Intraindividual Variability in the Development of

Adolph, K. E., Cole, W. G. & Vereijken, B. (in press). Intra-individual variability in the development of motor skills in childhood. In M. Diehl, K. Hooker, & M. Sliwinski (Eds.), Handbook of Intra-individual Variability Across the Lifespan. New York:
Routledge/Taylor & Francis Group. Preprint formatted by authors.
Intraindividual Variability in the Development of
Motor Skills in Childhood
Karen E. Adolph Whitney G. Cole
Beatrix Vereijken
Since the inception of the field, researchers have considered intraindividual variability to be a signature component
of motor development. In this chapter, we describe three types of developmental changes in intraindividual
variability in infancy and childhood and speculate about the potential functions of each pattern of developmental
change. Intraindividual variability decreases for repeated performance of the same movements; the apparent
function is improvement in consistency and control. Variability continues or increases when selecting and modifying
actions in response to variations in local conditions; the presumed function is greater behavioral flexibility and
adaptability. Developmental changes in the structure of intraindividual variability reflect changes in the temporal
organization and coordination of movements; the putative function is to ensure sufficient consistency for successful
performance and sufficient flexibility to cope with changing circumstances. We use case studies to illustrate each
type of developmental change in intraindividual variability and to highlight the time scales of motor actions
typically associated with each pattern of developmental change. We conclude the chapter by identifying important
shortcomings of current developmental research, including lack of integration across recognized patterns of
developmental change, overreliance on artificial laboratory tasks, and lack of experimental evidence for how
different types of variability promote or hinder development. We offer suggestions for future research that promises
to facilitate a deeper understanding of the role of intraindividual variability in motor development.
Keywords: variability, motor development, infants, children, consistency, flexibility, time scales
Introduction And Chapter Overview
a novice walker, speed and step length vary willy-nilly
from step to step causing a random pattern of variability,
but in an experienced infant walker, step length and speed
increase in tandem over the sequence and step-to-step variability has a coherent structure (Badaly & Adolph, 2008).
As these examples of infant walking make clear, the term
“intraindividual variability” can refer to very different aspects of motor behavior occurring at very different temporal
and spatial scales. Researchers can observe intraindividual
variability in the milliseconds of a single step, the seconds
that comprise a sequence of steps, and the months and years
of developmental changes in walking. Variability is apparent in repetitive, cyclical movements such as walking or riding a bicycle, discrete movements such as executing a reach
or doing a cartwheel, and in continuous movements such as
maintaining a sitting or standing posture. We see variability
in the real-time strategies children select for grasping the
handle of a spoon (e.g., palm up or palm down) and in the
developmental strategies infants select for posture and locomotion (e.g., crawling on their bellies or on hands and knees).
Research on motor development encompasses all of
it, but not all at once by a single research team within a
single study. Instead, researchers who are interested in the
development of motor control tend to focus on decrease
in intraindividual variability and to rely on standardized
laboratory tasks in which consistency is paramount. Studies on developmental improvements in adaptability tend
to focus on continuance or increase in intraindividual
variability and to test infants in novel and variable conditions in which behavioral flexibility is critical. Studies
on developmental changes in the organization of move-
However you look at it, children’s movements are variable. Intraindividual variability is a signature component
of motor development. A toddler’s foot traces a variable
trajectory through the air, each shaky step is a different
wobbly variant of the previous one, and the shape of the
walking path varies from trial to trial. For some aspects of
motor skill, intraindividual variability decreases over development: An infant’s foot trajectory becomes smoother
in the course of a step, each step in a sequence becomes
more similar to the last, and the walking path becomes more
consistent across trials (Cheron et al., 2001; Looper, Wu,
Barroso, Ulrich, & Ulrich, 2006). For other aspects of skill,
variability continues or increases over development: Foot
trajectory varies to clear obstacles in the path, steps vary to
accommodate variations in the terrain, and the shape of the
walking path varies to navigate a cluttered room (Adolph
et al., 2012; Dominici, Ivanenko, Cappellini, Zampagni,
& Lacquaniti, 2010). For still other aspects of motor skill,
the structure of variability changes over development: In
Correspondence concerning this article should be addressed
to Karen E. Adolph, Department of Psychology, New York University, 4 Washington Place, Room 410, New York, NY 10003.
Electronic mail may be sent to [email protected].
Work on this chapter was supported by National Institute of
Health and Human Development Grant R37-HD33486 to Karen
E. Adolph. We gratefully thank Gladys Chan for her assistance
with figures.
1
2
ment tend to focus on the structure of variability and to
record long time series of movements in which optimal
levels of predictability and complexity are important.
Descriptions of variability and how it changes are common in research on motor development. So, rather than
present a comprehensive review of the field, we have selected illustrative case studies to highlight our points. Given
that intraindividual variability is endemic in motor development, the overall aim of the chapter is to examine the potential functions of developmental changes in intraindividual
variability—how developmental changes in intraindividual
variability might serve to promote motor skill acquisition.
We begin the next section by asserting that variability—both intra- and interindividual variability—is integral
to motor development and that researchers have always
treated it as such. We describe the work of the early pioneers and how their legacy of studying variability has influenced modern investigators. The subsequent three sections
reflect our focus on the potential functions of developmental changes in intraindividual variability. We describe
how motor development can entail decrease in variability, continuance or increase in variability, and changes in
the structure of variability. Our case studies illustrate the
types and time scales of intraindividual variability that are
typically associated with the particular pattern of developmental change. Each section ends with speculations about
the potential functions of that developmental pattern. We
conclude the chapter with suggestions for future research,
which promise to facilitate a deeper understanding of the
role of intraindividual variability in motor development.
Variability Is Integral To Motor Development
The most familiar and iconic representation of motor development is the so-called “milestone chart.” As illustrated
in Figure 4.1, the common features of such charts are the ordering of skills by age of attainment and the associated age
norms for each skill. Typically, the salient aspect of the chart
is the average age at which children display each skill (here,
shown by the black vertical lines along each bar). However, most milestone charts also depict an important measure
of interindividual variability: the normative range in age of
attainment (total length of each bar). For postural and locomotor skills, age bands span several months and show
considerable overlap. Moreover, the depicted age ranges
would be considerably wider if onset ages were normed
across different cultural groups and reflected the true diversity of human development (Adolph & Robinson, in press).
Although the great pioneers in motor development—Gesell, McGraw, and Shirley—are responsible for our legacy
of milestone charts, age norms, and the related fictions of
“the average child” and prescribed developmental “stages”,
these same researchers viewed intra- and interindividual
variability as fundamental to development (Newell, Liu, &
Mayer-Kress, 2003; Thelen & Adolph, 1992). Gesell, for example, is most famous for popularizing normative summa-
Walks Alone
Stands Alone
Cruising
Pulls to Stand
Stands with Support
Crawling
Sits without Support
Rolls Over
Prone, Chest Up; Uses Arms for Support
Prone, Lifts Head
0
1
2
3
4
5
6
7
8
9
Age (months)
10
11
12
13
14
15
16
Figure 4.1. Typical infant milestone chart. Vertical lines depict
average age at onset. Horizontal length of bars represents the
normative age range. Skills are ordered from least to most mature based on the average onset age. Data from Bayley (1969)
and Frankenburg et al. (1992). Adapted from Adolph, Karasik,
& Tamis-LeMonda (2010).
ries of development, as in the 9-month-old who “sits alone”
and “opposes thumb in seizing a cube” and the 10-monthold who “pulls self up to stand” and “plucks pellet with precise pincer prehension” (Gesell & Thompson, 1929, p. 132).
But Gesell (1925) also recognized that infants frequently express behaviors belonging to other age norms and wrote extensively about “developmental peers of incongruent age” .
Interindividual Variability
Since the inception of the field, interindividual variability—differences between children—has been a central focus
of research. The early pioneers acknowledged that individual
infants need not progress lockstep through each stage in the
same order. Stages can be skipped, revisited, and expressed
in variable orders (Ames, 1937; Gesell & Ames, 1940; McGraw, 1945). Gesell and McGraw liberally sprinkled their
work with data tables and figures illustrating individual differences in onset age and duration of expression (Gesell &
Thompson, 1938; McGraw, 1941). Many early works list
the onset age of every infant studied (Ames, 1937; Shirley,
1931, 1933) and contain graphs of individual infants’ developmental trajectories (Ames, 1937; McGraw & Breeze,
1941). Indeed, McGraw (1935) dedicated an entire book to
a description of twins Johnny and Jimmy, replete with foldout graphs of their individual trajectories. Detailed qualitative descriptions of individual infants also were common
(Burnside, 1927). Such case studies, including descriptions
of illnesses that prevented movement, periods of rapid
weight gain, and changes in body dimensions, led researchers to speculate that individual differences in onset ages
stem from differences in infants’ experience, physique, and
temperament (Gesell & Thompson, 1938; Shirley, 1931).
3
The pioneering researchers also reported individual differences in infants’ abilities within particular stages
of motor development. Average step length and walking
speed, for example, vary widely among infants at the same
stage of walking development (McGraw & Breeze, 1941).
Shirley (1931), who painstakingly recorded walking skill
based on footprints created by dipping infants’ feet in oil
and sprinkling graphite over the oily marks, reported that
each infant’s footprint paths were so distinctive that researchers could deduce which baby had produced them.
Although modern researchers no longer focus on skill
onset ages, the practice of depicting individual differences among infants is widely adopted and descriptions of
individual trajectories are commonplace. Toddlers, for example, discover different strategies to accomplish walking. Some initiate movement by lifting their legs in short,
tiny steps; some do it by twisting their torso and swinging
their legs from side to side; and some initiate movement
by leaning forward and falling into each step (McCollum,
Holroyd, & Castelfranco, 1995; Snapp-Childs & Corbetta, 2009). Similarly, weekly observations of infants over
the transition to reaching reveal that each infant has a different movement problem to solve (Thelen et al., 1993).
More active infants must dampen inertial forces from
spontaneous arm flaps to guide their hands to the target,
and more sedentary infants must generate muscle force to
lift their arms to the target. All infants learn to adjust the
force and compliance of their arms, but each finds an individual solution to the problem. Each begins at a different starting point and forges a unique developmental path.
Intraindividual Variability
The great pioneers also recognized the importance of intraindividual variability—variability expressed by individual children. Infants frequently straddle adjacent stages and
occasionally express behaviors from distant stages (Gesell
& Ames, 1940; McGraw, 1941). Gesell (1946) viewed overlapping stages as integral to development. Infants straddle
multiple stages because they display the behaviors characteristic of their current level of maturity, they revert to
less mature forms as part of a cyclical process of reciprocal interweaving, and they display behaviors from the next
stage as part of the mastery process (Gesell & Ames, 1940).
Recent research reveals a similar story. Motor skills do
not turn on and off like a faucet. Rather, skills sputter in and
out of infants’ repertoires, with variable, oscillating trajectories spanning days, weeks, or months (Adolph & Robinson,
2011; Adolph, Robinson, Young, & Gill-Alvarez, 2008). Infants typically express a new skill to criterion on one day,
but not on the next; likewise, they reject an old skill on one
day, but not on the next. The developmental transition from
crawling to walking, for example, takes most infants several
months. During the transition period, both skills are alternately expressed within and across days, and long after crawlers become walkers, they revert to crawling when needed.
Development Entails Decrease In Intraindividual Variability
As researchers and parents can attest, motor development
often entails decrease in intraindividual variability. When
infants first achieve new motor skills, movements are noisy
and unstable. Newly reaching infants often miss the target.
Novice sitters wobble and fall. As infants gain strength, control, and experience, movements become smoother and more
economical and performance becomes more successful and
consistent. After months or years, children eventually achieve
adult-like stability in execution and performance outcome.
From this perspective, intraindividual variability in early skill acquisition reflects poor motor control (Deutsch
& Newell, 2005; Vereijken, 2010). A key component of
motor development therefore is acquiring the necessary
control to reduce variability. Increase in control arises
from several factors: less noise in the sensory-motor system (Schmidt & Lee, 2011), better and faster use of perceptual feedback, and greater reliance on feedforward
mechanisms (Deutsch & Newell, 2005). The current section illustrates the classic developmental trend of decreasing intraindividual variability accompanying increasing
control in basic manual, postural, and locomotor skills.
Variability Within A Single Movement
Long before infants can reach successfully for objects,
they spontaneously flap their arms in bouts of flailing, variable movements. Arm flapping increases in the presence
of a toy and brings the hand closer to the toy—as though
infants carve goal-directed reaching out of the noisy, variable arm movements already in their repertoire (Bhat &
Galloway, 2006; Lee, Bhat, Scholz, & Galloway, 2008).
Between 11 and 24 weeks of age, infants finally make
consistent contact with objects (Berthier & Keen, 2006;
Clifton, Muir, Ashmead, & Clarkson, 1993; Konczak,
Borutta, Topka, & Dichgans, 1995). Flapping the arms as a
prelude to reaching eventually disappears (Thelen, Corbetta, & Spencer, 1996), but the arm movements for reaching
are still highly variable. Perhaps the most salient developmental change in infant reaching is a straighter, more direct
hand path to the goal (Berthier & Keen, 2006; Konczak et
al., 1995; von Hofsten, 1991). In adults, the length of the
reaching path is roughly equivalent to the initial distance between hand and target (Konczak et al., 1995). The hand rapidly accelerates toward the target and then slows and adjusts
for grasping. In newly reaching infants, however, the hand
lurches and swerves through the air in an indirect zigzag
path that can be 2.2 to 3.8 times longer than the straight-line
distance to the object (Konczak et al., 1995; von Hofsten,
1991). Each zigzag change in direction results in acceleration and deceleration of the hand. Over development, infants’ reaches become straighter (Figure 4.2A-C). As the
4
reaching path becomes more direct, the number of speed
peaks and jerk (change in the rate of acceleration) decrease
(Berthier & Keen, 2006; von Hofsten, 1991). Straightness
continues to improve and by two to three years of age, the
length of the hand path is only 1.2 to 1.4 times the direct path
to the object (Berthier & Keen, 2006; Konczak & Dichgans,
1997). Variability in infant reaching also illustrates a general
developmental trend: Improvements in proximal body segments occur before distal ones. Here, shoulder movements
become smooth and consistent before infants tackle the
challenge of controlling the elbow (Berthier & Keen, 2006).
After infants successfully reach for an object, they face
the problem of how to grasp it. Development progresses
from immature whole-hand power grips where the object
is pressed between fingers and palm to more mature pincer grips where the object is held between thumb and forefinger (Halverson, 1931). However, from 6 to 20 months
of age, infants employ multiple configurations of both
power and pincer grips. Even 6-month-olds can perform
pincer grips—but they do not do so consistently (Butterworth, Verweij, & Hopkins, 1997). In fact, on repeated
presentations of objects, 6- to 11-month-olds switch grip
types as frequently as they repeat a single grip. From 12
to 24 months of age, infants still employ different grip
types from trial to trial, but the mature pincer grip becomes increasingly dominant (Butterworth et al., 1997).
Tool use provides another example of decreasing vari-
A
ability. When the handle of a tool points away from their
dominant hand, adults consistently adjust their initial grip
to ensure “end-state comfort” (Rosenbaum, Chapman, Weigelt, Weiss, & van der Wel, 2012). In contrast, between 9
months and 4 years of age, children use a variety of strategies to grasp a spoon oriented away from their dominant
hand, many of which reflect “start-state,” not end-state,
comfort and some of which fail to bring food to mouth
without spilling (Keen, 2011; Keen, Lee, & Adolph, 2014).
Eight-year-olds use mature grasps more frequently, but
still revert to awkward, immature strategies on some trials. Development entails gradually winnowing out ineffective grip strategies and retaining more effective ones.
Variability In Repetitive Movements
When adults walk on a treadmill or in a continuous,
straight line over uniform ground, their movements are cyclical and repetitive. Although not perfectly identical, each
step is similar to the last. Movements on one side of the
body mirror those on the other side. But when infants first
begin walking, nearly every aspect of their gait is variable
and inconsistent—the placement of their feet on the ground,
the timing of leg movements through the air, the forces
produced by their leg movements, and the muscle actions
that produce those forces (Chang, Kubo, Buzzi, & Ulrich,
2006; Hallemans, Aerts, Otten, De Deyn, & De Clercq,
B
y
C
end
end
y
2 cm
x
start
2 cm
start
x
y
end
z
end
2 cm
y
2 cm
z
start
Figure 4.2. Sample traces of variability in the reaching path of one infant observed longitudinally. (A) The plane in which reaches are
represented. The top row shows front views of each reach, and the bottom row shows side views of each reach. (B) Front and side
views of a reach at 19 weeks of age; the length of the reach path is 1.9 times the initial distance from the hand to the object. (C) Front
and side views of a reach from the same infant at 31 weeks of age; the reach path is 1.2 times the distance to the object. Adapted from
von Hofsten (1991). With permission.
5
2004; Ivanenko et al., 2004). Intraindividual variability decreases most rapidly in the first few months after walking
onset, followed by a long period of gradual improvements
in which consistency finally approximates that of adults.
As illustrated in Figure 4.3A, new walkers’ step length
is highly variable. As shown in Figure 4.3B, variability in
step length (indexed with the standard deviation or coefficient of variation) decreases after a few months of walking experience (Adolph, Vereijken, & Shrout, 2003; Chang
et al., 2006; Clark, Whitall, & Phillips, 1988; Looper
et al., 2006). Stance time (when one foot is on the floor)
and double support time (when both feet are on the floor)
also show rapid and dramatic decrease in variability over
the first three months of walking experience (Chang et al.,
2006). Although new walkers achieve 50% phasing on average—meaning that the time when each foot touches down
is about halfway through the other foot’s step—phasing
fluctuates by 9% across a sequence of steps. After three
months of walking experience, phasing varies by only 3%,
and is no longer more variable than that of adults (Clark
et al., 1988). The most dramatic improvements in consistency occur in the months following the onset of walking,
but variability continues to decrease throughout childhood.
Variability in stride time decreases between 3 and 14 years
of age (Hausdorff, Zemany, Peng, & Goldberger, 1999),
and walking speed becomes more consistent across trials
until 15 years of age (Muller, Muller, Baur, & Mayer, 2013).
Variability in the coordination of joint movements within
each leg also decreases over the course of learning to walk.
The decrease is linked with practice keeping balance while
taking independent steps, not merely practice with the stepping movements. Prior to walking onset, infants can walk
with support if an adult holds their hands. Thus, infants can
practice executing stepping movements without controlling
their own balance. In supported walking, movements of
the hip, knee, and ankle are highly variable—the relative
phasing of joint movements is irregular and standard devia-
tions are large (Ivanenko et al., 2004; Ivanenko, Dominici,
Cappellini, & Lacquaniti, 2005). Infants show no change in
intraindividual variability at walking onset, but variability
decreases rapidly over the next few months and patterns of
intralimb coordination become more cyclical and consistent
(Clark & Phillips, 1993; Ivanenko et al., 2004; Ivanenko et
al., 2005; Lasko-McCarthey, Beuter, & Biden, 1990; Polk et
al., 2008). Infants display roughly adult-like consistency in
patterns of intralimb coordination after 3 to 6 months of walking experience (Clark & Phillips, 1993; Polk et al., 2008).
Variability In Continual Movements
Adult sitters are not perfectly motionless. High-resolution recordings of body position reveal continual postural
sway—small, slightly variable movements of the torso. But
in infant sitters, intraindividual variability is visible to the
naked eye—like watching a coin teeter on edge before it
topples. Detailed recordings of infants’ head position reveal the nature of the developmental changes (Saavedra,
van Donkelaar, & Woollacott, 2012). As shown in Figure
4.4, before infants can sit independently, postural sway is
variable but unidirectional: After a few seconds, infants
lose balance and collapse forward. Next is a “rise and
fall” period where infants pull themselves upright, only
to fall forward again. After infants achieve sufficient control to maintain a sitting posture, they wobble precariously around the upright position, frequently over- and under-correcting their position until they fall. Finally, infants
achieve functional, more adult-like sitting where sway varies within a tighter area around midline. These stages are
continuous and infants’ postural sway can reflect multiple
stages within a single session. The amplitude of postural
sway continues to decrease between 2 and 14 years of age
(Riach & Hayes, 1987) and the total path length drawn
by variation in the center of pressure decreases between
6 and 10 years of age (Hong, James, & Newell, 2008).
A
Y (cm)
60
40
20
0
B
Y (cm)
60
40
20
0
100
200
300
400
500
X (cm)
Figure 4.3. Variability in the footprint paths of infants tested in the standard gait paradigm. Infants walked over a pressure sensitive
carpet that recorded the placement of each footfall. (A) Typical footfall patterns in a novice walker (2 weeks of walking experience).
(B) Typical footfall patterns in a more experienced infant walker (2 months of walking experience). Adapted from Adolph & Robinson
(2013). With permission.
6
Like sitting, learning to stand entails decrease in intraindividual variability. When infants first pull themselves
to a stand, they display large swaying movements around
the upright position, even while holding a support rail.
Now they sway around the ankles rather than the hips as
in sitting. Standing sway variability continues unabated
through subsequent milestones in postural development—
when infants can stand alone and when they take their first
walking steps (Barela, Jeka, & Clark, 1999). In fact, variability in the developmental transition to upright balance
imposes a transient perturbation to sitting posture. When
infants first begin walking independently, they show a brief
increase in the variability of postural sway while sitting;
their new walking skill seems to briefly disrupt control
of sitting, leading to a spike in intraindividual variability
(L. C. Chen, Metcalfe, Jeka, & Clark, 2007). But after 6
weeks of walking experience, the amount of standing postural sway decreases substantially (Barela et al., 1999).
Intraindividual variability in standing sway velocity
also improves after walking onset (L. C. Chen, Metcalfe, Chang, Jeka, & Clark, 2008; Metcalfe et al., 2005).
At walking onset, infants’ swaying movements are rapid
and changes in velocity are erratic. With walking experience, sway velocity decreases and becomes more consistent, indicating a shift from rapid, reactive corrections to
slower, more prospective compensations. Refinements
in the amount of sway variability continue through childhood until finally decreasing to adult levels (Hong et al.,
2008; Newell, Slobounov, Slobounova, & Molenaar, 1997).
Potential Functions
Consistency and control. Researchers consider decrease
in intraindividual variability to be a signature of skilled performance and control for good reason. For basic motor skills,
the costs and benefits seem obvious. Decrease in variability
accompanies more consistent, economical, accurate, and
adult-like performance. Reciprocally, high variability incurs
immediate penalties. Variable infant reachers risk missing
the target. Wobbly infant sitters and toddlers risk toppling
over. Moreover, variability in one movement has the potential to disrupt other movements, with cascading effects
into domains far removed from motor control. Variability in
sitting disrupts reaching, grasping, and object manipulation
(Rachwani et al., 2013; Soska & Adolph, 2014), and lack
of experience with objects, in turn, impedes learning about
object form (Soska, Adolph, & Johnson, 2010). Data from
clinical populations supports the notion of decreased variability and improved motor control. Children with Down’s
Syndrome, for example, frequently show elevated amounts
of intraindividual variability in basic motor skills relative
to typically developing children (Looper et al., 2006).
A caveat. Despite clear evidence linking intraindividual variability with motor control, a caveat is in order. The
endpoint of development is not elimination of variability.
A
B
C
D
Figure 4.4. Characteristic patterns of variability in postural sway
in the development of infant sitting. Plots show overhead views
of the path of the head relative to the base of support (represented
by drawn circle) during one trial from four infants. (A) Earliest
period in the development of sitting. Infant sits upright for only a
few seconds before collapsing forward (119 days prior to the onset
of independent sitting). (B) Second period in the development of
sitting in which infants attempt to recover from errors in postural sway by pulling the torso backward and forward in a rise and
fall pattern (68 days prior to sitting onset). (C) Third period in the
development of sitting in which the infant’s body wobbles around
the upright (54 days prior to sitting onset). (D) Fourth period—sitting onset and thereafter in which infants have attained functional
independent sitting (0 days prior to sitting onset). Adapted from
Saavedra et al (2012).
Intraindividual variability only represents error for particular skills measured in particular ways. The presumed
importance of reducing intraindividual variability is due
in part to the way researchers study basic motor skills. In
most standard paradigms, participants’ bodies, the environment, and task constraints are held constant: Infants
reach repeatedly toward similar objects at midline; they
step on a motorized treadmill or walk on straight paths
over uniform ground. The task demands require consistency. Intraindividual variability is therefore considered
to be errors in execution or noise in the sensorimotor
system and is taken as evidence of poor motor control.
As we argue in the following sections, the role of variability in motor development is not so simple. Alternate
ways of testing motor skill and measuring variability reveal
a richer, more complex developmental story about intraindividual variability (Deutsch & Newell, 2005; Vereijken,
2010). Although variability in many basic motor skills decreases with experience, other aspects of intraindividual
variability show different patterns of developmental change.
Development Entails Continuance Or Increase In Intraindividual Variability
Intraindividual variability is a defining characteristic of
spontaneous activity such as fetal motility, infant exploration, and gross motor play (Adolph & Robinson, in press;
Hadders-Algra, 2011). Indeed, the so-called “stereotypies”
displayed by typically developing infants comprise such diversity of movement across so many body parts that they
are not stereotyped at all (Piek & Carman, 1994; Thelen,
1979). To the contrary: Rigidly stereotyped movements
during spontaneous fetal and infant activity are a harbinger of developmental disability (Hadders-Algra, 2011).
Intraindividual variability is also a central component of goal-directed motor behavior. In fact, for successful performance of most everyday, real world activities,
movements cannot be performed in the same way over
7
and over because local conditions are always in flux. Instead, motor behavior must be adapted to variations in the
body, environment, and task. Think of modifying your
movements to type with long fingernails, steer through
a cluttered living room, or lob a tennis ball beyond your
opponent’s reach. Consider alternative strategies for tying
your shoes while pregnant, descending a high rock wall,
or pitching to a left-handed hitter. Variable movements
and a variety of strategies allow motor actions to be tailored flexibly to the current situation. This section focuses
on continuance and increase in intraindividual variability.
No Costs Or Benefits
Some motor behaviors have no external incentives or
developmental pressures to decrease or increase variability.
With no immediate costs or benefits for intraindividual variability, children typically maintain variability at the original
levels until they acquire a new skill or the situation changes.
Belly crawling provides an apt example of a motor behavior with continued variability. Many infants belly crawl
for several weeks prior to crawling on hands and knees.
Because the torso rests on the ground during part or all of
each belly-crawling cycle, balance constraints are minimal.
As a consequence, infants achieve forward propulsion using
arms, legs, belly (and sometimes head) in a seemingly limitless variety of combinations (Adolph, Vereijken, & Denny,
1998; Freedland & Bertenthal, 1994). Belly crawlers display
variable patterns of interlimb coordination from cycle to cycle, trial to trial, and day to day. A single infant might exhibit
a near-trot with arms and legs on diagonal sides of the body
moving together, then a quasi-bound with first arms then legs
moving together, then an army crawl moving only the arms
and dragging the legs. One or both arms might drag with
the legs pushing against the floor, sometimes inspiring infants to bear weight on the face. The infant might inchworm
forward by popping up on the toes and plopping down on
the abdomen. Intraindividual variability continues unabated
throughout the belly-crawling period; even after 10 weeks
of experience, infants continue to add new forms to their
repertoire (Adolph et al., 1998). In contrast, once infants
begin crawling on hands and knees, they quickly converge
on a near-trot (Adolph et al., 1998; Freedland & Bertenthal,
1994; Patrick, Noah, & Yang, 2009, 2012). Presumably, balance becomes imperative with the abdomen off the floor and
moving diagonal limbs together keeps the torso more stable.
Natural walking behavior provides another salient example. When infants are tested in the standard laboratory task—
walking repeatedly in a straight line at a constant speed—the
consistency of gait measures improves over the first three
months of independent walking (Chang et al., 2006). But
in everyday activity, infants do not walk in a straight line at
a constant speed. Instead, natural walking paths are highly variable (Adolph et al., 2012). Infants generate twisting,
turning, omnidirectional paths (Figure 4.5). They loop back
over areas they have already visited. They make frequent
starts and stops. Each step is not like the last in part because
turns, back steps, side steps, double steps, and so on necessitate variation. Natural walking patterns are variable regardless of infants’ individual level of walking experience. Variability in walking in the standard laboratory task decreases,
but natural locomotor exploration retains its variability.
Modifying Ongoing Movements
Body size and proportions are not static. The natural
environment is not uniform. Task constraints are not constant. Intraindividual variability is necessary for adaptive action because variations in the body, environment,
and task change the biomechanical constraints on action.
Movements must be modified online to suit the demands
of the current situation. Variations in movement are generated purposely to achieve a desired goal, and the outcome
is increase in the amount of intraindividual variability.
Carrying an asymmetrical load, for example, changes the
body’s functional dimensions. Adult walkers compensate
by varying body posture: They lean in the direction opposite
the load (while carrying a suitcase in their left hand, adults
lean to the right, etc.). School children routinely carry heavy
backpacks (more than 15% of body weight). To compensate
for the backload, they modify posture by leaning forward
and heavier backpacks elicit more forward leaning (Brackley, Stevenson, & Selinger, 2009). Although infants can
vary body posture by leaning in different directions, they
do not do so in response to loads applied to their bodies. In
contrast to older children and adults, infants do not compensate for asymmetrical loads comprising 15% of body weight
(Garciaguirre, Adolph, & Shrout, 2007). Instead, they lean
into the load and accommodate as best they can. As a consequence of failing to modify posture—failing to generate
appropriate movements in response to changing body constraints—gait patterns are disrupted and infants misstep and
fall (Garciaguirre et al., 2007; Vereijken, Pedersen, & Storksen, 2009). Backloads are more disruptive than front or side
loads, presumably because backloads are most novel; during
play, infants spontaneously carry loads to the front and
sides of their bodies, but they do not wear baby backpacks.
A sloping ground surface also changes demands for
keeping balance: the steeper the slope, the more challenging the demands. Walkers can curb forward momentum by
taking smaller, slower steps, but infants have little experience walking down steep slopes (Gill, Adolph, & Vereijken,
2009). Although infants have the capacity to modify step
length and walking speed, at first they do so haphazardly
across a wide range of shallow and steep slopes (Figure
4.6A). Variations in movements are applied erratically. For
example, in Figure 4.6A, the trials with 5 and 21 steps—
the minimum and maximum step numbers—were both at
the same 18° slope. After a few weeks of practice walking
down a range of shallow and steep slopes, infants learn to
modify their steps in accordance with 2° changes in degree
of slant (Figure 4.6B). Variable movements are generated
8
Figure 4.5. Variability in natural infant walking. Trace depicts one typical 13-month-old’s walking path through a laboratory playroom
during 10 minutes of spontaneous play. Adapted from Adolph et al (2012). With permission.
in precise accordance with changes in the environment
(Figure 4.6C-D). Moreover, gait modifications show anticipatory planning and control. Infants begin braking before stepping onto a steep slope and they take longer steps
toward the bottom of the slope (see cluster of footprints
before crossing the brink of the slope and widely spaced
footprints toward the landing platform in Figure 4.6D).
Like variations in the body and environment, variations in the task require children to generate variable
movements and apply the variations systematically. For
example, if the task is to squeeze through a narrow doorway, infants—like adults—modify locomotion by turning
sideways in accordance with doorway dimensions (Comalli, Franchak, Char, & Adolph, 2013; Franchak & Adolph,
2012). If the task is to balance along a narrow ledge, infants
and adults also turn sideways, but now they are less likely to attempt increments that are too narrow for passage.
Infants also adapt movements to variations in manual
tasks. When 10-month-olds intend to toss a ball into a large
bin, they reach faster for the ball than if they intend to fit
the ball into a tube because the subsequent act of throwing
requires less precision than the act of fitting (Claxton, Keen,
& McCarty, 2003). Similarly, if toddlers intend to place a
block into a container, they reach faster for the block than if
they intend to balance it on top of a block tower (Y. P. Chen,
Keen, Rosander, & von Hofsten, 2010). Refinements in
reaching speed continue from 4 to 11 years of age, depending
on whether children intend to fit an object into a tight or loose
space, or toss it into a bin (Wilmut, Byrne, & Barnett, 2013).
Variety Of Means
Sometimes modifications of ongoing movements are
insufficient to achieve the goal. Such situations require alternative strategies. Infants must select alternative actions
from their motor repertoire or generate new solutions on
the fly. Indeed, variety of means is a hallmark of adaptive motor behavior (Gibson & Pick, 2000; Piaget, 1952).
One might expect, for example, that crawling infants
cannot carry objects because their hands are used for locomotion. However, crawlers do spontaneously carry objects—on average 6 times per hour—and they discover a
variety of means to do so (Karasik, Adolph, Tamis-LeMonda, & Zuckerman, 2012). Infants crawl with an object in
hand, tucked under one arm, in the mouth, or by pushing
it along the floor. Crawlers also vary their method of lo-
9
Potential Functions
Flexibility and adaptability. Increased movement variability is a good thing when it functions to tailor motor behavior to the current situation. Indeed, intraindividual variability is an imperative for everyday action because it allows
behavior to be flexibly adapted to current constraints. Outside the laboratory, local conditions are never constant; rigid adherence to a consistent way of moving would be maladaptive. Instead, children must perceive the current status
A
Session 1
r = .22
Steps
18
12
6
0
0
B
8
16
24
32
16
24
32
Session 10
r = .90
18
Steps
comotion by hitching in a sitting position while holding
objects in hand, and by cruising while holding furniture
with one hand and an object in the other. Walking infants
carry objects more frequently than crawlers, averaging
43 carrying bouts per hour, and they also display a variety of means. Walkers carry one large object in two hands
or two small objects in each hand; they hold one object
in one hand while leaving the other hand free, grasping a
caregiver’s hand, or holding onto furniture (Karasik et al.,
2012; Mangalindan, Schmuckler, & Li, 2014). Individual
infants display multiple methods of carrying. Diversity of
carrying methods is more prevalent in crawlers for whom
carrying is more difficult, and more experienced crawlers use more carrying methods (Karasik et al., 2012).
When faced with the novel predicament of descending a
large step, high cliff, or steep slope, experienced infant walkers discover a variety of means for coping with the obstacle
(Adolph, 1997; Kretch & Adolph, 2013). Of course, on safe
increments, they can descend using their typical walking
method and on risky increments, they can simply avoid going. However, experienced infants typically do not avoid
descent. They find alternative means by scooting down in a
sitting position, backing down feet first, sliding down prone
headfirst, crawling (hands first), kneeling (legs first), or squatting using the hands on the starting platform. Across trials
within a session, experienced walkers displayed 2 to 7 strategies (4 on average) for descending cliffs (Kretch & Adolph,
2013). Across sessions, walkers displayed 2 to 6 strategies (5 on average) for descending slopes (Adolph, 1997).
Variety of means is so endemic in infant motor behavior
that infants sometimes surprise researchers by the creativity and inventiveness of their strategies. Sixteen-month-olds
spontaneously use a sturdy wooden handrail to augment
their balance while walking over narrow bridges (Berger &
Adolph, 2003; Berger, Adolph, & Kavookjian, 2010). But
when presented with a wobbly rubber handrail—designed
to be unusable for postural support—infants nevertheless
find a variety of ways to exploit the handrail for support
(Berger, Adolph, & Lobo, 2005). They hunch over the rail
letting it dip to their knees, pull themselves along like a
mountain climber, lean back like a windsurfer, carefully
distribute weight across the whole arm, and so on. Similarly, 11-month-old “cruisers” exploit a low handrail—
designed to be too low to provide postural support—by
cruising on their knees (Berger, Chan, & Adolph, 2014).
12
6
0
0
C
8
Slope Angle (deg)
6°
D
24°
Figure 4.6. Gait modifications for walking down slopes in one
infant. (A) Scatterplots showing number of steps to walk down
slopes in the first test session and (B) in the tenth test session,
approximately one month later. Each circle represents the result of one trial. Correlation coefficients refer to a linear fit.
(C) Infant’s footprints while approaching and walking down a
shallow, easily traversable 6° slope (20% of the steepest slope
infant could navigate) at the eleventh test session. (D) Infant’s
footprints while navigating a steep, more challenging 24°
slope at the eleventh test session (80% of the steepest slope
infant could navigate). Each symbol represents one step. Solid vertical lines demarcate portions of the walkway. The leftmost portion represents the flat starting platform. The middle
section represents the slope. The rightmost portion represents
the flat landing platform. Adapted from Gill et al (2009). With
permission.
of their bodies and skills relative to the current status of
the environment, determine which actions are possible and
which are not, and exploit possibilities for action to suit their
goals (Adolph & Robinson, in press; Gibson & Pick, 2000).
Skilled adults make it look easy. But none of it is easy
for young children, and all of it requires learning and development. Infants, for example, require months of everyday
locomotor experience before they accurately distinguish
safe from risky slopes or a step from a cliff (Adolph, 1997;
Kretch & Adolph, 2013). With sufficient experience in
a variety of everyday situations, infants can adapt movements on the fly to novel and variable changes in their body
and environment—a platform shoe on one foot, slippery
slopes, or walking down slopes wearing Teflon-soled shoes
10
or a lead-weighted vest (Adolph & Avolio, 2000; Adolph,
Joh, & Eppler, 2010; Adolph, Karasik, & Tamis-LeMonda, 2010; Cole, Gill, Vereijken, & Adolph, in press).
Like modifying ongoing action, a variety of means—
different locomotor methods for tackling a slope, different
strategies for carrying an object, and so on—functions to promote behavioral flexibility. More ways to tackle a problem
increases the probability of finding a satisfactory solution.
New strategies typically arise in the course of coping with
the task (Adolph, 1997; McGraw, 1935). Infants sometimes
generate new strategies by testing various options (e.g.,
start onto an obstacle in one position and then change their
minds and try something else) and sometimes discover new
strategies serendipitously (e.g., attempt their typical method
but due to a combination of gravity and inertia, find themselves doing something else that is ultimately more useful).
Given a variety of options in their repertoire, which
strategy do children select? If the penalty for error is sufficiently salient (e.g., falling into a precipice), experienced
infants avoid strategies that prohibit success (e.g., walking over a cliff) and select strategies that minimize errors (avoidance) or allow them to achieve their goal (e.g.,
backing feet first over the edge of a drop-off). If multiple
strategies are possible (backing, scooting, sliding, etc.), infants display multiple strategies (Adolph, 1997; Kretch &
Adolph, 2013). Moreover, if the penalty is not salient from
infants’ perspective (e.g., being wedged into an impossibly
narrow opening, increased energetic cost of superfluous
movement), experienced infants frequently select strategies
that are ineffective or lead to errors, despite the availability of more effective strategies (Franchak & Adolph, 2012).
Raw materials for learning. Regardless of infants’ intent, continuance or increase in intraindividual variability
can serve an exploratory function (Corbetta, 2009; Goldfield, 1995; Harbourne & Stergiou, 2009; Thelen, 1996;
Vereijken, 2010). Variation provides the raw materials for
learning. Variable movements and a variety of strategies
generate information about the self, the environment, and
the relations between them. Infants must learn about their
bodies and skills—the various body parts, the space in
which body parts can move, and how movements can be
coordinated within and across body parts. Variable movements provide opportunities for learning about the self in
action, and this information can be exploited later for making movements more proficient or for tailoring actions to
local conditions. Variability in belly crawling, for example, gives infants an immediate boost in speed and amplitude of movements when they begin crawling on hands
and knees (Adolph et al., 1998). Variability in step length
and speed in novice walkers can be exploited by experienced walkers to curb forward momentum on slopes (Gill
et al., 2009). Similarly, infants must learn about the environment—where things are and what can be done with
them. Variability in natural locomotion provides infants
with visual access to more of the room, ensures hands-on
experience with a variety of surfaces, and promotes interactions with varied objects, surfaces, and people (Adolph
et al., 2012; Karasik, Tamis-LeMonda, & Adolph, 2011).
From the perspective of learning and development,
continuance or increase in intraindividual variability is imperative because a selection process requires
variants to act upon. When variation runs out, selection grinds to a halt. Put another way: Strategy selection
is not a one-time event. Currently ineffective strategies
do not creep away to die. They decrease in frequency
or lie dormant until needed once again (Siegler, 2006).
Development Entails Changes In The Structure Of Intraindividual Variability
The previous sections described change in intraindividual variability by quantifying change in the amount of
variability such as decrease in the length of the hand path
in infant reaching or in the area of the head path in infant
sitting. However, development can also entail changes in
the structure of variability. Describing change in the structure of variability requires researchers to understand how
some aspect of motor behavior changes moment to moment over real time—that is, the sequence of values in a
time series. This section describes two ways of considering developmental change in the structure of variability.
One way of considering the structure of variability concerns the relations among variables in a time series. Look
again at the jerky arm path of the infant reaching for a toy
in Figure 4.2B and the wobbly head trajectory of the infant
trying to sit in Figure 4.4A-C. Outside the laboratory, infants are not strapped into a chair to stabilize the torso while
reaching for objects; instead, they maintain their own sitting
posture while reaching. Likewise, outside the lab, infants do
not sit for long periods without turning their heads to look
around and moving their arms to manipulate objects. Typically, infants learn to reach during the same developmental
period when they are learning to sit. Wobbles in the sitting
posture affect infants’ hand path while reaching; reciprocally, moving their arms to reach or turning their heads to
look affects infants’ sitting posture. Arm, head, and torso
movements do not happen in isolation, so the development
of motor control is not apportioned limb by limb (Bertenthal
& von Hofsten, 1998; von Hofsten, 2003). Instead, movements of various body parts co-occur so that motor control
requires the coordination of movements in real time (Bernstein, 1967; Turvey, 2007). This time-locked interplay
between simultaneous movements in different parts of the
body is one way of considering the structure of variability.
A second way of considering the structure of variability
concerns dependencies in a time series. Look once more at
the jerky hand path in Figure 4.2B and the wobbly trace of
the head position in Figure 4.4A-C. Consider also the trail of
footprints created by the toddler approaching the brink of a
steep slope in Figure 4.6D. In each example, the jerks, wobbles, and footfalls are related in real time. A jerk of the hand
11
in one direction necessitates a correction in the opposite direction; if the baby over- or undershoots, additional corrections are needed. In other words, what happened milliseconds earlier affects the hand path a few moments later. The
wobbly sitter shows a similar time dependency. As the infant’s torso starts to fall forward, the body jerks backward in
an effort to compensate. Postural sway in the current moment
depends on what happened a moment prior, and influences
what has to happen next. Likewise, changes in the toddler’s
step length are organized in time and space, getting smaller
as the toddler approaches the brink of the slope. Researchers can investigate developmental changes in the structure
of variability by quantifying dependencies within a variable in a time series at different developmental time points.
Emerging Correlations Between Variables
Successful and adaptive movements require different
body parts to work together, coordination among joints in
a limb, and different aspects of movement to be controlled
simultaneously. Consequently, a primary challenge for skill
acquisition is learning to coordinate the body as a single unit
rather than a collection of separate parts. The different variables that are crucial for successful performance must become
increasingly correlated in time. A famous example of such
emerging correlations across variables is in pistol shooting
(Arutyunyan, Gurfinkel, & Mirskii, 1969). Expert marksmen
do not consistently hit the bullseye by reducing extraneous
body movements to zero. Instead, their shooting is reliable
because variations in, say, the shoulder angle, are instantly
corrected by compensatory movements in the wrist joint.
Similarly, motor development involves emerging correlations between variables in time. For example, in a challenging balance control task, 5-year-olds can stand on one
leg longer than 3-year-olds (Slobounov & Newell, 1994).
But how do they do it? Older children exhibit less postural
sway than younger ones, but concurrent with reduced sway,
older children also produce more movements and larger
movements in the torso and limbs. Five-year-olds elevate
and swing their arms, bend at the knee, lift their heel, and
hop in place. These body movements are compensatory
responses that are tightly related in time to postural sway.
The net result is a reduction in the amount of postural
sway. In contrast, younger children stiffen their torso and
decrease leg and arm movements, but to no avail. Without
a time-locked correlation between variable limb movements and postural sway, they sway more and lose balance.
Long before children can balance on one foot, learning
to stand on two feet involves emerging correlations between
variables over a time series. With a hand resting on a support
surface, pre-walking infants and novice walkers sway first
and then lean onto the support surface to assist their balance.
But after 6 weeks of walking experience, this relation reverses so that pressing on the support surface precedes body
sway. By establishing a time-locked relation between variability in forces applied to the support surface and variability
in postural sway, infants generate perceptual information to
aid prospective control of standing (Barela et al., 1999). The
consequence is a reduction in the amplitude of postural sway.
Even before infants can stand with support, they can
coordinate different aspects of motor behavior in a new
task—producing vertical jumps in a “baby bouncer” (a
seat held by a spring to the top of a doorway). At first infants dangle in the bouncer, unsure of what to do, spontaneously producing occasional, random leg movements.
Through these sporadic, variable kicks and jumps, infants
discover the temporal and spatial relations between their leg
movements and the resulting bounce. Soon, they learn to
maximize the amplitude of their bounces by changing the
structure of their movement variability. The average time
between bounces remains the same, but the variability of
inter-bounce intervals decreases as infants learn to synchronize jumps with the spatial trajectory of the bounce. If
infants jump at the wrong moment, the forces they generate will cancel out the extension of the spring. Moreover,
after a few sessions, infants begin to group their bounces
into extended bouts. These emerging correlations between
the timing and spatial trajectory of infants’ movements
accentuate the bounce by matching the stiffness and elasticity of the spring (Goldfield, Kay, & Warren, 1993).
Emerging Time Dependencies Within Variables
Structure in variability can also emerge in the form of
time dependencies within a variable, such as the infant’s
step length while approaching the slope in Figure 4.6D.
As with emerging correlations across variables, time dependency in one variable over a time series is acquired
gradually with age and experience. Walking provides an
intuitive example of the emergence of time dependency
over development. When someone walks a long distance,
the successive stride times (the time interval between
foot contacts of the same foot) create a time series. In the
first few years of independent walking, the stride time of
the current step influences the stride time of the steps that
immediately follow but not steps farther forward in time.
This time dependency is short and decays quickly such
that each subsequent step becomes increasingly independent from earlier steps. After several years of walking experience, the dependency between successive stride times
stretches farther forward over the time series, reflected in
increased long-range correlations between stride times
and in a slower rate of decay (Hausdorff et al., 1999).
As illustrated with stride time variability, the sequence
of values in a time series determines the structure of the
time series. Imagine three types of waveforms: a perfectly regular oscillating wave, a totally random series of jiggles, and the somewhat wiggly waveforms somewhere in
between. All three types of time series could share the same
amount of variability as measured by the range in values,
the standard deviation, or other summaries of the amount of
variability around a central point. But such summaries do
12
not take the sequence of values in the time series into account. The perfectly regular waveform is highly structured,
utterly predictable, and low in complexity. In contrast, the
totally random time series has no structure, is completely
unpredictable, and—because it is white noise—is undefined in terms of complexity. The somewhat wiggly waveforms that reside between these extremes contain more or
less structure in terms of predictability and complexity.
The development of predictability in the structure of intraindividual variability follows an inverted U-shaped function (Cignetti, Kyvelidou, Harbourne, & Stergiou, 2011;
Harbourne & Stergiou, 2003). The time series of infants’
first attempts at a new skill more closely resemble a random
waveform than a regular oscillating wave. Thus, early periods of development typically are marked by an increase
in dependencies within a time series accompanied by a decrease in the amount of intraindividual variability (based on
measures such as the standard deviation and area of postural
sway) such that the time series becomes more predictable and
complex while the movements become more consistent and
smooth. Indeed, the emerging structure in the variability of
the time series may underlie improvements in consistency.
However, with continued development, movements cannot
continually increase in predictability because then behavior
would become robotic and inflexible. Thus, at later periods
of skill acquisition, predictability and consistency typically
decrease while complexity is maintained, ensuring sufficient flexibility in the system for behavior to be adaptive.
The development of sitting exemplifies the inverted
U-shaped trajectory. The erratic postural sway of new sitters
gradually gives way to a more periodic and predictable sway
pattern as sitting becomes better controlled (Cignetti et al.,
2011; Harbourne & Stergiou, 2003). However, after infants
can sit without falling, the developmental trajectories of predictability and consistency begin to reverse. Postural sway
becomes less predictable again and the linear measures of
sway become less consistent because infants now have
sufficient control to turn their head and move their arms.
The complexity of sitting continues to increase throughout childhood (Hong, James & Newell 2008). Younger
children make disproportionally many small corrections,
presumably because they rely on reactive control to keep
balance. This results in more postural sway around the
upright position (a longer path traveled by the center of
pressure) accompanied by low complexity. With age and
experience, children become increasingly able to exploit
passive forces, resulting in less postural sway and more
complexity. In adults, sway patterns are even less periodic
and predictable, but more complex (Cignetti et al, 2011).
Potential Functions
Optimal variability for performance. Motor development is characterized by concurrent changes in the amount
and structure of variability as infants master new skills.
From infants’ initially erratic and wobbly movements,
different aspects of motor behavior gradually become related across variables in a time series and more time dependencies emerge. Some of these relations reduce intraindividual variability to make movements more predictable
and consistent, whereas other relations increase variability to allow for greater complexity, behavioral flexibility,
and adaptability. Eventually, children acquire the optimal
structure of variability to support skilled performance.
Although the structure of intraindividual variability can
change with development, values at the high or low ends
of the variability spectrum are not a good thing. Healthy
behavior reflects both stability and the capacity for change.
Excessive noise in the structure of intraindividual variability incurs costs of inconsistency and unpredictability. Insufficient variability incurs costs of overly rigid, restricted
behavior, inflexibility, and low complexity. Intraindividual
variability, thus, adheres to the “Goldilocks principle” of
a just-right amount of variability (Fetters, 2010). Children
need sufficient consistency to make behavior reliable, and
sufficient flexibility to deal with variations in local conditions and disruptions of the status quo (Stergiou & Decker,
2011; Stergiou, Yu, & Kyvelidou, 2013; Vereijken, 2010).
Recent work suggests that suboptimal levels of intraindividual variability—too little or too much—can distinguish
children with typical development from children with disabilities and delay. Infants diagnosed with cerebral palsy or
at risk for the disorder show more regular, periodic structure in postural sway in early stages of sitting compared
with typically developing infants (Deffeyes, Harbourne,
DeJong, et al., 2009; Deffeyes, Harbourne, Kyvelidou,
Stuberg, & Stergiou, 2009; Kyvelidou, Harbourne, Willett, & Stergiou, 2013). In contrast, children with Downs
Syndrome display more irregular, non-periodic structure
in leg movements while walking on a treadmill compared
with typically developing children (Buzzi & Ulrich, 2004).
Optimal variability for development. In addition to contributing to performance, just-right levels of intraindividual variability might also contribute to healthy learning
and development (Corbetta, 2009; Harbourne & Stergiou,
2009). The same structure of variability at an earlier period
of development is not necessarily optimal at a later period.
For example, low predictability early in skill development
can facilitate learning which aspects of action must be coordinated in time and how earlier movements affect later
ones. This information is critical at early stages of skill acquisition. How else could infants acquire the information
without exploring the possibilities for movement? Thus, too
much predictability in children with disabilities may hamper learning. In typically developing children, greater predictability at later points in development can provide information about how to succeed at a task. The eventual return
to lower levels of predictability now provides information
about how to cope with novel changes in local conditions.
13
Conclusions And Future Directions: Understanding Intraindividual Variability In Motor
Development
Every behavioral and physiological event, of course,
produces intraindividual variability. But motor development is unique in developmental science by adhering to
a deep and long-standing commitment to describing and
explaining developmental change in intraindividual variability (e.g., Deutsch & Newell, 2005; McGraw, 1935;
Thelen, 1996). Perhaps because motor behavior so naturally lends itself to representations of trajectories over
time and space and perhaps because movements vary so
plainly on both qualitative and quantitative dimensions,
intraindividual variability has always been a central focus of research on motor development. This long and rich
history encompasses very different sorts of research programs focused on very different aspects of motor behavior
expressed over very different temporal and spatial scales.
Across this broad research landscape, the data reveal
three distinct patterns of developmental change. Decrease
in intraindividual variability is typically associated with
increase in motor control. As intraindividual variability
decreases, children’s movements become more proficient
and economical and performance becomes more accurate
and successful. Continuance or increase in intraindividual
variability is accompanied by increase in adaptability and
behavioral flexibility. Modifications of ongoing movements and a variety of strategies make it more likely that
children can cope with new situations and achieve their
goal. Changes in the structure of intraindividual variability go hand-in-hand with changes in the temporal organization and coordination of movements. The sequence of
values in a time series must be sufficiently unpredictable
and complex to ensure the flexibility to cope with variations in local conditions, but not so much that the system cannot produce movements reliably and consistently.
Despite the breadth of developmental research on intraindividual variability, several important things are missing
from the community’s research program. One shortcoming
is the lack of integration across the various instantiations
of intraindividual variability and the methods for studying
variability. Although most researchers acknowledge the
range of meanings and methods, few researchers take an
integrative approach. Instead, research teams tend to focus on only one aspect of intraindividual variability, say,
improvements in consistency (Berthier & Keen, 2006) or
in adaptability (Adolph, 1997). Notable exceptions are
a recent handful of studies that report both changes in
consistency and in the structure of variability (e.g., Deffeyes, Harbourne, Kyvelidou, et al., 2009; Hong et al.,
2008), but we know of no studies that examine developmental changes in all three types. The field now has the
tools and conceptual frameworks for studying multiple
types of intraindividual variability within individual children in a single study. A more integrative approach would
enable deeper understanding of motor skill acquisition
and the mechanisms underlying developmental change.
A second shortcoming is researchers’ over-reliance
on artificial situations in laboratory tasks. We construct
over-simplified and unnatural situations to ensure experimental control. But strapping pre-sitters into a seat and presenting the same toy repeatedly at midline is not how infants
really learn to reach. We standardize test situations to allow
generalization across children at various periods of development. But forcing infants to step on treadmills or along a
straight, continuous path is not how babies really learn to locomote. Sitting for long periods on a force plate and standing for long periods with a hand on a support rail are not
representative of sitting and standing outside the laboratory.
Encouraging infants to descend slopes dozens of times in a
row does not reflect the way they really encounter variation
and novelty in the everyday environment. Ecological validity is not a criticism unique to research on intraindividual
variability, but it impedes progress by reifying the laboratory paradigms and conflating findings from limited situations
with real-world motor behavior. The risk is that we construct theories about developmental change in intraindividual variability that do not generalize to real world activities.
A third shortcoming is our lack of knowledge about how
different types of intraindividual variability promote and
hinder healthy development. Most research findings are consistent with the idea that some amount of some type(s) of intraindividual variability is conducive to the healthy learning
and development of basic motor skills. However, we have
scarce experimental evidence to support this notion—that
is, studies with children randomly assigned to treatment and
control groups that promote or eliminate a particular type
of intraindividual variability. Identifying differences in the
amount and/or structure of variability between children with
typical development and children with disabilities is a promising first step. But it is not wholly satisfying. In one study,
approximate entropy (a measure of structure) distinguishes
healthy and impaired postural sway (Deffeyes, Harbourne,
DeJong, et al., 2009), but in another study it does not (Deffeyes, Harbourne, Kyvelidou, et al., 2009); instead, the Lyapunov exponent differs between groups. And the same measures of approximate entropy and the Lyapunov exponent
can differ in different directions between different clinical
populations (Buzzi & Ulrich, 2004; Deffeyes, Harbourne,
Kyvelidou, et al., 2009). The next step is to construct theories that direct researchers’ attention to particular measures
of intraindividual variability and that facilitate discovery
about the sources and consequences of intraindividual variability. Without such progress, researchers cannot design
appropriate therapeutic interventions for children at risk.
Finally, researchers need to generate experimental evidence about the functions of intraindividual variability. We
focused our chapter on potential functions of developmental
changes in intraindividual variability. Dozens of developmental studies provide correlational evidence concerning
the real-time functions of variability (Berthier & Keen,
14
2006; Cole et al., in press). A long history of experimental
research exists for studying the effects of variable practice
on motor learning (Schmidt & Lee, 2011). What we need
now is experimental evidence concerning the functions
of developmental changes in intraindividual variability.
References
Adolph, K. E. (1997). Learning in the development of
infant locomotion. Monographs of the Society for
Research in Child Development, 62(3, Serial No.
251). doi: 10.2307/1166199
Adolph, K. E., & Avolio, A. M. (2000). Walking infants
adapt locomotion to changing body dimensions.
Journal of Experimental Psychology: Human
Perception and Performance, 26, 1148-1166. doi:
10.1037/0096-1523.26.3.1148
Adolph, K. E., Cole, W. G., Komati, M., Garciaguirre,
J. S., Badaly, D., Lingeman, J. M., . . . Sotsky,
R. B. (2012). How do you learn to walk? Thousands of steps and dozens of falls per day.
Psychological Science, 23, 1387-1394. doi:
10.1177/0956797612446346
Adolph, K. E., Joh, A. S., & Eppler, M. A. (2010). Infants’
perception of affordances of slopes under high
and low friction conditions. Journal of Experimental Psychology: Human Perception and Performance, 36, 797-811. doi: 10.1037/a0017450
Adolph, K. E., Karasik, L. B., & Tamis-LeMonda, C. S.
(2010). Using social information to guide action:
Infants’ locomotion over slippery slopes. Neural Networks, 23, 1033-1042. doi: 10.1016/j.
neunet.2010.08.012
Adolph, K. E., & Robinson, S. R. (2011). Sampling development. Journal of Cognition and Development,
12, 411-423. doi: 10.1080/15248372.2011.608190
Adolph, K. E., & Robinson, S. R. (in press). Motor development. In L. Liben & U. Muller (Eds.), Handbook of child psychology and developmental
science (7th ed., Vol. 2 Cognitive processes, pp.
XXX-XXX). New York: Wiley.
Adolph, K. E., Robinson, S. R., Young, J. W., & Gill-Alvarez, F. (2008). What is the shape of developmental
change? Psychological Review, 115, 527-543.
doi: 10.1037/0033-295X.115.3.527
Adolph, K. E., Vereijken, B., & Denny, M. A. (1998).
Learning to crawl. Child Development, 69, 12991312. doi: 10.1111/j.1467-8624.1998.tb06213.x
Adolph, K. E., Vereijken, B., & Shrout, P. E. (2003).
What changes in infant walking and why. Child
Development, 74, 474-497. doi: 10.1111/14678624.7402011
Ames, L. B. (1937). The sequential patterning of prone
progression in the human infant. Genetic Psychology Monographs, 19, 409-460.
Arutyunyan, G. H., Gurfinkel, V. S., & Mirskii, M. L.
(1969). Organization of movements on execution
by man of an exact postural task. Biofizika, 14,
1162-1167.
Badaly, D., & Adolph, K. E. (2008). Beyond the average:
Walking infants take steps longer than their leg
length. Infant Behavior and Development, 31,
554-558. doi: 10.1016/j.infbeh.2007.12.010
Barela, J. A., Jeka, J. J., & Clark, J. E. (1999). The use of
somatosensory information during the acquisition
of independent upright stance. Infant Behavior
and Development, 22, 87-102. doi: 10.1016/
S0163-6383(99)80007-1
Berger, S. E., & Adolph, K. E. (2003). Infants use handrails as tools in a locomotor task. Developmental
Psychology, 39, 594-605. doi: 10.1037/00121649.39.3.594
Berger, S. E., Adolph, K. E., & Kavookjian, A. E. (2010).
Bridging the gap: Solving spatial means-ends
relations in a locomotor task. Child Development, 81, 1367-1375. doi: 10.1111/j.14678624.2010.01478.x
Berger, S. E., Adolph, K. E., & Lobo, S. A. (2005). Out of
the toolbox: Toddlers differentiate wobbly and
wooden handrails. Child Development, 76, 12941307. doi: 10.1111/j.1467-8624.2005.00851.x
Berger, S. E., Chan, G., & Adolph, K. E. (2014). What
cruising infants understand about support for
locomotion. Infancy, 19, 117-137. doi: 10.1111/
infa.12042
Bernstein, N. A. (1967). The coordination and regulation
of movements. Oxford: Pergamon Press.
Bertenthal, B. I., & von Hofsten, C. (1998). Eye, head and
trunk control: The foundation for manual development. Neuroscience and Biobehavioral Review,
22, 515-520. doi: 10.1016/S0149-7634(97)000389
Berthier, N. E., & Keen, R. E. (2006). Development of
reaching in infancy. Experimental Brain Research,
169, 507-518. doi: 10.1007/s00221-005-0169-9
Bhat, A. N., & Galloway, J. C. (2006). Toy-oriented changes during early arm movements: Hand kinematics.
Infant Behavior and Development, 29, 358-372.
doi: 10.1016/j.infbeh.2006.01.005
Brackley, H. M., Stevenson, J. M., & Selinger, J. C.
(2009). Effect of backpack load placement on
posture and spinal curvature in prepubescent
children. Work, 32, 351-360. doi: 10.3233/WOR2009-0833
Burnside, L. H. (1927). Coordination in the locomotion
of infants. Genetic Psychology Monographs, 2,
279-372.
Butterworth, G., Verweij, E., & Hopkins, B. (1997). The
development of prehension in infants: Halverson revisited. British Journal of Developmental
Psychology, 15, 223-236. doi: 10.1111/j.2044835X.1997.tb00736.x
15
Buzzi, U. H., & Ulrich, B. D. (2004). Dynamic stability
of gait cycles as a function of speed and system
constraints. Motor Control, 8, 241-254.
Chang, C. L., Kubo, M., Buzzi, U., & Ulrich, B. (2006).
Early changes in muscle activation patterns of
toddlers during walking. Infant Behavior and
Development, 29, 175-188. doi: 10.1016/j.infbeh.2005.10.001
Chen, L. C., Metcalfe, J. S., Chang, T. Y., Jeka, J. J., &
Clark, J. E. (2008). The development of infant
upright posture: Sway less or sway differently?
Experimental Brain Research, 186(2), 293-303.
doi: 10.1007/s00221-007-1236-1
Chen, L. C., Metcalfe, J. S., Jeka, J. J., & Clark, J. E.
(2007). Two steps forward and one back: Learning to walk affects infants’ sitting posture. Infant
Behavior and Development, 30, 16-25. doi:
10.1016/j.infbeh.2006.07.005
Chen, Y. P., Keen, R., Rosander, K., & von Hofsten,
C. (2010). Movement planning reflects skill
level and age changes in toddlers. Child Development, 81, 1846-1858. doi: 10.1111/j.14678624.2010.01514.x
Cheron, G., Bouillot, E., Dan, B., Bengoetxea, A., Draye,
J. P., & Lacquaniti, F. (2001). Development
of a kinematic coordination pattern in toddler
locomotion: Planar covariation. Experimental
Brain Research, 137, 455-466. doi: 10.1007/
s002210000663
Cignetti, F., Kyvelidou, A., Harbourne, R. T., & Stergiou,
N. (2011). Anterior-posterior and medial-lateral
control of sway in infants during sitting acquisition does not become adult-like. Gait and Posture,
33, 88-92. doi: 10.1016/j.gaitpost.2010.10.002
Clark, J. E., & Phillips, S. J. (1993). A longitudinal study
of intralimb coordination in the first year of
independent walking: A dynamical systems analysis. Child Development, 64(4), 1143-1157. doi:
10.1111/j.1467-8624.1993.tb04192.x
Clark, J. E., Whitall, J., & Phillips, S. J. (1988). Human
interlimb coordination: The first 6 months of independent walking. Developmental Psychobiology,
21, 445-456. doi: 10.1002/dev.420210504
Claxton, L. J., Keen, R., & McCarty, M. E. (2003).
Evidence of motor planning in infant reaching
behavior. Psychological Science, 14, 354-356.
doi: 10.1111/1467-9280.24421
Clifton, R. K., Muir, D. W., Ashmead, D. H., & Clarkson,
M. G. (1993). Is visually guided reaching in early
infancy a myth? Child Development, 64, 10991110. doi: 10.1111/j.1467-8624.1993.tb04189.x
Cole, W. G., Gill, S. V., Vereijken, B., & Adolph, K. E. (in
press). Coping with asymmetry: How infants and
adults walk with one elongated leg. Infant Behavior and Development.
Comalli, D. M., Franchak, J. M., Char, A., & Adolph, K.
E. (2013). Ledge and wedge: Younger and older
adults’ perception of action possibilities. Experimental Brain Research, 228, 183-192. doi:
10.1007/s00221-013-3550-0
Corbetta, D. (2009). Invited commentary. Physical Therapy, 89, 282-284. doi: 10.2522/ptj.20080130.ic
Deffeyes, J. E., Harbourne, R. T., DeJong, S. L., Kyvelidou, A., Stuberg, W. A., & Stergiou, N. (2009).
Use of information entropy measures of sitting
postural sway to quantify developmental delay in
infants. Journal of NeuroEngineering and Rehabilitation, 6, 34-47. doi: 10.1186/1743-0003-6-34
Deffeyes, J. E., Harbourne, R. T., Kyvelidou, A., Stuberg,
W. A., & Stergiou, N. (2009). Nonlinear analysis
of sitting postural sway indicates developmental
delay in infants. Clinical Biomechanics, 24, 564570. doi: 10.1016/j.clinbiomech.2009.05.004
Deutsch, K. M., & Newell, K. M. (2005). Noise, variability, and the development of children’s perceptual-motor skills. Developmental Review, 25,
155-180. doi: 10.1016/j.dr.2004.09.001
Dominici, N., Ivanenko, Y. P., Cappellini, G., Zampagni,
M. L., & Lacquaniti, F. (2010). Kinematic strategies in newly walking toddlers stepping over different support surfaces. Journal of Neurophysiology, 103, 1673-1684. doi: 10.1152/jn.00945.2009
Fetters, L. (2010). Perspective on variability in the development of human action. Physical Therapy, 90,
1860-1867. doi: 10.2522/ptj.2010090
Franchak, J. M., & Adolph, K. E. (2012). What infants
know and what they do: Perceiving possibilities
for walking through openings. Developmental Psychology, 48, 1254-1261. doi: 10.1037/
a0027530
Freedland, R. L., & Bertenthal, B. I. (1994). Developmental changes in interlimb coordination: Transition
to hands-and-knees crawling. Psychological
Science, 5, 26-32. doi: 10.1111/j.1467-9280.1994.
tb00609.x
Garciaguirre, J. S., Adolph, K. E., & Shrout, P. E. (2007).
Baby carriage: Infants walking with loads. Child
Development, 78, 664-680. doi: 10.1111/j.14678624.2007.01020.x
Gesell, A. (1925). The mental growth of the pre-school
child. New York: Macmillan.
Gesell, A. (1946). The ontogenesis of infant behavior. In
L. Carmichael (Ed.), Manual of child psychology
(pp. 295-331). New York: John Wiley.
Gesell, A., & Ames, L. B. (1940). The ontogenetic organization of prone behavior in human infancy. The
Journal of Genetic Psychology, 56, 247-263. doi:
10.1080/08856559.1940.10534500
Gesell, A., & Thompson, H. (1929). Learning and growth
in identical infant twins: An experimental study
by the method of co-twin control. Genetic Psychology Monographs, 6, 11-124.
16
Gesell, A., & Thompson, H. (1938). The psychology of
early growth including norms of infant behavior
and a method of genetic analysis. New York:
Macmillan.
Gibson, E. J., & Pick, A. D. (2000). An ecological approach to perceptual learning and development.
New York: Oxford University Press.
Gill, S. V., Adolph, K. E., & Vereijken, B. (2009). Change
in action: How infants learn to walk down
slopes. Developmental Science, 12, 888-902. doi:
10.1111/j.1467-7687.2009.00828.x
Goldfield, E. C. (1995). Emergent forms. Oxford: Oxford
University Press.
Goldfield, E. C., Kay, B. A., & Warren, W. H. (1993). Infant bouncing: The assembly and tuning of action
systems. Child Development, 64, 1128-1142. doi:
10.1111/j.1467-8624.1993.tb04191.x
Hadders-Algra, M. (2011). Variation and variability: Key
words in human motor development. Physical Therapy, 90, 1823-1837. doi: 10.2522/
ptj.20100006
Hallemans, A., Aerts, P., Otten, B., De Deyn, P. P., & De
Clercq, D. (2004). Mechanical energy in toddler
gait: A trade-off between economy and stability?
The Journal of Experimental Biology, 207, 24172431. doi: 10.1242/jeb.01040
Halverson, H. M. (1931). An experimental study of prehension in infants by means of systematic cinema
records. Genetic Psychology Monographs, 10,
107-283.
Harbourne, R. T., & Stergiou, N. (2003). Nonlinear analysis of the development of sitting postural control.
Developmental Psychobiology, 42, 368-377. doi:
10.1002/dev.10110
Harbourne, R. T., & Stergiou, N. (2009). Movement
variability and the use of nonlinear tools: Principles to guide physical therapist practice. Physical
Therapy, 89, 267-279. doi: 10.2522/ptj.20080130
Hausdorff, J. M., Zemany, L., Peng, C. K., & Goldberger,
A. L. (1999). Maturation of gait dynamics: Strideto-stride variability and its temporal organization
in children. Journal of Applied Physiology, 86,
1040-1047.
Hong, S. L., James, E. G., & Newell, K. M. (2008).
Age-related complexity and coupling of children’s sitting posture. Developmental Psychobiology, 50, 502-510. doi: 10.1002/dev.20310
Ivanenko, Y. P., Dominici, N., Cappellini, G., Dan, B.,
Cheron, G., & Lacquaniti, F. (2004). Development of pendulum mechanism and kinematic
coordination from the first unsupported steps in
toddlers. The Journal of Experimental Biology,
207, 3797-3810. doi: 10.1242/jeb.01214
Ivanenko, Y. P., Dominici, N., Cappellini, G., & Lacquaniti, F. (2005). Kinematics in newly walking
toddlers does not depend upon postural stability.
Journal of Neurophysiology, 94, 754-763. doi:
10.1152/jn.00088.2004
Karasik, L. B., Adolph, K. E., Tamis-LeMonda, C. S., &
Zuckerman, A. (2012). Carry on: Spontaneous
object carrying in 13-month-old crawling and
walking infants. Developmental Psychology, 48,
389-397. doi: 10.1037/a0026040
Karasik, L. B., Tamis-LeMonda, C. S., & Adolph, K.
E. (2011). Transition from crawling to walking and infants’ actions with objects and people. Child Development, 82, 1199-1209. doi:
10.1111/j.1467-8624.2011.01595.x
Keen, R. (2011). The development of problem solving in
young children: A critical cognitive skill. Annual
Review of Psychology, 62, 1-21. doi: 10.1146/
annurev.psych.031809.130730
Keen, R., Lee, M. H., & Adolph, K. E. (2014). Planning an
action: A developmental progression in tool use.
Ecological Psychology.
Konczak, J., Borutta, M., Topka, H., & Dichgans, J.
(1995). The development of goal-directed reaching in infants: Hand trajectory formation and joint
torque control. Experimental Brain Research, 106,
156-168. doi: 10.1007/BF00241365
Konczak, J., & Dichgans, J. (1997). The development
toward stereotypic arm kinematics during
reaching in the first 3 years of life. Experimental Brain Research, 117, 346-354. doi: 10.1007/
s002210050228
Kretch, K. S., & Adolph, K. E. (2013). Cliff or step?
Posture-specific learning at the edge of a dropoff. Child Development, 84, 226-240. doi:
10.1111/j.1467-8624.2012.01842.x
Kyvelidou, A., Harbourne, R. T., Willett, W. L., & Stergiou, N. (2013). Sitting postural control in infants
with typical development, motor delay, or cerebral palsy. Pediatric Physical Therapy, 25, 46-51.
doi: 10.1097/PEP.0b013e318277f157
Lasko-McCarthey, P., Beuter, A., & Biden, E. (1990).
Kinematic variability and relationships characterizing the development of walking. Developental Psychobiology, 23, 809-837. doi: 10.1002/
dev.420230805
Lee, H. M., Bhat, A., Scholz, J. P., & Galloway, J. C.
(2008). Toy-oriented changes during early arm
movements IV: Shoulder-elbow coordination.
Infant Behavior and Development, 31, 447-469.
doi: 10.1016/j.infbeh.2007.12.018
Looper, J., Wu, J., Barroso, R. A., Ulrich, D., & Ulrich,
B. D. (2006). Changes in step variability of new
walkers with typical development and with down
syndrome. Journal of Motor Behavior, 38, 367372. doi: 10.3200/JMBR.38.5.367-372
Mangalindan, D. M., Schmuckler, M. A., & Li, S. A.
(2014). The impact of object carriage on independent locomotion. Infancy. doi: 10.1016/j.
17
infbeh.2013.12.008
McCollum, G., Holroyd, C., & Castelfranco, A. M.
(1995). Forms of early walking. Journal of
Theoretical Biology, 176, 373-390. doi: 10.1006/
jtbi.1995.0206
McGraw, M. B. (1935). Growth: A study of Johnny and
Jimmy. New York: Appleton-Century Co.
McGraw, M. B. (1941). Neural maturation as exemplified
in the reaching-prehensile behavior of the human
infant. Journal of Psychology, 11, 127-141. doi:
10.1080/00223980.1941.9917023
McGraw, M. B. (1945). The neuromuscular maturation of
the human infant. New York: Columbia University Press.
McGraw, M. B., & Breeze, K. W. (1941). Quantitative
studies in the development of erect locomotion.
Child Development, 12, 267-303.
Metcalfe, J. S., Chen, L. C., Chang, T. Y., McDowell, K.,
Jeka, J. J., & Clark, J. E. (2005). The temporal
organization of posture changes during the first
year of independent walking. Experimental Brain
Research, 161, 405-416. doi: 10.1007/s00221004-2082-z
Muller, J., Muller, S., Baur, H., & Mayer, F. (2013).
Intra-individual gait speed variability in healthy
children aged 1-15 years. Gait and Posture, 38,
631-636. doi: 10.1016/j.gaitpost.2013.02.011
Newell, K. M., Liu, Y. T., & Mayer-Kress, G. (2003). A
dynamical systems interpretation of epigenetic
landscapes for infant motor development. Infant
Behavior and Development, 26, 449-472. doi:
10.1016/j.infbeh.2003.08.003
Newell, K. M., Slobounov, S. M., Slobounova, E. S., &
Molenaar, P. C. M. (1997). Stochastic processes in
postural center-of-pressure profiles. Experimental Brain Research, 113, 158-164. doi: 10.1007/
BF02454152
Patrick, S. K., Noah, J. A., & Yang, J. F. (2009). Interlimb
coordination in human crawling reveals similarities in development and neural control with
quadrupeds. Journal of Neurophysiology, 101(2),
603-613. doi: 10.1152/jn.91125.2008
Patrick, S. K., Noah, J. A., & Yang, J. F. (2012). Developmental constraints of quadrupedal coordination
across crawling styles in human infants. Journal of Neurophysiology, 107, 3050-3061. doi:
10.1152/jn.00029.2012
Piaget, J. (1952). The origins of intelligence in children.
New York: International Universities Press.
Piek, J. P., & Carman, R. (1994). Developmental profiles of spontaneous movements in infants.
Early Human Development, 39, 109-126. doi:
10.1016/0378-3782(94)90160-0
Polk, J. D., Spencer-Smith, J., DiBerardino, L., Ellis, D.,
Downen, M., & Rosengren, K. S. (2008). Quantifying variability in phase portraits: Application to
gait ontogeny. Infant Behavior and Development,
31, 302-306. doi: 10.1016/j.infbeh.2007.10.005
Rachwani, J., Santamaria, V., Saavedra, S. L., Wood, S.,
Porter, F., & Woollacott, M. H. (2013). Segmental
trunk control acquisition and reaching in typically
developing infants. Experimental Brain Research,
228, 131-139. doi: 10.1007/s00221-013-3544-y
Riach, C. L., & Hayes, K. C. (1987). Maturation of
postural sway in young children. Developmental
Medicine and Child Neurology, 29, 650-658. doi:
10.1111/j.1469-8749.1987.tb08507.x
Rosenbaum, D. A., Chapman, K. M., Weigelt, M., Weiss,
D. J., & van der Wel, R. (2012). Cognition,
action, and object manipulation. Psychological
Bulletin, 138, 924-946. doi: 10.1037/a0027839
Saavedra, S. L., van Donkelaar, P., & Woollacott, M.
H. (2012). Learning about gravity: Segmental
assessment of upright control as infants develop
independent sitting. Journal of Neurophysiology,
108, 2215-2229. doi: 10.1152/jn.01193.2011
Schmidt, R. A., & Lee, T. D. (2011). Motor control and
learning: A behavioral emphasis (5 ed.). Champaign, IL: Human Kinetics.
Shirley, M. M. (1931). The first two years: A study of
twenty-five babies. Postural and locomotor development (Vol. 1). Minneapolis, MN: University of
Minnesota Press.
Shirley, M. M. (1933). Locomotor and visual-manual functions in the first two years. In C. Murchison (Ed.),
A handbook of child psychology (2nd ed., Vol. 1,
pp. 236-270). Worcester, MA: Clark University
Press.
Siegler, R. S. (2006). Microgenetic analyses of learning.
In D. Kuhn & R. S. Siegler (Eds.), Handbook of
child psychology. Vol. 2. Cognition, perception,
and language (6th ed., pp. 464-510). New York:
John Wiley & Sons.
Slobounov, S. M., & Newell, K. M. (1994). Dynamics of
posture in 3- and 5-year-old children as a function
of task constraints. Human Movement Science,
13, 861-875. doi: 10.1016/0167-9457(94)90022-1
Snapp-Childs, W., & Corbetta, D. (2009). Evidence of
early strategies in learning to walk. Infancy, 14,
101-116. doi: 10.1080/15250000802569835
Soska, K. C., & Adolph, K. E. (2014). Postural position constrains multimodal object exploration
in infants. Infancy, 19, 138-161. doi: 10.1111/
infa.12039
Soska, K. C., Adolph, K. E., & Johnson, S. P. (2010).
Systems in development: Motor skill acquisition
facilitates three-dimensional object completion.
Developmental Psychology, 46, 129-138. doi:
10.1037/a0014618
Stergiou, N., & Decker, L. M. (2011). Human movement
variability, nonlinear dynamics, and pathology: Is
there a connection? Human Movement Science,
18
30, 869-888. doi: 10.1016/j.humov.2011.06.002
Stergiou, N., Yu, Y., & Kyvelidou, A. (2013). A perspective
on human movement variability with applications in infancy motor development. Kinesiology
Review, 2, 93-102.
Thelen, E. (1979). Rhythmical stereotypies in normal human infants. Animal Behavior, 27, 699-715. doi:
10.1016/0003-3472(79)90006-X
Thelen, E. (1996). The improvising infant: Learning about
learning to move. In M. R. Merrens & G. G.
Brannigan (Eds.), The developmental psychologists: Research adventures across the lifespan (pp.
21-36). New York: McGraw-Hill.
Thelen, E., & Adolph, K. E. (1992). Arnold L. Gesell: The
paradox of nature and nurture. Developmental
Psychology, 28, 368-380. doi: 10.1037/00121649.28.3.368
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. doi:
10.1111/j.1467-8624.1993.tb04188.x
Thelen, E., Corbetta, D., & Spencer, J. P. (1996). Development of reaching during the first year: Role of
movement speed. Journal of Experimental Psychology: Human Perception and Performance, 22,
1059-1076. doi: 10.1037/0096-1523.22.5.1059
Turvey, M. T. (2007). Action and perception at the level of
synergies. Human Movement Science, 26, 657697. doi: 10.1016/j.humov.2007.04.002
Vereijken, B. (2010). The complexity of childhood development: Variability in perspective. Physical Therapy, 90, 1850-1859. doi: 10.2522/ptj.20100019
Vereijken, B., Pedersen, A. V., & Storksen, J. H. (2009).
Early independent walking: A longitudinal study
of load perturbation effects. Developmental Psychobiology, 51, 374-383. doi: 10.1002/dev.20377
von Hofsten, C. (1991). Structuring of early reaching movements: A longitudinal study. Journal of Motor Behavior, 23, 280-292. doi:
10.1080/00222895.1991.9942039
von Hofsten, C. (2003). On the development of perception
and action. In K. J. Connolly & J. Valsiner (Eds.),
Handbook of developmental psychology (pp. 114140). London: Sage.
Wilmut, K., Byrne, M., & Barnett, A. L. (2013). To throw
or to place: Does onward intention affect how a
child reaches for an object? Experimental Brain
Research, 226, 421-429. doi: 10.1007/s00221013-3453-0