Variability in Postural Control During Infancy: Implications for Development, Assessment, and Intervention

Abstract

Variability is commonly considered a key to typical motor development. However, multiple definitions and quantification systems have limited the clinical interpretation of variability and the translation of developmental research to assessment and intervention. The purposes of this perspective article are to highlight the importance of statistical variability and complexity in postural control during development and to describe implications for assessment and intervention during infancy and early childhood. Five tenets are proposed describing the role of variability in postural control to support movement experiences, exploration, and global development. Evidence for assessment and intervention focused on variability in postural control are introduced.

The importance of variability in infant development has been evident in the research literature for years.^1–3 Whether using a cognitive perspective,³ a perceptual motor perspective,^1,4 or a motor perspective,^5–7 variability has emerged as a driving factor of developmental change. In this perspective article, we focus on variability of postural control and the description and evolution of that control in the first months of life as the child becomes an active participant in the world. Because all interaction with the environment includes some aspect of postural control, children with deficits in postural control are limited in their experiences from the very start of life. Atypical variability may be a key component in identifying postural control problems at an early age and planning successful interventions.

The purposes of this perspective article are to highlight the importance of variability in postural control during development and to describe implications for assessment and intervention during infancy and early childhood. In this exposition of early postural variability, we propose and provide supporting evidence for 5 key tenets:

1. Variability is a part of normal development and postural control.

2. Variability in postural control is needed for functional skill acquisition.

3. Variable and adaptive postural control facilitates exploration through movement opportunities, perception, and action.

4. A lack of complexity in postural control may be an early marker of developmental disabilities.

5. Enhancing the complexity of variability may lead to functional changes and improvement in motor function.

Variability Is a Part of Normal Motor Development and Postural Control

Variability is a hallmark of “normal” motor development across the life span and has been described and interpreted in multiple ways.^1,4,5,7–13 From the fetus moving in utero to the adult walking, variability can be observed in most human movements and across various time scales.^1,4 On a short time scale such as a single event lasting a few seconds or minutes, variability can be observed within the movement. For example, a spontaneous arm, leg, and head movement of a fetus or young infant can be characterized as irregular and variable in amplitude and velocity.^5,8,9 On a longer time scale, variability can be observed during a task-oriented sequence, such as moving across the room. For example, belly-crawling infants use a variety of movement patterns (reciprocal, symmetric, arms only, legs only), even within the same belly-crawling sequence, to accomplish the goal of moving across a space.¹⁰ Variability also can be observed over the long term as a skill emerges over months. The development of reaching¹¹ and gait¹² are frequently characterized by early variability in biomechanical measures, which decreases as skill proficiency increases after weeks of practice. Variability is so pervasive in development that it has been considered to be a catalyst for changing skill over time, rather than simply a correlate or result of change.¹³

In addition to the variety of time scales over which variability can be measured, different definitions and interpretations of the term “variability” have made it challenging to explain changes in variability occurring during development or in physical therapist practice. Is greater variability always better? Should a more skilled and experienced child have less variability? Should variability be measured the same way for different types of tasks?² In this article, we will specifically focus on describing variability of postural control in early life, from the newborn period until skilled, independent sitting is achieved. We will consider postural control from the perspective of a spontaneous behavior (versus a skilled task), prospective or in advance of movement (rather than reactive), and as a behavior that can be measured on various time scales. Table 1 provides definitions for many of the terms used in this perspective article.

Table 1.

Definitions of Key Terminology Used for This Perspective Article

Postural control can be measured in several ways, but one of the most frequently used methods to evaluate prospective postural control during movement is to quantify the displacement of center of pressure (COP) over a relatively short time scale (seconds or minutes). The COP is the location of the vertical reaction vector on the surface on which the individual is positioned and is a weighted average of the forces acting on the surface.²³ Movement of the COP is the body's neuromuscular response to the position of the body's center of mass.²³ Movement of the COP over time reflects postural adjustment and creates a time series that can be analyzed for cyclical patterns.¹⁸

For example, an infant ring sitting on the floor while looking between his mother and a distant toy is continually exhibiting postural adjustments. When he moves his gaze from his mother to the toy, his COP is displaced. When his gaze returns to his mother, the starting position of his COP is dictated by the position of the COP while he looked at the toy. However, the path of his COP as he returns his gaze to his mother may differ from the path of the COP when he looked from his mother to the toy. The use of various pathways during multiple repetitions of this task can be measured as high complexity, which requires constant postural adjustment. Although smaller in magnitude than in a sitting infant, COP displacement can be measured in supine infants during spontaneous movements, reaching, kicking, and changes in trunk or head position.^24,25 These postural adjustments can be seen as a purposeful process by which an individual explores and utilizes the limits of his or her dynamic stability during spontaneous or goal-directed movements.^26,27

The variability of the COP displacement as described above can be investigated in terms of both magnitude (statistical variability) and organization (complexity). Statistical variability is evaluated using traditional linear measures of centrality such as standard deviations. Complexity can be quantified using nonlinear analysis that determines the evolution of behavior over time.²⁸ In Figure 1, we use a schematic to plot the change in posture (variable unidentified) against time. This plot uses the mean and standard deviation to describe differences in the 2 tracings. The top graph indicates that the behavior is very regular and repeatable except for one large spike. The bottom graph indicates that the behavior is less regular and not very repeatable, with a more random-looking pattern. However, both graphs have the exact same mean and standard deviation, despite having a very different appearance.

Figure 1.

Schematic of time series of changes in posture (variable undefined): (A) postural movement that is repetitive, with the exception of a single large postural sway; (B) postural movement that is complex or nonrepetitive. Note that the means and standard deviations of both plots are equal, demonstrating a limitation of these measures of variability to describe the internal structure of the postural control within the time series.

In translating this comparison to the clinic, consider the postural control of an infant with poverty of movement (too regular, few strategies) plus one large startle or sway compared with a similar infant who makes small, dynamic changes in posture (not too regular and with a variety of strategies) related to an environmental goal. The standard deviation (indicating an average of the statistical variability) of these 2 infants' COP displacement would be the same because the standard deviation does not take into account the time-dependent evolution of the behavior or the pattern within the time series. However, the differences in the postural control strategies used by the infants in this example may affect their ability to select a behavior that allows interaction with objects while sitting upright, as we will highlight in later sections of this perspective article.

Various methods are now available to describe the internal structure of the variability we see within behaviors. Nonlinear tools are able to quantify the variations in pattern and the evolution of movement and behaviors over time that are continuous, such as postural control (see Harbourne and Stergiou¹⁶ for a review of tools). These methods build on narrative descriptions of variability in postural control by quantifying qualities of postural control, such as complexity. Complexity can be described by the regularity of the pattern of variability and by the number of strategies used over time (Tab. 1). In combination, linear and nonlinear analysis quantify postural control to provide a more complete understanding of the adaptive strategies used in postural control than either method could provide alone.

For example, an infant who is ring sitting and has an anterior posterior postural sway trajectory with large statistical variability (larger standard deviation) and a small amount of complexity might be rocking forward and backward in a repetitive fashion. In comparison, a similar infant who is ring sitting and has an anterior posterior sway with a smaller amount of statistical variability (small standard deviation) and larger amount of complexity, making smaller and less repetitive changes in COP, might be moving his COP to adapt to small changes in the postural control required to stay sitting upright while looking around the room. A supine infant with large statistical variability and a small amount of complexity in caudal-cephalic COP displacement might be alternating between a position of trunk flexion and extension in a repetitive fashion. In contrast, a supine infant with small statistical variability and larger complexity in caudal-cephalic COP movement might maintain a position of trunk flexion while looking around the room and exhibiting small adaptive weight shifts to prevent rolling to the side and support reciprocal kicking. These small and less repetitive changes in the COP reflect the adaptation of posture necessary to actively select and pick up sensory information from the surrounding environment, which can contribute to planning the next action. This example demonstrated the necessity of complex variability in developing postural control to enhance the development of multiple action systems.

We urge researchers and clinicians to consider behavioral variability, statistical variability, and complexity to fully describe an infant's control of posture. The next 4 sections of this perspective article will focus on how we believe this combined approach to considering postural control will lead to advances in early detection and better intervention for infants and young children.

Variability in Postural Control Is Needed for Functional Skill Acquisition

Although postural control is important to the development of functional skills, it is not postural control alone that supports function. We propose that highly complex variability of postural control drives change and serves as a foundation for function in the first year of life.^4,19,20 From this early complexity, normal development of postural control can proceed to select successful strategies for functional skill.²⁹ Research findings presented in this section support the importance of early complexity for the development of action systems and demonstrate how complexity supports the infant's ability to select the best strategy for a task.

An important concept emerging from studies using nonlinear analysis to describe developing skills is that early complexity or a tendency toward randomness precedes eventual adaptive skill.^{24,28,30–33} Although behaviors may appear random in early development, there is structure within that variability that is highly complex.^24,30,33 This complexity is a building block for eventual skilled behavior to emerge and enable a system to adapt to changing task demands, environmental conditions, or individual constraints (limitations).³³ Figure 2 illustrates our perspective on how behavioral variability (photos), statistical variability, and complexity can be considered in analyzing infant posture in supine and sitting positions. The first row of pictures in the figure depicts infant supine postures, and the second row depicts sitting postures. The rows below each picture describe a category of variability (statistical variability or complexity), as well as the use of adaptive postural control strategies to explore the world.

Figure 2.

Statistical variability, complexity, and adaptability for supine and sitting motor behaviors. Each image represents an infant's attempt to control posture during a specific behavior. Each behavior is characterized by the statistical variability and complexity observed during this type of motor behavior. In addition, each behavior is categorized based on the authors' perspective of the infant's ability to adapt postural control strategies to task demands.

Center-of-pressure data have been used to quantify early postural control in supine and sitting positions, allowing examination of the spontaneous and unconstrained movements of infants in the first year of life.^24,30 Researchers examined the continuous, dynamic, and adaptive process of developing postural control using a combination of linear and nonlinear analysis techniques. Dusing and colleagues²⁴ examined the COP movement variability in infants while positioned supine in the first 1 to 3 weeks of life during unrestricted and self-initiated spontaneous movements. Full-term infants who were healthy were able to maintain their COP in a small area, with minimal postural sway (low statistical variability).²⁴ However, the structure of the COP movement variability within this area was irregular (high complexity), as measured by approximate entropy.²⁴ We propose that the complexity of postural control observed during the first weeks of life supports the developing action systems of the infant, including orienting, manipulation, and appetitive functions. Multiple postural control strategies support the infant's attempts to move in variable ways while exploring control of his or her own movements.

Harbourne and Stergiou³⁰ used COP data to examine the development of early postural control in sitting. The data collection paradigm allowed the infants to sit as independently as possible while attending to a toy or person and collecting time series data on the COP. With monthly data collections from 4 to 8 months of age, the researchers described significant changes that took place in the way posture was controlled. The standard linear measures of the statistical variability of COP, such as the range and the standard deviation, showed that the majority of the infants (about 60%) demonstrated an increase in statistical variability in the anterior-posterior direction. The other 40% of the infants who were developing typically decreased the statistical variability of their anterior-posterior COP movement. However, the nonlinear variables of approximate entropy (quantifying regularity) and Lyapunov exponent (quantifying strategy selection; Tab. 1) indicated early irregularity and multiple strategies at the beginning of sitting, which consistently decreased over time regardless of the change in statistical variability. This finding suggests that the infants demonstrated multiple complex postural control strategies while learning to sit. However, once an infant became a competent sitter, he or she selected a few of the previously identified strategies to use routinely in sitting, making the postural control strategies appear more predictable during routine sitting. By the time infants have selected these preferred or routine strategies, they also have learned from their errors that some of the strategies are not successful.³⁴

This pattern of early complexity parallels the findings in supine postures in early infancy. Although a stable strategy (or set of successful strategies) emerges eventually, this does not mean that the infant cannot move out of these strategies. Because of the early complexity, the infant has access to multiple strategies that allow adaptation, depending on the events within the environment or the constraints of the situation.

These findings support the importance of early complexity for the development of action systems before the onset of the infant's first few functional skills. Thus, it is not postural control alone that supports function but rather highly complex variability of postural control that drives change and serves as a foundation for function in the first year of life.^4,17,18 From this early complexity, normal development of postural control can proceed to select successful strategies for functional skill.³⁰

Variable and Adaptive Postural Control Facilitates Exploration Through Movement Opportunities, Perception, and Action

Variability in postural control strategies is needed to maximize opportunities for exploration. As highlighted in the previous section, infants who are developing typically have both statistically variable and complex postural control in the first weeks of life in supine postures and during early sitting behaviors. Although some may view the variety of postural control strategies used by infants as errors they encounter while trying the find the most efficient strategy, we suggest the opposite.²⁸ The variety of postural control strategies used by infants who are developing typically is key to their global development.^1,35 The “errors” made while attempting to maintain a position or engage in a task are valuable as long as they provide the infant with a variety of experiences and lead to some success in reaching a goal. However, if the errors are too predictable and lack variety, complexity is lacking.

The capacity to orient the body to gather information and act on the world allows infants to gradually learn, explore, and build skills.^4,36,37 As such, postural control must dynamically and continuously change for an individual to engage with the world in multiple ways. A combination of experience, sensory information, and postural control support the development of motor skills such as sitting and reaching. While learning to sit, infants learn through active touch about surfaces to lean on and push against, which helps to select strategies to stabilize the body (postural control) under different conditions.²¹ Infants with sitting experience have improved ability to distinguish between novel and familiar objects because they have sensorimotor experience reorienting their body to pick up information via vision and active touch from the world around them.³⁶ Likewise, early locomotor activities (reciprocal versus symmetrical belly crawling) provide sensorimotor information that influences early reaching patterns.^35,38

These examples of the complex relationship among variable postural control, sensory perception, motor action, and cognition demonstrate the role of postural control in supporting the development of multiple action systems. Although reduced postural control is consistently associated with activity limitations in people with varying diagnoses and ages, reduced postural control in infancy is even more likely to result in long-term developmental deficits.^39–44 Therefore, the therapists' concern with early complex and adaptive postural control is not optional; it is a necessary component of every movement intervention.

How can this understanding of the interaction of variability in postural control with exploratory behavior be applied in the clinic? We propose that one way is to de-emphasize the focus of guidance to facilitate “normal” patterns of postural behavior, but rather change to suggestions or light touch cues to encourage a variety of strategies for postural control supporting functional movement. We encourage therapists to provide infants and children with a wide variety of opportunities to demonstrate exploratory actions that will increase complexity, a necessary precursor to the selection of an efficient movement strategy. Although not all of these movements will look efficient to an observer, the child seeking his or her own solution by using many strategies (even the unsuccessful ones) is important. Therefore, putting a child on a ball and tilting them to the side to elicit specific reactions would not be recommended as a technique during intervention and actually should be discouraged because it is not explorative, variable, or connected to any action system that is initiated by the child. Instead, we propose that using the construct of complexity, the therapist can modify the environmental supports, the goals, and the options for exploration, as well as giving touch cues to explore variable options (Fig. 3). Using the construct of complexity for assessment and intervention is explored in the next 2 sections.

Figure 3.

Example of therapeutic intervention to encourage the infant to perceive active weight shift forward in supported sitting. (A) The infant pushes posterior. (B and C) The infant experiences variable active practice shifting weight forward through the upper extremities and legs to achieve an environmental goal. (D) The infant maintains his trunk more upright while attempting to sit and reach.

A Lack of Complexity in Postural Control Might Be an Early Marker of Developmental Disabilities

Evaluation of complexity in postural control used during early motor development may aid in the early identification of infants who will have long-term postural control, movement, or cognitive deficits. Infants who are developing typically demonstrate high behavioral variability observed during spontaneous fetal and newborn movements.⁴⁵ Lack of behavioral variability in movements during the first 4 months of life may be indicative of developmental or neurological deficits in both full-term and preterm infants.^45,46

Recent work has begun to explore the statistical variability and complexity of postural control of infants born preterm during the first months of life using a combination of linear and nonlinear methods.²⁴ The COP movement was evaluated in supine postures during a single session between 1 and 3 weeks of age adjusted for prematurity. Infants born preterm exhibited larger statistical variability of the COP movement in the caudal-cephalic direction during spontaneous movements compared with infants born full-term. However, nonlinear analysis (approximate entropy) of the data revealed the presence of a more repetitive, less complex movement of the COP in the infants born preterm.

Video 1 includes 2 infants whose COP displacement was assessed during spontaneous movements at term age. The infant born full-term demonstrates a variety of movements and postural control strategies and has a small but complex COP displacement. The infant born preterm demonstrates repetitive and simple movements with limited postural control, resulting in a larger COP displacement and limited complexity, compared with the infant born full-term. Clinical observation of these infants reveals that the infant born full-term is able to lift his legs and kick reciprocally, move his head freely, and bring his hands to his mouth, all while staying in a supine position. The infant born preterm attempts to lift his legs a few times, but is unable to maintain the limbs off the support surface and appears to use the same unsuccessful strategy each time. This strategy results in the infant rolling from a supine position to his side each time both legs are off the surface. The infant born preterm is unable to rotate his head from one side to the other or lift one leg at a time. Video 1 exemplifies the use of complex and adaptive postural control strategies in the infant born full-term, but not the infant born preterm. Video 1 also demonstrates a link between behavioral variability and complexity. The preterm infant lacks behavioral variability and complexity, both of which are observed in the full-term infant. The limited postural control strategies used by the infant born preterm may result in fewer opportunities to explore how his body moves, reducing the infant's perceptual experiences and possibly altering the development of the postural control action system.

Harbourne and colleagues^30,34 have examined the postural control of infants with developmental delay and infants with cerebral palsy (CP) who are learning to sit. Using both linear and nonlinear variables together, the researchers were able to more completely describe the emergence of sitting, as well as differentiate between infants with developmental delays and infants with CP. The nonlinear variables served to provide information about small improvements in postural control over time that were not apparent with standard clinical tests such as the Gross Motor Function Measure.^30,47

Although the use of nonlinear analysis of COP data is not readily accessible in the clinic, literature identifying a lack of complexity in specific populations may demonstrate a need for earlier and more frequent motor assessments.^24,48 Research is ongoing on the relationship between complexity in postural control and clinically feasible measures of behavior and developmental assessments. During routine early intervention and developmental follow-up assessments, clinicians should observe infants during their spontaneous movements, keeping variability in mind. Infants with repetitive postural control strategies should be monitored very closely for emerging developmental delays that may result from their lack of experience with a variety of postural control strategies.

Enhancing the Complexity of Variability May Lead to Functional Changes and Improvement in Motor Function

Limited complexity during the development of postural control has been identified in infants born preterm and with CP, limiting their exposure to perceptual motor experiences in the first years of life.^24,30,34,49 In light of this evidence, we propose that therapeutic interventions for infants and young children with postural control and movement deficits should facilitate opportunities for infants to experience a wide variety of movements requiring different postural control strategies. The purpose of the intervention should be to increase the amount and variability of the experiences rather than focus only on one “correct” movement or postural control strategy. Complexity should be encouraged while learning a new skills. Intervention should encourage infants to alter their postural control based on the task demands, not to apply the same strategy to all tasks. Therapists can support infants' attempts at variable postural control by modifying the environment to facilitate a new experiences and postural control strategies while working toward a functional goal.

We propose that intervention to support variable postural control may be delivered through a combination of home-based and direct therapeutic interventions. Routine caregiving contributes to the development of skills as caregivers provide infants with experiences in a variety of positions, including supine, prone, upright carrying, and sitting with decreasing amounts of support. Increased experience with a variety of reaching tasks and environments has been shown to advance reaching and object manipulation in full-term and preterm infants in the first 6 months of life.^50–53 In each of these studies, parents provided increased experiences for their infants to practice reaching or general movement. The infants were not taught a specific movement. It was likely the increased opportunity for movements without restriction of the movement type or “errors” that provided these infants with a faster rate of reaching development then control infants.^50–53 Although this example does not specifically focus on postural control, it clearly demonstrates the need for appropriately timed experience and exploration to advance development. We propose caregivers should be encouraged and educated on how to provide infants with varied movement opportunities that will support the development of complex postural control and environmental exploration. For infants with poverty of movement or limited self-directed movement strategies, therapeutic guidance may be needed to aid them in perceiving and experiencing additional movement strategies. Exploration, either independent or guided, is important to expand infants' perceptions and experience with movement and to aid in the development of coordinated movement patterns in preterm infants.⁵⁴

Harbourne et al⁵⁵ (see their article in this issue) conducted a comparison of home program intervention and perceptual motor intervention provided by a therapist. Infants with CP or who were at risk for CP and under the age of 2 years were participants in an 8-week program started at the initial stage of sitting development. Infants in the traditional home program group received practice in static sitting opportunities during interaction with toys and equipment available in the home with the family. Sitting support was gradually reduced over time, but the focus of the home activities was not on variability. The infants in the perceptual motor group who received an intervention utilizing the concepts of variability, complexity, and refining multiple strategies for sitting showed greater improvements in the measures representing those variables compared with the infants in the home program group. Although both groups of infants showed similar gains in a behavioral measure (ie, the Gross Motor Function Measure sitting subscale), the variability measures from the COP data reflected subtle changes in postural control that give insight to adaptive control strategies (see Video 2).

The results of the study by Harbourne et al⁵⁵ suggest that therapists can address variability of postural control during intervention and successfully alter the progression of these features of postural control is an important step in examining outcomes for children with motor difficulties. Table 2 provides an overview of some treatment approaches we propose a therapist might use to support the development of variability in postural control. Figure 3 shows a child who is learning to sit. At the initiation of the session, the infant demonstrated that he perceived his back as a weight-bearing surface. This infant strongly pushes back against the surface to initiate any movement in a statistically variable but too regular (not complex) pattern (Fig. 3A). The first intervention can be a slight change in the environment, to help the infant to perceive that the space in front of him in sitting is supportive, and can be explored for postural support and toy exploration (Figs. 3B and 3C). Finally, touch cues, change of task, and environmental setup can help the infant find a successful strategy, which is to lean the trunk slightly toward the object of the reach each time he wants interaction (Fig. 3D). This successful strategy may have a slightly lower amount of statistical variability, but behaviorally he is able to apply a more adaptive strategy to solve the problem of attaining the toy during multiple reaches in slightly different parts of the play space in front of him. The structure of that variability also is more complex, similar to the time series in Figure 1B. This more complex pattern is more functional because he can adapt his posture so he can reach the toy.

Table 2.

Intervention Examples to Encourage Complex Postural Control

Although it is impossible within the scope of this article to fully interpret the ideas of complexity and variability in terms of assessment and intervention for postural control, we propose the following principles may be used for guidance:

1. Development is nonlinear, and there are individual differences among children, so each child may take a slightly different path to the same endpoint. Thus, earlier attempts at a new posture may have high irregularity, reflecting complexity of postural control.

2. Small and varied exploratory movements are essential building blocks for any new postural skill. These movements are characteristic of complexity, a type of variability necessary for adaptation within a dynamic environment. Ample opportunities to explore diverse movements (multiple strategies) are necessary to support the development of postural control.

3. Errors are made with these exploratory and divergent trials, but these errors are valuable as long as they lead to some successes in reaching an environmental or behavioral goal.

4. Postural control can be affected by multiple factors, including environment, sensation, perception, selective muscle control, practice, variability of that practice, cognition, overall health, and temperament.

5. Postural control is primarily prospective (as opposed to reactive) and used to engage with the environment and to support action systems (manipulation, attention, locomotion, orienting); thus, intervention should focus on infant-directed action so that prospective control is inherent in the therapeutic plan.

6. Postural control is the background of all other action systems and thus should be a primary focus as functional skills are changing over time.

Need for Research and Future Directions

Although progress has been made in understanding the role of variability in the postural control of infants and young children, additional research is needed. Longitudinal studies of the development of variability in postural control during young infancy are needed to identify early predictors of postural control deficits before motor delays are present. The relationship between postural control in infancy and functional skills such as reaching, rolling, sitting, and moving between positions needs to be quantified. The interaction among postural control, cognitive function, childhood functional activities, and coordination must be considered. Most importantly, we must evaluate the efficacy of interventions to enhance postural control and maximize children's abilities and participation.

Summary

Infants who are developing typically begin life with high complexity and redundant possibilities for postural control, which appear initially to be random, excessive, and nonfunctional. However, these movements serve as a base to explore within the environmental context and to develop a repertoire of strategies for each task given the infants' personal constraints. Thus, during spontaneous movement, infants are learning about the link between body segments and determining the optimal ways to adjust their body over a dynamic base of support in multiple positions. They also are using sensory systems to build perception-action linkages so they can predict postural needs for a variety of actions and to support reaching, looking, and moving through the environment. This continuum of changes in the complexity of postural control as new skills emerge reflects normal dynamics of development.

Research on the complexity of postural control strategies suggests that early complexity or a lack of repetitive movement patterns is a hallmark of normal development. Complexity of postural movements is higher at 1 to 3 weeks of age in full-term infants who are healthy than in preterm infants. The ability to adaptively use efficient postural control strategies increases as infants develop the ability to sit independently, regardless of the magnitude of their postural movements. The changing use of adaptability during development reflects the use of complex variability as skill develops. Infants who have or are at risk for developmental delays use more repetitive postural control strategies, limiting their movement experience.

In understanding the tremendous complexity and exploration of strategies that occur in typical development, we propose therapists can develop interventions that capitalize on variability and complexity. Whether the intervention focuses on a specific segment of the body, general strategies, or environmental modifications, principles that optimize the early complexity that develops toward successful adaptive strategies are likely to benefit the infant or child.