Linda Fetters

Chapter contents

Infants are dynamically changing

Human infants experience continual and profound changes in all systems during the first year of life. They begin to master control of their bodies in relation to the world while their central and peripheral nervous systems, as well as their musculoskeletal system, are in dynamic development.

Development requires exploration

Mastery of action during this dynamic development is achieved through extensive exploration and discovery (Adolph et al., 1993; Gibson, 1988, 1997; Gibson and Pick, 2000). Mastery of action is a process that relies on extensive practice and the variability that accompanies practice. The creative research of Adolph et al. (2003) suggests that emerging walkers explore their limits of balance and locomotion through practice of more than six hours a day, walking the equivalent of 29 football fields. Skilled action depends as much on error as is does on successful achievement of the chosen goals (Fetters, 2010). Overshooting or undershooting when reaching for a toy affords information to the infant, not only about the location of the toy but also about the forces necessary to successfully grab the toy.

Altered development for the infant and child with cerebral palsy

These dynamic processes of development are altered for the infant and child with cerebral palsy (CP), and this alteration has serious consequences for development. The limited exploration that is typical for the child with cerebral palsy reduces the available variability that is essential to the discovery and mastery of action. These constraints on variability have serious consequences on all systems. Children with cerebral palsy are smaller and weigh less than typically developing children (Campbell et al., 1989). The tissues of the musculoskeletal system are altered, leading to reduced contractibility, weaker muscles and potential contractures (Lieber and Bodine-Fowler, 1993; Lieber and Fridén, 2002) (see Chapter 6).

Compromised systems also constrain exploration through space for children with CP, which has potential impact on perceptual, social and cognitive development. Thelen and Smith (1994) describe the development of meaning and cognition by stating ‘Meaning has its origins in actions and is made manifest—created—in real time and through activity (p. 323)’. Development of the embodied mind (Johnson, 1987; Thelen, 2000) is indeed compromised for the child with constraints on action development such as the child with CP.

Early intervention is essential

The earliest intervention possible provides the best opportunity to change the developmental trajectory for all of these systems. Early intervention not only promotes the development of leg action but also promotes the neural foundations for this action. Eyre et al. (2000, 2001) (see Chapter 2) have demonstrated the critical importance of maintaining cortical inputs for the development of spinal circuits that maintain leg movements. Lesions of either the brain or spinal cord during fetal and neonatal development dramatically alter the formation and function of sensorimotor pathways. Lesions of the brain or spinal cord during critical periods of development lead to increased levels of cell death in the damaged pathways that alter the competitive dynamics, resulting in the retention of aberrant pathways (Bayatti et al., 2008). Evidence now exists that the corticospinal system is already active and shaping spinal circuits by the late prenatal period, but that these dynamics are derailed by pre- to postnatal insults (Eyre et al., 2000). Therapeutic interventions that motivate movement are potential substrates for driving these circuits during their most dynamic phase of plasticity.

We are developing an intervention to promote leg action during the very first weeks and months after birth. Our research seeks to understand the development of the control of leg action in order to use this foundational knowledge to inform the earliest possible intervention to improve leg action for infants at risk for CP. We have developed a line of research investigating the earliest leg movements of infants born full term (FT) and infants born prematurely who are at high risk for the development of CP. Infants born prematurely with very low birth weight (VLBW; less than 1,500 grams or about 3⅓ lbs), specifically those with white matter abnormalities, are at increased risk for CP, in particular spastic diplegia and other types of cerebral palsy (Spittle et al., 2011). Infants and children with spastic diplegia have reduced lower extremity movement, movement in atypical patterns, and delayed onset of mobility including walking.

Although it is typically thought that the hands are the first to explore and bring the world into view for infants, Thelen and colleagues suggest that exploration with the feet actually precedes exploration with the hands (Galloway and Thelen, 2004; Galloway et al., 2002; Heathcock and Galloway, 2009). The quality and quantity of lower extremity (LE) movement are critical for the development of mobility. Independent mobility enables the toddler to freely engage in self-directed learning through independent exploration of the environment. Mobility through the environment has been demonstrated to enhance socialization, perception and cognition for the developing infant, with an increase in each of these developmental domains with onset of crawling and walking (Bertenthal et al., 1984; Herbert et al., 2007).

One emergent property of typically developing leg action is the selective control of segments of the leg. We have demonstrated that this selective control is aberrant in prematurely born VLBW infants with white matter disorder (WMD) (Fetters et al., 2004) and other research groups have demonstrated the importance of selective control for independent locomotion in older children with CP (Fowler and Goldberg, 2009). Our work and the work of others support that the more dominant newborn synergies of total lower extremity flexion and total lower extremity extension are replaced with more selective control of segments with, for example, the hip joint moving into extension and the knee joint moving into flexion (Fetters et al., 2004; Heriza, 1988; Jeng et al., 2004; Vaal et al., 2000). Creeping, crawling, moving into sitting or standing and walking all require this disassociation of segments as selective control emerges. During infancy, these joint combinations have been described as moving from the dominant newborn in-phase interjoint coupling to a more out-of-phase interjoint coupling (Heriza, 1988; Thelen et al., 1983). An in-phase movement of the legs is characterized by all joints, hip, knee and ankle, moving in flexion or all joints moving in extension. These phase transitions have been studied using kinematics and kinetics. Using kinematics, the emergence of out-of-phase coupling is seen to begin with the hip–ankle joints at 1 to 2 months of age, followed by the hip–knee and knee–ankle joints over the next few months (Jeng et al., 2002; Piek, 1996).

Early intervention through contingency learning

This selective control of the legs is occurring during the period in which infants are discovering that their actions have consequences in the environment: that is, they discover the relationship between their actions and an environmental event. This relational learning is referred to as the learning of contingencies. The learning of contingencies occurs very early, as infants learn the connections between their actions and the effects of their actions on the world (Goldfield et al., 1993; Milewski and Siqueland, 1975; Siqueland and DeLucia, 1969). For example, 1-month-old infants demonstrate the ability to modify their sucking frequency to maintain the visual presence of a slide of a coloured shape (Milewski and Siqueland, 1975) and even to use sucking to effect the amount of luminance in a slide presentation (Siqueland and DeLucia, 1969). In a now classic study, Goldfield et al. (1993) demonstrated that 8-month-old infants learn to bounce at the resonant frequency of a baby-bouncer suspended by a spring. After a period of exploration characterized by high variability, infants learned to produce longer bouts of bouncing with a concomitant decrease in bounce amplitude and period variability. Exploration in contingency studies such as this is typically captured through measures of variability such as standard deviations. Improvement of performance and learning are then characterized as actions with reduced variability.

We know very little, however, about the characteristics of the exploratory processes associated with learning contingencies in the first months of life; it is unclear if exploration is a relevant parameter in learning, and, if it is, how exploration actually affects learning. Previous research with infants has sought to investigate the behaviours that immediately precede learning. For example, significant correlations have been found between the amount of familiarization decrement (lack of sustained levels of sucking to maintain a visual image) that occurs during the acquisition phase of a learning experiment and the amount of response recovery when a novel stimulus is introduced (Milewski and Siqueland, 1975). More recently, the effect of immediate previous kicking experience of one leg in a learning paradigm was seen to increase the rate of learning by the other leg (Angulo-Kinzler, 2001). Thus, performance in a learning paradigm may be influenced by prior experience, but this tells us little about the specific characteristics of the prior behaviour and its relation to learning. In particular, although the importance of exploratory actions in infant learning has been stressed by many authors, and the reduction in variability after a period of exploration has been documented, there is scant data supporting a causal relationship between exploration and learning or data detailing the necessary amount or timing of the exploratory actions adequate for learning.