspective. The word 'direct' refers to the fact that objects, places and events in the environment can be perceived without the need of cognitive mediation to make perception meaningful, such as in the information-processing approach. Information in the environment is not static in time and space, but specifies events, places and objects. The child has to learn to pick up and select the appropriate information by exploration. The concept of exploration plays a key role in many publications by Esther. She advocates that through active and directed exploration of the environment the child learns to detect relevant information, and to couple the information to movements. For instance, this is nicely shown by Thelen in her earlier ethologically orientated studies (1979) where she studied the occurrence of rhythmical behaviors like kicking, rocking, arm waving and banging in natural settings. She showed that these behaviors occurred most frequently around transitions where new behaviors emerged. Rocking on hands and feet, for example, emerged in development just before the onset of crawling. This rocking might represent the infant's exploration of his action capabilities and the (proprioceptive) information it produces. This eventually leads to a new stable action pattern (crawling). Through exploration the infant discovers how perception and action are coupled.

A third source of inspiration, and maybe her major source, is the dynamic systems theory (Scott Kelso, 1995). The aim of the dynamic systems approach is to characterize spatio-temporal and functional patterns of motor behavior in terms of their stability properties by formalizing the time-evolution of relevant variables into dynamical equations of motion. Stationary, stable states or patterns of activity, as well as abrupt transitions between different states accompanied with loss of stability (induced by changes in external conditions), have been successfully modeled in this way (Kelso, 1995). The perspective portrays co-ordination as a process that constrains the potentially free variables of a system into a behavioral unit. A collective variable (order parameter) is the parameter that captures the observed behavior (coordination pattern), while a control parameter is the parameter that leads the system through different co-ordination patterns. Within this approach, the behavioral pattern is regarded as a stable collective state attained by the system under certain constraints (boundary conditions) and informational settings (Zanone, Kelso, & Jeka, 1993). When the control parameter passes through a critical point, a co-ordination pattern that was stable becomes unstable causing a sudden discrete transition to a qualitatively different, stable co-ordination pattern. Such a change appears without any prescription from outside, but is acquired by the system self, i.e., through self-organization. Esther Thelen's group provided empirical evidence (among others, e.g. Van Geert, 1999; Van der Maas, 1993) that developmental processes are not smooth and monotonic, but are better characterized by phenomenon such as discontinuities, transitions, instabilities, and regressions (Thelen, 1995; Thelen & Smith, 1994, 1995; Thelen & Ulrich, 1991). These phenomena are characteristic of non-linear dynamical processes and scientists from this perspective often search for control and order parameters. The identification of rate-limiting factors can help to identify these control parameters (like relative muscle force as result of the ratio of fat/muscle tissue by Thelen et al., 1984).

Within this theoretical perspective, Esther's carried out a string of ingenious cross-sectional and longitudinal experiments. The development of infants' leg movements as a research paradigm dominated in the 1980s, with careful tracking of infants' kicking, stepping, and walking patterns. In the 1990s, she extended her research program from cyclic leg movements towards perceptually guided, goal-directed arm movements in reaching and grasping (for instance; Thelen et al., 1993). An important concept in these studies was variability. When the data are grouped with respect to the transition point an increase in variability is found around the transition point. Thelen and Smith (1994) suggested that variability around a transition is a prerequisite in order to jump to a qualitatively new level of behavior. In their view, variability is highly functional: it provides the infant/child with possibilities for exploration, that is, the opportunity to discover new qualitatively different solutions.


Esther's most recent work was directed towards embodied cognition, showing how the complex interactions between looking, remembering, and acting

shape infants' perseverative errors in the A not B search task. With this work Esther challenged one of the most highly researched phenomena in the study of infant cognitive development: Piaget's theory of object permanence with this work. In the traditional work (Piaget, 1954), Piaget described how his 9.5 months old child, Laurent, made reaches for his hidden watch. The watch was hidden several times under a garment at location A and Laurent searched for it. Subsequently, the watch was hidden at a nearer and highly similar location B while Laurent watched. The infant made a curious reaching 'error'; after a delay between hiding and reaching, Laurent searched for the watch at location A, the location where it was first found. Piaget concluded that it is a universal phenomenon that is due to limitations in Laurent's concept of an object. The explanation focuses attention on Laurent's internal cognitive machinery, the way he thinks about the world.

Following the work of Piaget the hide and seek task has been repeated countless times, with myriad variations. The error turned out was found to be a robust phenomenon in its standard form, but highly complex in its context dependency. Each of the highlighted components seems to be crucial to the occurrence of the error (see for a review Smith, Thelen, Titzer, McLin, & Smith, 1999). After decades of research no cognitive theory could explain the full pattern of experimental results, neither a consensus on the nature of the error nor on its meaning was attained. If the A not B error is a measure of the infant's object concept, how can it be that infants have a more mature object representation when seemingly small task conditions are altered?

Thelen and Smith challenged the notion that the error reflects the infant's immature understanding of the concept of an object. They focused on the infant's reaching activity and not on the contents of the infant's mind (Smith and Thelen, 1993; Thelen and Smith, 1994). A critical experiment was performed to demonstrate, for example, that the same patterns of behavior are evident when waving a continually in-view object at location A or B, instead of hiding a toy (Smith et al., 1999). So, the error is not about what infants have and don't have as enduring object concepts but what they are doing and have done. From a dynamic perspective on goal-directed reaching, perseveration in the A not B task is a continuous, self-organizing process which involves multiple interactions including visual input, looking behavior, posture, and the memory dynamics of repeating the same action several times in succession. This means that no single element has causal priority.

In collaboration with Gregor Schoner, a powerful mathematical model of A-not-B (i.e. the dynamical field theory) was offered and tested (Thelen, Schoner, Scheier, & Smith, 2001). The field model simulates the decision of infants to reach to location A or B and the level of inclination to reach in any direction is constantly changing over time. This constantly changing set of inclinations (i.e. the movement planning field) depends on a number of factors, including (1) the current state of the field, (2) the task input, capturing the perceptual cues that specify the possible hiding locations, (3) the selective input, capturing the hiding event or the waving of the lid, (4) the memory of previous reaches become an input to the next trial, so in absence of other cues the system is likely to stick to previous choices (5) interactions between the locations make sure that a highly activated location will exert a strong inhibitory influence over the other point, so one location wins out.

This model can reproduce the A not B error. Before there is any object hidden there is activation at both A and B. As the experimenter directs attention to A by for example hiding the toy, it produces high activation at A. When the activation peak crosses a threshold the infant reaches to A. Each time the infant reaches to A, the memory becomes an input and the activation level at A builds. After several A trials the experimenter draws attention to location B. But as that cue decays, the lingering memories of location A begin to dominate the field and over time the inclination shifts back to location A. Experimenters can make the error come and go by manipulating the inputs to the fields (Smith and Thelen, 2003). For example the A not B error in 10 month old infants diminishes when the number of prior reaches to A decreases, when the two hiding locations are made more distinctive and, when the hiding event at B is more attractive (i.e. hiding cookies instead of toys). Younger infants are not able to maintain stable peaks. In the A not B task they are dominant on B trials through their motor memory of reaches to A. Older infants maintain a stable peak after the cue at B so they reach to B despite delay and motor memory. Clearly the development of keeping the relevant, but no longer visual, cues in memory is a critical developmental achievement. But also developmental changes in the perceptual differentiation of the targets and attention processes may contribute to a shift from perseveration to non-perseveration. The consequences are that the division between what is conceptual and what is perceptual-motor may be very hard to draw.

The dynamic field theory predicts that the dynamics that create the error in infants are general processes involved in goal-directed reaching at all ages. Indeed, even toddlers make perseverative errors in the A not B task when the A and B locations are not marked by lids but instead are two locations in a homogeneous task space such as a sandbox (Spencer, Smith, and Thelen, 2001). The model captures both the robustness of the task and its exquisite context-sensitivity. This is in itself a great achievement, but the model also provided a number of novel predictions, which were born out in further experiments (for instance: Schutte, Spencer & Schoner, 2003; Spencer & Schoner, 2003). Esther's principle of multicausality demands a rethinking of what is meant by knowledge and development. The model accomplishes infant behavior in the A not B task without invoking constructs of object representation or other knowledge structures. Rather the infants' behavior of knowing or not-knowing is emergent from the dynamics of the reaching task itself, including the history of previous actions. With her work on embodied cognition, Thelen showed that knowing cannot be separated from perceiving, acting, and remembering.

In this tribute we provide a taste of Esther's Thelen impact and contribution to the field of developmental science. Her works towards an encompassing dynamic theory for the development of action, perception, and cognition will be a source of inspiration for several research groups over the world. She left us, undoubtedly, with a vibrant, healthy state of current research on motor development. Esther's influence and contributions to scientific work and academic life will be measurable beyond these outputs for many years to come.


Bernstein, N.A. (1967). The co-ordination and regulation of movements. Oxford: Pergamon Press.

Gesell, A. (1929). Maturation and infant behavior pattern. Psychological Review, 36, 307-319

Gibson, J.J. (1979). The ecological approach to visual perception. Boston: Houghton Mifflin

Gibson, E.J. (1988). Exploratory behavior in the development of perceiving, acting, and the acquiring of knowledge. In, M.R. Rosenzweig & L.W. Porter (Eds.), Annual review of psychology (pp. 1-41). Palo Alto, California: Annual review, Inc.

Kawai, M., Savelsbergh, G. J. P., & Wimmers, R. H. (1999). Newborns spontaneous arm movements are influenced by the environment. Early Human Development, 54, 15-27.

Kelso, J.A.S. (1995). Dynamic Patterns. The self-organization of Brain and Behavior. Cambridge: MIT press.

Kugler, P.N., & Turvey, M.T. (1987). Information, natural law, and the self-assembly of rhythmic movements. Hillsdale, NJ: Lawrence Erlbaum.

Kugler, P.N., Kelso, J.A.S., & Turvey, M.T. (1982). On the control and coordination of naturally developing systems. In J.A.S. Kelso & J.E. Clark (Eds.), The development of movement control and coordination (pp. 5-78). New York: John Wiley and Sons.

McGraw, M. (1935). Growth: A study of Johnny and Jimmy. New york: Appleton-Century-Crofts.

Morange, F. & Bloch, H. (1996). Laterlization of the approach movement and the prehension movement in infants from 4 to 7 months. Early Development and Parenting, 5, 81-92 Piaget, J. (1954). The Construction of Reality in the Child. New York: basic Books.

Provine, R. R. & Westerman, J. A. (1979). Crossing the midline: Limits of early eye-hand behavior. Child Development, 50, 437-441.

Savelsbergh, G. J. P., & van der Kamp, J. (1994). The effect of body orientation to gravity on early infant reaching. Journal of Experimental Child Psychology, 58, 510-528.

Schutte, A.R., Spencer, J.P., & Schoner, G. (2003). Testing the dynamic field theory: Working memory for locations becomes more spatially precise over development. Child Development, 74, 1393-1417.

Shirley, M.M. (1931). The first two years: A study of twenty-five babies. Vol. 1. Postural and locomotor development. Minneapolis: University of Minnesota Press.

Smith, L. B., Thelen, E., Titzer, R., & McLin, D. (1999). Knowing in the Context of Acting: The task Dynamics of the A not B error. Psychological Review, 106(2), 235-260.

Smith L. B., & Thelen, E. (2003). Development as a dynamic system. Trends in Cognitive Sciences, 7(8), 343-348.

Spencer, J.P. & Schoner, G. (2003). Bridging the representational gap in the dynamic systems approach to development. Developmental Science, 6, 392-412.

Spencer, J. P., Smith, L. B., & Thelen, E. (2001) Tests of a dynamic systems account of the A not B error: the influence of prior experience on the spatial memory abilities of 2-years-old. Child Development, 72, 1327-1346.

Thelen, E. (1979). Rhythmical stereotypies in normal human infants. Animal Behaviour, 27, 699-715.

Thelen, E. (1989). The (RE) discovery of motor development: Learning new things from on old field. Developmental Psychology, 25, 946-949.

Thelen, E. (1995). Motor development: A new synthesis. American Psychologist, 50, 79-95

Thelen, E., & Fisher, D.M.. (1982). Newborn stepping: An explanation for disappearing reflex. Developmental Psychology, 15, 447-453

Thelen, E., & Smith. L. (1995). A dynamic systems approach to development of cognition and action . Cambridge: MIT Press.

Thelen E, & Smith LB. (1994). A Dynamic systems approach to the development of cognition and action. Cambridge, MA:MIT Press

Thelen , E., Fisher, D., & Ridley Johson, R. (1984) The relationship between physical growth and a newbone reflex. Infant Behavior & Development, 7, 479-493.

Thelen, E., Ridley Johson, R. & Fisher, D., (1983). Shifting patterns of bilateral coordination and lateral dominance in the leg movements of young children. Developmental Psychobiology, 16, 29-46.

Thelen E., Schoner, G., Scheier, C., Smith, L. B. (2001). The dynamics of embodiment: a field theory of infant perseverative reaching. Behavioral Brain Sciences, 24, 1-86.

Thelen, E., Corbetta, D., Kamm, K., Spencer, J.P., Schneider, K., & Zernicke, R.F. (1993). The transition to reaching: mapping intention and intrinsic dynamics. Child Development, 64, 1058-1098.

Thelen, E. & Ulrich, B (1991). Hidden Skills: A dynamic systems analysis of treadmill stepping during the first year. Monographs of the Society for Research in Child Development, 56, (1, Serial No. 223)

Turvey, M.T. (1990). Coordination. American Psychologist, 45, 938-953.

Van Geert, P.C.L. (1999). Dynamic systems of development. Change between complexity and chaos. New York: Harvest Wheatsheaf

Van der Maas, H.L.J. (1993). Catastrophe analysis of stepwise cognitive development. Academic thesis, University of Amsterdam.

Van Hof, P, van der Kamp, J., & Savelsbergh, G.J.P. (2002) The Relation of Unimanual and Bimanual Reaching to Crossing the Midline. Child Development, 73, 1353-1362.

Wimmers, R.H., Savelsbergh, G.J.P., Kamp, van der, J., & Hartelman, P. (1998). A cusp catastrophe model as a model for transition in the development of prehension. Developmental Psychobiology , 32, 2335

Zanone, P.G., Kelso, J.A.S., & Jeka, J.J. (1993). Concepts and methods for a dynamical approach to behavioral coordination and change. In G.J.P Savelsbergh (Ed.), The development of coordination in infancy (pp. 89-136). Amsterdam: North-Holland.

Zelazo, P. R. (1983). The development of walling: New findings and old assumptions. Journal of Motor Behavior, 15,99-t37.

Author Information:

G.J.P Savelsbergh, Institute for Fundamental and Clinical Human Movement Sciences, Faculty of Human Movement Sciences, Free University of Amsterdam, the Netherlands, Institute for Biophysical and Clinical Research into Human Movement, Department for Exercise & Sport Sciences, Manchester Metropolitan University, United Kingdom.

Human Movement Sciences, Vrij University

Van der Boechorststraat 9, 1081 BT

Amsterdam Netherlands


Fax: +31-20-598



COPYRIGHT 2005 Behavior Analyst Online
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Savelsbergh, G.J.P.
Publication:The Behavior Analyst Today
Date:Sep 22, 2005
Previous Article:Constructing single-subject reversal design graphs using microsoft word[TM]: a comprehensive tutorial.
Next Article:Comprehensive ABA programs: integrating and evaluating the implementation of varied instructional approaches.

Related Articles
Minds on the Move.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters