Research Report

Early Intensive Postural and

Movement Training Advances Head
Control in Very Young Infants
Hui-Min Lee, James Cole Galloway
H-M. Lee, PT, PhD, was a graduate
student in the Department of
Background. Daily experiences are thought to play an important role in motor Physical Therapy, University of
development during infancy. There are limited studies on the effect of postural and Delaware, Newark, Delaware, at
movement experiences on head control. the time this study was con-
ducted. Address all correspon-
Objective. The purpose of this study was to quantify the effects of postural and dence to Dr Lee at: noralee0724@
yahoo.com.
movement experiences on head control through a comprehensive set of measure-
ments beginning when infants were 1 month old. J.C. Galloway, PT, PhD, Infant
Motor Behavior Laboratory, and
Biomechanics and Movement Sci-
Design. This was a prospective, longitudinal, 2-cohort study. ences Graduate Program, Depart-
ments of Physical Therapy and
Methods. Twenty-two full-term infants who were healthy were randomly Psychology, Center for Biomedical
assigned to either a training group or a control group. Infants were observed every Engineering Research, University
other week from 1 to 4 months of age. Head control was assessed using a standard- of Delaware.
ized developmental assessment tool, the Test of Infant Motor Performance (TIMP), as [Lee H-M, Galloway JC. Early
well as behavioral coding and kinematics of infants’ head postures and movements in intensive postural and movement
a supported sitting position. Caregivers performed at least 20 minutes of daily training advances head control in
postural and movement activities (training group), or social interaction (control very young infants. Phys Ther.
2012;92:935–947.]
group) for 4 weeks.
© 2012 American Physical Therapy
Results. The training group had higher TIMP scores on head control–related items Association
during the training period and after training stopped compared with the control Published Ahead of Print:
group. Starting from the during training phase, the training group infants had their March 30, 2012
heads in a vertical and midline position longer compared with the control group Accepted: March 23, 2012
Submitted: June 13, 2011
infants. After training stopped, the training group infants actively moved their heads
forward more often and for larger distances.

Limitations. The experiences outside daily training were not monitored, and the
results may be specific to the experimental setup for infants with typical
development.

Conclusions. Young infants are able to take advantage of postural and movement
experiences to rapidly advance their head control as early as 4 to 6 weeks of postnatal
life. Infant positioning, caregiver handling, and caregiver-infant interactions were
likely contributing factors. This database of comprehensive measures may be useful
in future trials focused on head control in infants with special needs.

Post a Rapid Response to

this article at:
ptjournal.apta.org

July 2012 Volume 92 Number 7 Physical Therapy f 935

Downloaded from http://ptjournal.apta.org/ by Emma Horton on July 19, 2012
Early Intensive Postural and Movement Training in Very Young Infants

T
he development of head con- tively weak and poorly controlled. experiences on head control, how-
trol begins during the prenatal Thus, newborns often are provided ever, are rare. The general purpose
period. Head movement is one external support when they are held. of the current study was to address
of the first fetal movements observed By 3 months old of age, infants typ- the need for a longitudinal investiga-
as early as 7 to 8 weeks of postmen- ically maintain their heads in an tion of the effect of postural and
strual age, with increasing move- upright, midline position when held movement experiences on head con-
ment varieties seen by the 11th upright. Clinically, this upright, mid- trol during early infancy.
week.1 Infants display dramatic line position has been considered an
changes in their head control during important basic level of head con- Head control is the first major motor
the first months of their postnatal trol,3,4 which continues to increase milestone on most assessments of
life. Not surprisingly, head control is throughout the first year and early skills.11–14 Head control impair-
critical for a range of early behaviors, beyond.5–9 ments often are cited as an early risk
including those involving vision, oro- factor for future developmental
motor skills, and the trunk and arms. The above description of the “natu- delays.15 For example, head control
For example, a dynamically stable ral” timing of the emergence of head on the Test of Infant Motor Perfor-
head may provide less mechanical control should not be taken as uni- mance (TIMP) in the first months of
perturbation to the trunk and arms versal and immutable. Cross-cultural life is a factor in predicting which
and may support the function of studies on the handling of young children will be diagnosed with
vision.2 Less perturbation and infants by non-Western cultures sug- cerebral palsy by 2 years of age.16
improved use of vision, in turn, may gest that the above descriptions In several studies, infants born pre-
allow for better body control, which, strongly reflect culture norms. One maturely or developing cerebral
in turn, assists infants in learning implication of these studies is that palsy at a later age displayed poorer
more complex behaviors. early head control is influenced by head control compared with
the degree to which infants are pro- infants with typical develop-
Conventional wisdom, based on vided with postural and movement ment.17–22 Delayed head control
studies that typically involve infants opportunities (reviewed in Adolph greatly influences many aspects of
raised in Western cultures, holds that et al10). As outlined below, systemic a child’s life.23–25
a newborn’s neck muscles are rela- investigations of the effect of early
Despite the importance of head con-
trol in early intervention,4 compre-
The Bottom Line hensive studies are limited and are
typically cross-sectional investiga-
tions of infants at term age.17–22,26
What do we already know about this topic? The assessment of head control usu-
ally was conducted as a part of
For infants with typical development, movement experiences play an
larger, more general neuromotor or
important role in motor development. For example, infants provided with
neurobehavioral evaluation.11,13,14
additional advanced postural or movement experiences reach earlier than Therefore, results often were com-
infants provided with additional social experiences. posed of general posture control or
What new information does this study offer?
Available With
This study provides the first longitudinal quantification of the effect of This Article at
postural and movement experiences on the development of head control. ptjournal.apta.org
Very young infants with typical development who were provided with
additional experiences rapidly displayed advanced head control. • eAppendix: Training Activities for
the Training Group (20 Minutes
If you’re a caregiver, what might these findings mean Total per Day)
for you? • eTable: Summary of the Effects
of Postural and Movement
The daily experiences that families provide to their infants during play or Training on Head Control
therapy may influence the emergence of even the earliest skills such as
• Demonstration Video of Infants
head control. Daily play involving a range of positions and movements in the Control Group and the
may influence the motor development of infants with special needs. Training Group During Chair Play

936 f Physical Therapy Volume 92 Number 7 July 2012

Downloaded from http://ptjournal.apta.org/ by Emma Horton on July 19, 2012
 Early Intensive Postural and Movement Training in Very Young Infants
posture response scores, with few
specifics on head control. Further-
more, results on specific head con-
trol impairments were not always
consistent, such as in the case of
impairments in pull-to-sit and sitting
tasks.17–22,26
The above review points out that
most studies on very young infants
determined the presence or absence
of milestones, which do not provide
information on the process by which
these infants develop impaired head
control or what factors lead to
impaired head control. Without a
detailed understanding of these fac-
tors, there is little evidence from
which to build early interventions.
This is the current situation with
respect to treating infants with head
control impairments. The current
study addressed the need for a more
comprehensive and longitudinal
investigation on the development of
head control from 1 to 4 months of
age with multilevel measures. As a
first step, this study focused on build-
ing a normative database using
infants with typical development.
If very young infants were receptive
to additional postural and movement
experiences, this receptiveness
would provide support for the study
of the effects of these experiences as
“training” for infants with poor head
control. In general, daily experience
is critical for the emergence of basic
motor skills in early childhood.7,27,28
Studies of infants who were develop-
ing typically have shown the effects
of postural and movement experi-
ences on a range of behaviors,
including stepping, kicking, reach-
ing, sitting, crawling, and postural
muscle responses.27,29 –37 These stud-
ies demonstrated positive effects of
training after as well as before a
behavior initially emerged.
Training after a behavior is displayed
can advance the quality of the prac-
ticed behavior.33,35,36 Although less
July 2012
Downloaded from http://ptjournal.apta.org/ by Emma Horton on July 19, 2012
studied, training before the onset of
a behavior can advance the emer-
gence and the quality of the prac-
ticed behavior.27,31,32,37 For example,
infants had earlier onset of crawling
after receiving 3 weeks of training.31
Six-week-old infants who received 7
weeks of sitting training displayed
longer upright sitting duration imme-
diately posttraining.37 We know of
no empirical investigation of the
training effects on head control in
typical development.
In pediatric intervention, postural
and movement training are generally
well-accepted strategies for promot-
ing behaviors for young infants.38,39
Few studies, however, have focused
on the effect of specific training for
head control in young infants, and
none were comprehensive and lon-
gitudinal. As mentioned previously,
cross-culture research on motor
development supports the use of
physical activities and active han-
dling as a potentially effective inter-
vention for head control impair-
ments.40 – 44 These studies suggest
that child-rearing practices through
daily formal training (eg, training to
sit or walk) or informal handling (eg,
how infants are carried) affect the
onset ages of many motor mile-
stones, including the development
of head control. The specific pur-
pose of the current study, therefore,
was to test the hypothesis that daily
postural and movement experiences
would significantly influence the
development of early head control.
Results from this study of infants
with typical development could be
useful for future intervention studies
of infants at risk for long-term impair-
ments in head control.
Method
Participants
Twenty-four full-term infants with no
known sensory or motor impair-
ments were recruited after their par-
ents signed informed parental con-
sent forms approved by the
Volume 92 Number 7
University of Delaware Human Sub-
jects Review Board. Due to schedul-
ing conflicts, 2 infants did not con-
tinue the study. Twenty-two infants
were randomly assigned to either a
training group (n⫽11, 6 male and 5
female) or a control group (n⫽11, 8
male and 3 female). Two infants
were Asian, 2 were African Ameri-
can, and the remaining infants were
Caucasian.
Procedures
Infants were seen for testing every
other week for 3 months, from 1 to 4
months of age. There was no signif-
icant difference in averaged age at
each testing session between groups
(P⫽.67–.95). There were a total of 9
testing sessions, including 2 home
testing sessions and 7 laboratory test-
ing sessions (see Table for the sum-
mary of procedures).
Home Testing Sessions
One experimenter visited infants at
their home when they were 1 and
1.5 months old. During the home
sessions, a standardized develop-
mental test (ie, TIMP) was con-
ducted. When infants were 2 to 4
months old, the TIMP was con-
ducted in the laboratory (see “Labo-
ratory Testing Sessions” section for
details), and there were no more
home testing sessions.
The TIMP is a sensitive, predictive,
and valid tool for detecting the
changes and differences in motor
performance in young infants.45,46 In
previous studies, the TIMP discrimi-
nated among infants with different
degrees of risk for poor motor
impairments at the time of initial
testing,46 – 48 as well as predicted
future motor impairments.16,49 –53
The TIMP scores also were sensitive
to detect changes in development in
infants with typical development
over a 2-week period.48 In our study,
we used specific TIMP items related
to head control to quantify the devel-
opment of head control in full-term
Physical Therapy f 937
Early Intensive Postural and Movement Training in Very Young Infants

Table.
Summary of Study Proceduresa

Variable Results

Age of infants (mo) 1 1.5 2 2.5 ... 4

Average age in each group (wk) CG: 4.47 CG: 4.57 CG: 8.78 CG: 10.71 CG: 16.73
TG: 4.46 TG: 4.48 TG: 8.69 TG: 10.70 TG: 16.69

Study week 1 1 3 3 5 7 ... 13

Visit number Home 1 Laboratory 1 Home 2 Laboratory 2 Laboratory 3 Laboratory 4 ... Laboratory 7

Location Home Laboratory Home Laboratory Laboratory Laboratory ... Laboratory

Weeks of training 0 (pretraining) 0 (pretraining) 2 2 4 Posttraining Posttraining

(training was
introduced right
after this visit)

Measures Chair play K K K K ... K

B B B B B

Developmental test TIMP TIMP TIMP TIMP ... TIMP

a
CG⫽control group, TG⫽training group, K⫽kinematics, B⫽behavioral coding of head postures and movements, TIMP⫽Test of Infant Motor Performance.

infants who were healthy (see the
“Behavioral Coding” section for
details). Most of these items have
been shown to be key characteristics
for identifying differences between
children with and without cerebral
palsy.16
During the second home testing ses-
sion, caregivers were asked to dem-
onstrate training activities intro-
duced at the end of the first
Figure 1.
The setting of chair play. The picture shows the TOY condition.
laboratory testing session. The
experimenter checked for proper
training and answered training-
related questions.
Laboratory Testing Sessions
Caregivers brought their infants to
the Infant Motor Behavior Labora-
tory, Department of Physical Ther-
apy, University of Delaware, every 2
weeks while the infants were 1 to 4
months of age. There were 2 com-
ponents: chair play and the TIMP.
When infants were 1 and 1.5 months
old, only the chair play was per-
formed in the laboratory. Starting
from the session at 2 months of age,
the TIMP was added to laboratory
procedures.
Chair play. Motion capture proce-
dures (see “Kinematic Data Acquisi-
tion” section for details) were con-
ducted during the chair play. Infants
were seated in a customized infant
chair reclined 30 degrees (Fig. 1)
with their trunk secured with a wide
cloth band. Two synchronized video
cameras recorded right front and
bird’s-eye views of the infants for
behavioral coding. Three experimen-
tal conditions were conducted:
no-toy arm-free (NTAF), no-toy arm-
held (NTAH), and toy (TOY). The
NTAF and TOY conditions were
used to test for differences in head
control with and without a midline
toy present. The NTAF and NTAH
conditions were used to test for dif-
ferences in head control during free
and restricted arm movements.
Experimental conditions were ran-
domly ordered across infants and
sessions.

938 f Physical Therapy Volume 92 Number 7 July 2012

Downloaded from http://ptjournal.apta.org/ by Emma Horton on July 19, 2012
 Early Intensive Postural and Movement Training in Very Young Infants
There were six 15-second trials for
each condition. For each trial, the
experimenter held the infant’s head
in a neutral (midline and upright)
position before the start of a trial and
then gently released the head right
after a trial start. The starting posi-
tion of infants’ arms was not
restricted for the NTAF and TOY
conditions.
TIMP. The TIMP was conducted
and videotaped to quantify the gen-
eral motor development with spe-
cific focus on the development of
head control.
Kinematic Data Acquisition
A 6-camera (120-Hz) Vicon motion
capture system (Pulnix TM6701-120
Progressive, Vicon Motion Systems
Inc, Los Angeles, California) was
used to obtain 3-dimensional (3D)
position-time data of the infants’
head movements. The cameras were
arranged in a circular fashion, with 2
cameras on each side and 2 in the
front and in the back of the infant.
Infants were seated within a cali-
brated volume of 160 cm ⫻ 160
cm ⫻ 200 cm. The average range of
calibration residuals was no larger
than 0.35 mm for the given cali-
brated volume. Arrays of three 8-mm-
diameter, retroreflective, nonlinear
markers were placed on the center
of the forehead. Two single markers
were placed on the toy in line. The
array positions were kept the same
between a static trial that defined the
reference position of the arrays and
the movement trials. The static trial
data were collected with the head
and trunk maintained in a neutral
(upright and midline) position.
Training Group
Infants in the training group
received postural and movement
activities provided by their caregiv-
ers for 20 minutes daily for 4 weeks
starting when they were 1 month of
age. The specific training activities
were selected based on our previous
July 2012
Downloaded from http://ptjournal.apta.org/ by Emma Horton on July 19, 2012
studies27,32,54 using various types of
training for young infants, including
both postural training, requiring
increased use of neck, shoulder
girdle, and trunk muscles, and move-
ment training, requiring increased
use of arm movements for reaching
(see eAppendix, available at
ptjournal.apta.org, for detailed infor-
mation on training activities). In our
previous studies, infants received the
2 types of training separately, either
postural or movement training, and
began at an older age (8 weeks of age
or older). To provide infants with
additional upright experience, care-
givers were instructed to carry their
infant for an additional 20 minutes
daily with a front carrier (BabyBjörn,
Bredaryd, Sweden) provided to fam-
ilies of the training group infants.
Control Group
To control for the increased social
interaction that infants in the train-
ing group experienced during train-
ing, infants in the control group
received structured, face-to-face
communication with their caregivers
for 20 minutes daily for 4 weeks.
Caregivers were asked to place their
infants in a supine position and inter-
act with their infants visually and ver-
bally without physically contacting
their infant or presenting toys.
Daily activity journals were provided
to caregivers in both groups to track
the amount of activities performed
each day. The percentage of training
completed for the daily 20-minute
activities was similar between
groups (training group: X⫽103.2 %,
SD⫽43.1%; control group:
X⫽102.1%, SD⫽10.3%; z⫽⫺0.413,
P⫽.68). On average, the use of the
front carrier was completed for more
than 85% of the required time in the
training group.
Data Analysis
Kinematics. Standard kinematic
analysis was conducted for head
movements during chair play. A stan-
Volume 92 Number 7
dard low-pass, recursive filter (But-
terworth filter) with a cutoff fre-
quency of 4 Hz reduced high-
frequency noise. The position of the
head in 3D space was computed by
the singular value decomposition
method.55 A loss of no more than 10
frames consecutively was interpo-
lated using cubic spline interpola-
tion. A customized MATLAB pro-
gram (The Mathworks Inc, Natick,
Massachusetts) was used to calculate
the head kinematics. Variables of
head kinematics were analyzed
based on the data concatenated
across trials in each condition.
The kinematic dependent variables
were: (1) maximum lateral dis-
placement (in millimeters), defined
as the maximum displacement from
the midline along the x axis; (2)
maximum anterior-posterior dis-
placement (in millimeters), defined
as the maximum displacement along
the y axis; (3) speed of head move-
ment (in millimeters per second),
defined as the average 3D speed of
head movement; and (4) variability
of head 3D speed, defined as the
coefficient of variation of head 3D
speed.
Behavioral coding. Videotapes
were coded for head postures and
movements during chair play and
the TIMP. The following behavioral
coding dependent variables for head
control behaviors were chosen
based on observations from a pilot
study showing that infants fre-
quently display these behaviors
when seated in the infant chair: (1)
percentage of time in lean, defined
as amount of time infants laterally tilt
their heads from the midline more
than 15 degrees; (2) percentage of
time upright, defined as amount of
time infants keep their heads in an
upright, midline position and faced
forward; (3) percentage of time in
turn, defined as amount of time
infants turn their heads to the right
or left over 15 degrees from the mid-
Physical Therapy f 939
Early Intensive Postural and Movement Training in Very Young Infants
line with their head in an upright,
midline position; (4) percentage of
time in pop-up, defined as amount of
time infants move their head forward
and away from the back of the infant
chair with their head in an upright,
midline position; and (5) percentage
of time with other postures and
movements, defined as amount of
time infants show head postures and
movements other than lean, upright,
turn, or pop-up. Each variable was
coded frame by frame (1 frame⫽1/
120 second) and normalized with
respect to time for group
comparisons.
The total score of the TIMP was the
summed score of all of the 42 items
of the TIMP. The summed score of
the head control–related items was
separated out from the total score to
quantify head control from 1 to 4
months of age. Selected head control
items reflected the infants’ ability to
hold their head in midline or an
upright position without additional
visual or auditory stimulation in dif-
ferent body orientations (ie, supine,
prone, and sitting). The following 10
items were selected: item 1 (head in
midline in supine); items 15, 16, 17,
and 18 (head control: supported sit-
ting, posterior neck muscle, anterior
neck muscle, lower from sitting);
item 21 (head in midline without
visual simulation [supine]); item 32
(pull to sit); item 36 (head lift in
prone position); and items 41 and 42
(lateral head turning to the right and
left).
Reliability of behavioral coding.
The primary experimenter (H-M.L.)
was not blinded to group assign-
ment. To promote coding score
reliability, 2 blinded coders were
first trained to code with interrater
reliability of greater than 90% for
variables measuring head postures
and movements using 2 randomly
selected infants in each group. Two
random sessions were selected for
testing the reliability of head pos-
940 f Physical Therapy Volume 92 Number 7
Downloaded from http://ptjournal.apta.org/ by Emma Horton on July 19, 2012
Total Scores of the TIMP
Group
CG
150 * TG
*
125 *
Raw Score
100
75
50
Baseline During After Training After Training
Training Phase 1 Phase 2
Phase
Figure 2.
The box plots for the total scores of the Test of Infant Motor Performance (TIMP)
between the control group (CG) and the training group (TG) in each phase. The median
for each group is represented by the single line in the box. The boxes represent the
distribution of data from the 2 middle quartiles. The error bars represent the maximum
and minimum values and the top and bottom quartiles of data distribution. Asterisk
signifies a group difference (P⬍.05) based on the Mann-Whitney U test.
tures and movements coding. Fol- Statistical Analysis
lowing training, each coder was The focus of the current study was to
assigned half of the infants to code determine the training effects on the
for head postures and movements. development of head control by test-
The infants each coder received ing for differences between infants
were randomly selected from the 2 in the training and control groups.
groups. Head control–related items Preliminary analysis did not show
in the TIMP were scored by a third consistent patterns of significant dif-
coder for all infants. The interrater ferences among the 3 experimental
reliability of TIMP items then was conditions across the 7 experimental
tested with the scoring of 2 ran- sessions. That is, certain conditions
domly selected infants coded by a were significantly different for cer-
fourth coder. The reliability tain ages, but there was no clear pat-
([amount of agreement/(amount of tern of results across conditions and
agreement ⫹ amount of disagree- sessions. Thus, these 3 conditions
ment)] ⫻ 100) for the coding of head were combined for the final analysis.
postures and movements was The 7 experimental sessions were
greater than 90%: upright (⬎90%), combined into 3 phases for the final
lean (⬎91%), turn (⬎91%), and analysis based on when sessions
pop-up (⬎95%). Interrater reliability, occurred relative to the training peri-
calculated as intraclass correlation od: the baseline phase (1-month-old
coefficient between 2 coders for the session), the during training phase
head control–related items in the (1.5- and 2-month-old sessions when
TIMP, was .943. the training was being performed),
July 2012
 Early Intensive Postural and Movement Training in Very Young Infants

and the after training phase. The TIMP Scores of Head Control–Related Items
after training phase was split into 2 Group
50
phases—after training phase 1 (2.5- CG

and 3-month-old sessions), and after TG

training phase 2 (3.5- and 4-month- *
*
old sessions)—to match the length *
40
of the during training phase. We
combined data from 2 sessions into

Raw Score
phases to decrease the number of
comparisons and ease interpretation 30
in this initial study of head control.
Data from each of the 2 sessions per
phase were averaged. Future studies
20
may build on our results to use non-
combined data.

As the data did not follow a normal 10

distribution and the nonkinematic
data were rank orders or propor- Baseline During After Training After Training
tions, a nonparametric analysis Training Phase 1 Phase 2
(Mann-Whitney U test) was used to Phase
analyze all of the variables for group Figure 3.
differences in each phase. The Mann- The box plots for the summed scores of head control–related items in the Test of Infant
Whitney U test was conducted using Motor Performance (TIMP) in each phase between the control group (CG) and the
SPSS software (SPSS Inc, Chicago, training group (TG). Asterisk signifies a group difference (P⬍.05) based on the Mann-
Illinois) with Pⱕ.05. The effect size Whitney U test.
(r) was calculated using the equa-
tion: Z/(N)1/2, in which Z is the z
score and N is the total number of
group infants in all phases (during 36, and 41: z⫽⫺3.330 to ⫺2.186,
participants.56
training phase: z⫽⫺3.715, P⫽.000, P⫽.001 to .029, r⫽⫺.71 to ⫺.47).
r⫽⫺.79; after training phase 1:
Results z⫽⫺3.909, P⫽.000, r⫽⫺.83; after Behavior Coding of Head
Total Scores of the TIMP
training phase 2: z⫽⫺3.367, Postures and Movements
Results of total scores of the TIMP
P⫽.001, r⫽⫺.72), except for the Results of behavior coding of head
showed that the training group had
1-month-old baseline session. postures and movements showed
higher scores in the during training
that starting from the training phase,
phase (z⫽⫺3.309, P⫽.001, r⫽⫺.71)
In the during training phase (age training group infants had their
and the after training phases (phase
1.5–2 months), the training group heads in a vertical and midline posi-
1: z⫽⫺3.514, P⫽.000, r⫽⫺.75;
infants had significantly higher tion longer compared with control
phase 2: z⫽⫺1.977, P⫽.048,
scores on all 10 individual head group infants. Training group infants
r⫽⫺.42) compared with the control
control–related items except item 1 continued to display advanced head
group (Fig. 2).
(z⫽⫺3.447 to ⫺2.009, P⫽.001 to control in the phases after training
.045, r⫽⫺.73 to ⫺.42) compared stopped. The percentage of the
TIMP Scores of Head Control–
with control group infants. In after types of head postures and move-
Related Items
training phase 1 (age 2.5–3 months), ments infants displayed when they
Results of the TIMP scores of head
the training group had significantly were seated in the infant chair is
control–related items showed that
higher scores in 6 of the 10 items shown in Figure 4A–D.
the training group had higher scores
(items 15–18, 32, and 42: z⫽⫺3.811
during the training period and after
to ⫺2.591, P⫽.000 to .010, r⫽⫺0.81 In the baseline phase (age 1 month,
training stopped compared with the
to ⫺.55). In after training phase 2 Fig. 4A), 3 main behaviors character-
control group. Figure 3 shows that
(age 3.5– 4 months), the training ized both training and control group
training group infants had higher
group had significantly higher scores infants’ head postures and move-
scores on head control–related items
in 5 of the 10 items (items 15, 18, 32, ments in our infant chair: lean,
in the TIMP compared with control

July 2012 Volume 92 Number 7 Physical Therapy f 941

Downloaded from http://ptjournal.apta.org/ by Emma Horton on July 19, 2012
Early Intensive Postural and Movement Training in Very Young Infants

A Percentage of Types of Head Postures and Movements B Percentage of Types of Head Postures and Movements
in the Baseline Phase in the During Training Phase
100 100
90 90
80 80 *
70 70
Percentage

Percentage
Lean
60 60
Upright
50 50 Pop-up
40 40 * Turn
Others
30 30
20 20
10 10
0 0
CG TG CG TG
Group Group

C Percentage of Types of Head Postures and Movements D Percentage of Types of Head Postures and Movements
After Training Phase 1 After Training Phase 2
100 100
90 * 90
80 80
70 70

Percentage
Percentage

60 60
50 50
40 40
30 30
*
20 20
*
10 10
0
*
0
CG TG CG TG
Group Group
Figure 4.
The percentage of types of head postures and movements between the control group (CG) and the training group (TG) in the
baseline phase (A), the during training phase (B), and after training phases 1 (C) and 2 (D) when infants were seated in an infant
chair. Asterisk signifies a group difference (P⬍.05) based on the Mann-Whitney U test.

upright, and turn. Infants in both
groups spent most of the time
(approximately 50%) leaning their
head in various directions. The per-
centage of time infants kept their
heads in upright and turn was the
second and third highest, respec-
tively. There was no significant
group difference in any of the coded
head postures and movements dur-
ing the baseline phase.
In the during training phase (age
1.5–2 months, Fig. 4B), infants in
both groups spent the majority of
the time keeping their heads in
upright, lean, and turn, in the order
of the highest to lowest percentages,
respectively. The training group
infants demonstrated less lean
(z⫽⫺3.016, P⫽.03, r⫽⫺.64) and
more upright (z⫽⫺2.043, P⫽.041,
r⫽⫺.44) behaviors compared with
the control group infants. There
were no significant differences
between groups for turn and pop-up
behaviors during the training phase.
In after training phase 1 (age 2.5–3
months, Fig. 4C), infants in both
groups spent the majority of the time
with their heads upright, but the
order of the percentage of time in
the other head postures and move-
ments varied between groups. The
training group infants demonstrated
less lean (z⫽⫺2.017, P⫽.044,
r⫽⫺.43) and more pop-up (z⫽⫺
2.753, P⫽ .006, r⫽⫺.59) of their
heads in the chair compared with
the control group infants. There was
no significant difference in upright
and turn behaviors between groups
during after training phase 1.
In after training phase 2 (age 3.5– 4
months, Fig. 4D), infants in both
groups demonstrated upright, pop-
up, and turn as the main behaviors
characterizing their head postures
and movements in an infant chair.
The training group infants had more
pop-up (z⫽⫺3.354, P⫽.001,
r⫽⫺.72) and less turn (z⫽⫺3.243,
P⫽.001, r⫽⫺.69) of their heads in
the chair compared with the control
group infants. (See a demonstration
video of infants in the control group
and the training group during chair
play, available at ptjournal.apta.org.)
There was no significant difference
in lean and upright behaviors
between groups during after training
phase 2.

942 f Physical Therapy Volume 92 Number 7 July 2012

Downloaded from http://ptjournal.apta.org/ by Emma Horton on July 19, 2012
 Early Intensive Postural and Movement Training in Very Young Infants

A Maximum Head Displacement Along B Maximum Head Displacement Along

the x Axis the y Axis
Group
250 300
CG
TG
* *
200 * 250

Displacement (mm)
Displacement (mm)

150
200

100
150

50
100

0 Baseline During After After

Training Training Training
Baseline During After After Phase 1 Phase 2
Training Training Training
Phase
Phase 1 Phase 2
Phase
C Average Speed of Head Movements
100

80
*

*
Speed (mm/s)

60

40

20

Baseline During After After

Training Training Training
Phase 1 Phase 2
Phase
Figure 5.
The box plots for head kinematics between the control group (CG) and the training (TG) in each phase: the maximum head
displacement along the x axis (A), the maximum head displacement along the y axis (B), and the average speed of head movements
(C). Asterisk signifies a group difference (P⬍.05) based on the Mann-Whitney U test.

Head Kinematics In the baseline phase (age 1 month), In the during training phase (age
Results showed measurable quantita- the training group infants had 1.5–2 months), there was no signifi-
tive differences in head control abil- smaller maximum head displace- cant difference in any of the mea-
ity between the training and control ment along the x axis (z⫽⫺2.084, sured head kinematics between
groups, especially during the phases P⫽.037, r⫽⫺.44, Fig. 5A) and groups. In after training phase 1 (age
after the training stopped. Head slower average speed of head move- 2.5–3 months), the training group
kinematics with significant differ- ments (z⫽⫺2.097, P⫽.036, r⫽⫺.45, infants had larger maximum head
ences between groups are shown in Fig. 5C) compared with the control displacement along the y axis
Figure 5A–C. group infants. (z⫽⫺2.121, P⫽.034, r⫽⫺.45,

July 2012 Volume 92 Number 7 Physical Therapy f 943

Downloaded from http://ptjournal.apta.org/ by Emma Horton on July 19, 2012
Early Intensive Postural and Movement Training in Very Young Infants
Fig. 5B) compared with the control
group infants. In after training phase
2 (age 3.5– 4 months), the training
group infants had larger maximum
head displacement along the y axis
(z⫽⫺2.626, P⫽.009, r⫽⫺.56,
Fig. 5B) and faster average speed of
head movements (z⫽⫺3.668,
P⫽.000, r⫽⫺.78, Fig. 5C) compared
with the control group infants.
There was no significant difference
between groups in the coefficient of
variation of head movement speed in
all phases.
Discussion
Training Advances Head Control
in Infants Developing Typically
Results from the multiple measures
suggest that enhanced postural and
movement experiences (ie, training)
advanced the development of head
control. The training group infants
showed more advanced head control
compared with the control group
infants during the 4-week training
period, as well as months after train-
ing. Moreover, advances displayed
by training group infants showed an
interesting additive pattern with
advanced behavioral measures in the
during training phase, then
advanced behavioral and spatial
kinematics (ie, maximum displace-
ment) in after training phase 1, fol-
lowed by advances in behavioral,
spatial, and temporal kinematics (eg,
average speed) in after training
phase 2. Findings from total TIMP
scores indicated that the training
group infants were more advanced
in their general motor development
compared with the control group
infants. A summary of the results of
training effects on head control is
provided in the eTable (available at
ptjournal.apta.org). The results of
each phase are further discussed in
the following sections.
Training Effects in the During
Training Phase
Training had an immediate and sig-
nificant effect on infants’ head con-
944 f Physical Therapy Volume 92 Number 7
Downloaded from http://ptjournal.apta.org/ by Emma Horton on July 19, 2012
trol. Two major findings from the
during training phase are: (1) the
speed with which training experi-
ences led to advances in head con-
trol and (2) the rapid changes across
multiple measures.
First, changes in head control began
within weeks of providing infants
with additional postural and move-
ment experiences. Our data showed
that within the first 2 weeks of daily
training, the training group infants
displayed significant differences
from the control group infants. Spe-
cifically, the training group infants
had higher scores for all head
control–related items except item 1,
which is an easy item on which most
infants in both groups achieved the
highest score during the baseline
phase.48 Second, training effects
were not only rapid but also were
seen across multiple measures. Spe-
cifically, training altered both the
quality and the quantity of infants’
performance on the TIMP, as well as
how they used their head during
seated play (Fig. 4C and D). Taken
together, the training group infants
scored higher on head control items
in prone, supine, and sitting posi-
tions and during activities that
required active participation.
These changes, combined with the
cross-cultural research on caregiver
handling,10,40 – 44 strongly suggest
that a young infant’s head and neck
are neither as muscularly weak nor
as poorly controlled as traditionally
thought. Moreover, young infants
are able to take advantage of postural
and movement experiences to
advance their head control as early
as 4 to 6 weeks of postnatal life, if
not much earlier. The clinical impli-
cations will be discussed in a later
section.
Training Effects in After Training
Phases 1 and 2
In the weeks and months after train-
ing stopped, the training group
infants continued to show greater
head control and more advanced
head behaviors. Interestingly, these
training effects differed from those
noted during the training period.
The after training phase results sug-
gest that training has a lasting, posi-
tive effect on the development of
head control. First, the training
group infants not only achieved a
basic level of head control but also
were able to perform basic behaviors
for longer durations or with larger
degrees of movement than the con-
trol group infants. For example, in
after training phase 2, the training
group infants had higher scores in 5
of the 10 head control–related items.
These particular items are of interest,
as a Rasch analysis suggested they
were more difficult items to per-
form.48 Second, the training group
infants continued the advanced head
control from the during training
phase and displayed advanced move-
ment strategies while engaging in
activities in the infant chair in the
weeks and months after training
stopped. For example, in after train-
ing phase 1, for the first time, the
training group infants moved their
heads off the back of the chair (eg,
pop-up) longer than the control
group infants (Fig. 4C). Interestingly,
in after training phase 2, although
the training group infants had more
time in pop-up behavior, the control
group infants continued to spend
more time in turn behavior (Fig. 4D).
Pop-up behavior may be an espe-
cially interesting measure to follow
in future studies, given that it clearly
shows purposeful intent and
requires significant effort. It is impor-
tant to note, however, that pop-up
behavior may require supported sit-
ting in a reclined seat.
Lastly, kinematic and behavioral
results during the after training
phases suggest that the training
group infants continued to gain
strength and advanced control com-
pared with the control group infants
July 2012
 Early Intensive Postural and Movement Training in Very Young Infants
in the months after training. For
instance, the training group infants
not only held their heads off the back
of the chair for longer durations but
also moved with greater excursions
(Fig. 5B) and moved their heads
faster than the control group infants
(Fig. 5C). During the baseline phase,
when infants’ head control was gen-
erally poor, the faster speed of head
movements likely resulted from
insufficient control to counteract
gravity or a lack of motivation to
maintain an upright position. In after
training phase 2, head control in
both groups was beyond the basic
level such that the training group
infants were likely purposefully pro-
ducing faster movement speeds
through greater muscle activation
levels, for example.57
How Early Postural and
Movement Experiences
Advanced Head Control
Results from the current study pro-
vide empirical evidence that the
emergence of head control can be
advanced through early intensive
postural and movement training. In
addition, these results join those of
other studies to strongly suggest that
postnatal experience is an important
factor in typical development of
head control.30,35 To better under-
stand how our training had rapid and
significant effects during the training
period and continued effects months
after training, 3 training compo-
nents—positioning, handling, and
infant-caregiver interaction—are dis-
cussed next.
Infant positioning. In the during
training phase, the training group
infants received positioning experi-
ences that were advanced for infants
during the first months of their life.
Typically, in Western cultures,
young infants spend much of their
waking hours in a supine posi-
tion.58,59 More than 30% of 4-month-
olds never experienced a prone posi-
tion while awake, and 75% spent less
July 2012
Downloaded from http://ptjournal.apta.org/ by Emma Horton on July 19, 2012
than 20 minutes daily in a prone
position.60 Based on the training
activities in the current study, the
training group infants spent an addi-
tional 6 minutes in a prone position
and 12 minutes in an upright posi-
tion, which are body positions typi-
cal of older, more mobile infants.
Prone and upright positions would
be expected to increase the muscle
performance and coordination of an
infant’s neck, shoulders, and upper
trunk; provide a new view of the
environment; and provide novel ves-
tibular stimulation, all of which have
been associated with increased alert-
ness and motor development.61– 68
Caregiver handling. In addition
to the positions that infants were
placed in for play activities, the activ-
ities themselves altered the typical
manner in which caregivers handled
their infants. As stated previously, in
Western cultures, caregivers tend to
hold young infants carefully, sup-
porting their heads for at least the
first months of their life (reviewed in
Adolph et al10). In the current study,
caregivers of the training group
infants were instructed to perform
activities that facilitated head control
and were encouraged to handle their
infants daily with less passive sup-
port of the head and in a much more
active manner overall. The gains in
head control in the current study fit
with the cross-cultural research
showing active handling advances
head control.
Infant-caregiver interaction.
Lastly, the training may have influ-
enced infant-caregiver interaction
beyond physical handling. During
daily training time, caregivers of the
training group infants likely encour-
aged their infants to actively control
their head appropriately. As infants
developed better head control abil-
ity, their caregivers likely continued
to offer additional opportunities to
facilitate advanced head control
even outside of the prescribed “train-
Volume 92 Number 7
ing time.” Future studies can now
build on these foundation data to
develop valid and reliable methods
to track the type and duration of
infant-caregiver interaction both dur-
ing and after the daily training
period.
Clinical Implications
One major principle of early inter-
vention is to provide activities to
advance behaviors in infants and
children at risk for a developmental
delay before the delay is directly
observed. Several recent reviews,
including a meta-analysis, however,
have not found strong effects of tra-
ditional early intervention strategies
on motor development during
infancy, preschool age, or school
age.69 Authors highlighted the need
for comprehensive, sensitive mea-
sures to quantify both the short-term
training period and longer-term
follow-up period for intervention
programs for functional abilities.38,70
As discussed above, we know of no
longitudinal study with sensitive
measurements on the effectiveness
of interventions focused on head
control. The dependent variables
and training program from the cur-
rent study were specifically chosen
to provide a framework for future
trials with pediatric populations at
risk for lifelong motor impairments.
The head control measures in the
current study are appropriate for lon-
gitudinally tracking the development
and effectiveness of intervention for
head control ability during early
infancy. These measures are multi-
level, which provides a comprehen-
sive view of infants’ performance of
head control, and can be conducted
on infants as young as 1 month old.
Results from the current study sug-
gest these measures successfully
detected the training effects on head
control from general behavioral
changes to specific kinematic
differences.
Physical Therapy f 945
Early Intensive Postural and Movement Training in Very Young Infants
Limitations
Given that this is the first compre-
hensive, longitudinal study of the
training effect on the emergence of
head control, there are several limi-
tations that need to be considered
when generalizing these results and
planning for future studies. First, the
experiences outside the daily train-
ing sessions were not monitored.
Caregivers likely altered the infants’
daily experiences outside of the for-
mal training time, which would be
expected to influence head control.
Such procedural validity will be
important in future studies inter-
ested in actual intervention dosage.
Second, only head behaviors were
assessed. The development of head
control involves not only the head,
but also the limbs and trunk, as well
as sensory-perceptual and cognitive
development. It will be necessary to
investigate the range of related sys-
tems to fully understand the devel-
opment of head control.
Third, data were analyzed and
reported as group effects. Individual
infants likely responded to the train-
ing differently. Identifying sub-
groups and individual patterns
would assist the detection of infants
with head control impairments who
often form subgroups based on the
distribution and degree of motor
involvement. Fourth, our primary
analysis focused on between-group
studies. Secondary analysis of within-
group comparisons is equally impor-
tant. Fifth, the results are specific to
the experimental setup. Typical
head control emerges within multi-
ple, dynamic environments. Investi-
gations that include different physi-
cal and social contexts, including a
range of cultures, are necessary to
fully understand head control. Sixth,
this study focused on infants with
typical development as a first step. In
terms of clinical implications, the
key effects of training will not be
known until after studying infants
with special needs. Future studies
946 f Physical Therapy Volume 92 Number 7
Downloaded from http://ptjournal.apta.org/ by Emma Horton on July 19, 2012
can build on these findings but will
need to tailor the methods and
hypotheses to the patient popula-
tion, age, and level of experience of
the cohort of these clinical trials.
Both authors provided concept/idea/re-
search design, writing, and project manage-
ment. Dr Lee provided data collection and
analysis. Dr Galloway provided fund pro-
curement, facilities/equipment, institutional
liaisons, and consultation (including review
of the manuscript before submission). The
authors thank all of the infants and their
parents for their effort and enthusiasm for
participation. They also thank many under-
graduate assistants for their help with data
collection and processing.
Preliminary data from this study were pre-
sented at PT 2009: Annual Conference and
Exposition of the American Physical Therapy
Association; June 10 –13, 2009; Baltimore,
Maryland. Parts of the manuscript were pre-
sented at the XVIIth Biennial International
Conference on Infant Studies; March 10 –14,
2010; Baltimore, Maryland; and the North
American Society for the Psychology of Sport
and Physical Activity (NASPSPA) Annual
Conference; June 10 –14, 2010; Tucson,
Arizona.
DOI: 10.2522/ptj.20110196
References
1 Prechtl H. Prenatal and early postnatal
development of human motor behavior.
In: Kalverboer A, Gramsbergen A, eds.
Handbook of Brain and Behaviour in
Human Development. Norwell, MA: Klu-
wer Academic Publishers; 2001:415– 428.
2 Bertenthal B, Von Hofsten C. Eye, head
and trunk control: the foundation for man-
ual development. Neurosci Biobehav Rev.
1998;22:515–520.
3 Bobath K. A Neurophysiological Basis for
the Treatment of Cerebral Palsy. 2nd ed.
London, England: William Heinemann
Medical Books Ltd; 1980.
4 Kramer JF, Ashton B, Bander R. Training of
head control in the sitting and semi prone
positions. Child Care Health Dev. 1992;
18:365–376.
5 Gassier J. A Guide to the Psycho-Motor
Development of the Child. New York, NY:
Churchill Livingstone Inc; 1984.
6 Gesell A, Thompson H. The Psychology of
Early Growth. New York, NY: The Mac-
millan Company; 1938.
7 McMraw MB. The Neuromuscular Matu-
ration of the Human Infant. London,
United Kingdom: Haffner Publishing;
1943.
8 World Health Organization, Multicentre
Growth Reference Study Group. WHO
Motor Development Study: windows of
achievement for six gross motor develop-
ment milestones. Acta Paediatr Suppl.
2006;450:86 –95.
9 Woollacott MH, Shumway-Cook A.
Changes in postural control across the life
span: a system approach. Phys Ther. 1990;
7:779 – 807.
10 Adolph KE, Karasik LB, Tamis-LeMonda
CS. Moving between cultures: cross-
cultural research on motor development.
In: Bornstein MH, ed. Handbook of Cross-
Cultural Development Science. Hillsdale,
NJ: Lawrence Erlbaum Associates; 2009:
61– 88. Domains of Development Across
Cultures; vol 1.
11 Bayley N. Bayley Scales of Infant and Tod-
dler Development. 3rd ed. San Antonio,
TX: Harcourt Assessment Inc; 2006.
12 Campbell SK, Girolami GL, Kolobe THA,
et al. The Test of Infant Motor Perfor-
mance. Chicago, IL: Infant Motor Perfor-
mance Scales LLC; 1993.
13 Folio MR, Fewell RR. Peabody Develop-
mental Motor Scales: Examiner’s Manu-
al. 2nd ed. Austin, TX: Pro-ED; 2000.
14 Piper MC, Darrah J. Motor Assessment of
the Developing Infant. Philadelphia, PA:
WB Saunders Co; 1994.
15 Scherzer AL, Tscharnuter I. Early Diagno-
sis and Therapy in Cerebral Palsy: A Pre-
mier on Infant Developmental Problems.
2nd ed. New York, NY: Marcel Dekker;
1990.
16 Barbosa VM, Campbell SK, Smith E, Ber-
baum M. Comparison of Test of Infant
Motor Performance (TIMP) item
responses among children with cerebral
palsy, developmental delay, and typical
development. Am J Occup Ther. 2005;59:
446 – 456.
17 Aylward GP, Hatcher RP, Leavitt LA, et al.
Factors affecting neurobehavioral
responses of preterm infants at term con-
ceptional age. Child Dev. 1984;55:1155–
1165.
18 Forslund M, Bjerre I. Neurological assess-
ment of preterm infants at term concep-
tional age in comparison with normal full-
term infants. Early Hum Dev. 1983;8:
195–208.
19 Jeng S, Tsou Yan K, Teng R. Neurobehav-
ioral development at term in very low
birthweight infants and normal term
infants in Taiwan. Early Hum Dev. 1998;
51:235–245.
20 Palmer PG, Dubowitz LMS, Verghote M,
Dubowitz V. Neurological and neurobe-
havioural differences between preterm
infants at term and full-term newborn
infants. Neuropediatrics. 1982;13:183–
189.
21 Mercuri E, Guzzetta A, Laroche S, et al.
Neurologic examination of preterm
infants at term age: comparison with term
infants. J Pediatr. 2003;142:647– 655.
22 Ricci D, Romeo DM, Haataja L, et al. Neu-
rological examination of preterm infants at
term equivalent age. Early Hum Dev.
2008;84:751–761.
July 2012
 Early Intensive Postural and Movement Training in Very Young Infants
23 Leiper CI, Miller A, Lang J, Herman R. Sen-
sory feedback for head control in cerebral
palsy. Phys Ther. 1981;61:512–518.
24 Mital MA, Belkin SC, Sullivan MA. An
approach to head, neck and trunk stabili-
zation and control in cerebral palsy by use
of the Milwaukee brace. Dev Med Child
Neurol. 1976;18:198 –203.
25 Wooldridge CP, Russell G. Head position
training with the cerebral palsied child: an
application of biofeedback techniques.
Arch Phys Med Rehabil. 1976;57:407–
414.
26 Cameron EC, Maehle V. Comparison of
active motor items in infants born preterm
and infants born full term. Pediatr Phys
Ther. 2006;18:197–203.
27 Lobo MA, Galloway JC, Savelsbergh
GJ. General and task-related experiences
affect early object interaction. Child Dev.
2004;75:1268 –1281.
28 Shumway-Cook A, Woollacott MH. Motor
Control: Translating Reach Into Clinical
Practice. 3rd ed. Philadelphia, PA: Lippin-
cott Williams & Wilkins; 2007.
29 Angulo-Kinzler RM, Ulrich B, Thelen E.
Three-month-old infants can select spe-
cific leg motor solutions. Motor Control.
2002;6:52– 68.
30 Hadders-Algra M, Brogren E, Forssberg H.
Training affects the development of pos-
tural adjustments in sitting infants.
J Physiol. 1996;493:289 –298.
31 Lagerspetz K, Nygard M, Strandvik C. The
effects of training in crawling on the
motor and mental development of infants.
Scand J Psychol. 1971;12:192–201.
32 Lobo MA, Galloway JC. Postural and
object-oriented experiences advance early
reaching, object exploration, and means-
end behavior. Child Dev. 2008;79:1869 –
1890.
33 Needham A, Barnett T, Peterman K. A
pick-me-up for infant’s exploratory skills:
early simulated experiences reaching for
objects using “sticky mittens” enhances
young infants’ object exploration skills.
Infant Behav Dev. 2002;25:279 –295.
34 Rovee CK, Rovee DT. Conjugate reinforce-
ment of infant exploratory behavior. J Exp
Child Psychol. 1969;8:33–39.
35 Sveistrup H, Woollacott MH. Practice mod-
ifies the developing automatic postural
response. Exp Brain Res. 1997;114:33–
43.
36 Thelen E. 3-month-old infants can learn
task-specific patterns of interlimb coordi-
nation. Psychol Sci. 1994;5:280 –285.
37 Zelazo NA, Zelazo PR, Cohen KM, Zelazo
PD. Specificity of experience effects on
elementary neuromotor patterns. Dev Psy-
chol. 1993;29:686 – 691.
38 Blauw-Hospers CH, Hadders-Algra M. A
systematic review of the effects of early
intervention on motor development. Dev
Med Child Neurol. 2005;47:421– 432.
39 Damiano DL. Rehabilitative therapies in
cerebral palsy: the good, the not as good,
and the possible. J Child Neurol. 2009;24:
1200 –1204.
July 2012
Downloaded from http://ptjournal.apta.org/ by Emma Horton on July 19, 2012
40 Hopkins B. Culturally determined patterns
of handling the human infant. Journal of
Human Movement Studies. 1976;2:1–27.
41 Hopkins B, Westra T. Maternal handling
and motor development: an intracultural
study. Genet Soc Gen Psychol Monogr.
1988;114:379 – 408.
42 Hopkins B, Westra T. Motor development,
maternal expectations, and the role of han-
dling. Infant Behav Dev. 1990;13:117–
122.
43 Rabain-Jamin J, Wornham WL. Practice
and representations of child care and
motor development among West Africans
in Paris. Early Development and Parent-
ing. 1993;2:107–119.
44 Super CM. Environmental effects on motor
development: the case of “African infant
precocity.” Dev Med Child Neurol. 1976;
18:561–567.
45 Campbell SK, Kolobe THA. Concurrent
validity of the Test of Infant Motor Perfor-
mance with the Alberta Infant Motor
Scale. Pediatr Phys Ther. 2000;12:1– 8.
46 Campbell SK, Kolobe THA, Osten ET, et al.
Construct validity of the Test of Infant
Motor Performance. Physi Ther. 1995;75:
585–596.
47 Campbell SK, Hedeker D. Validity of the
Test of Infant Motor Performance for dis-
criminating among infants with varying
risk for poor motor outcome. J Pediatr.
2001;139:546 –551.
48 Campbell SK, Levy P, Zawacki L, Liao P-J.
Population-based age standards for inter-
preting results on the Test of Infant Motor
Performance. Pediatr Phys Ther. 2006;18:
119 –125.
49 Campbell SK, Kolobe THA, Wright B, Lina-
cre JM. Validity of the Test of Infant Motor
Performance for prediction of 6-, 9-, and
12-month scores on the Alberta Infant
Motor Scale. Dev Med Child Neurol. 2002;
44:263–272.
50 Barbosa VM, Campbell SK, Sheftel D, et al.
Longitudinal performance of infants with
cerebral palsy on the Test of Infant Motor
Performance and on the Alberta Infant
Motor Scale. Phys Occup Ther Pediatr.
2003;23:7–29.
51 Flegel J, Kolobe THA. Predictive validity of
the Test of Infant Motor Performance as
measured by the Bruininks-Oseretsky Test
of Motor Proficiency at school age. Phys
Ther. 2002;82:762–771.
52 Kolobe THA, Bulanda M, Susman L. Pre-
dicting motor outcome at preschool age
for infants tested at 7, 30, 60, and 90 days
after term age using the Test of Infant
Motor Performance. Phys Ther. 2004;84:
1144 –1156.
53 Lekskulchai R, Cole J. Effect of a develop-
mental program on motor performance in
infants born preterm. Aust J Physiother.
2001;47:169 –176.
54 Heathcock J, Lobo MA, Galloway JC.
Movement training advances the emer-
gence of reaching in infants born less than
33 weeks gestational age. Phys Ther. 2008;
88:310 –322.
Volume 92 Number 7
55 Soderkvist I, Wedin PA. Determining
movements of the skeleton using well-
configured markers. J Biomech. 1993;26:
1473–1477.
56 Rosenthal R. Effect sizes: Pearson’s corre-
lation, its display via the BESD, and alter-
native indices. Am Psychol. 1991;46:
1086 –1087.
57 Thelen E, Corbetta D, Spencer J. Develop-
ment of reaching during the first year: role
of movement speed. J Exp Psychol Hum
Percept Perform. 1996;22:1059 –1076.
58 Mildred J, Beard K, Dallwitz A, Unwin J.
Play position is influenced by knowledge
of SIDS sleep position recommendations. J
Paediatr Child Health. 1995;31:499 –502.
59 Salls JS, Silverman LN, Gatty CM. The rela-
tionship of infant sleep and play position-
ing to motor milestone achievement. Am J
Occup Ther. 2002;56:577–580.
60 Majnemer A, Barr RG. Influence of supine
sleep positioning on early motor mile-
stone acquisition. Dev Med Child Neurol.
2005;47:370 –376.
61 Bly L. The Components of Normal Move-
ment During the First Year of Life and
Abnormal Motor Development. Chicago,
IL: Neuro-Developmental Treatment Asso-
ciation; 1983.
62 Clark D, Kreutzberg J, Chee F. Vestibular
stimulation influence on motor develop-
ment in infants. Science. 1977;196:1228 –
1229.
63 Neal M. Vestibular stimulation and devel-
opmental behavior of the small premature
infant. Nurs Res Rep. 1968;3–5.
64 Fredrickson WT, Brown JV. Posture as a
determinant of visual behavior in new-
borns. Child Dev. 1975;46:579 –582.
65 Gregg CL, Haffner ME, Korner AF. The
relative efficacy of vestibular-
proprioceptive stimulation and the
upright position in enhancing visual pur-
suit in neonates. Child Dev. 1976;47:309 –
314.
66 Pederson D, TerVrugt D. The influence of
amplitude and frequency of vestibular
stimulation on the activity of two-month-
old infants. Child Dev. 1973;44:122–128.
67 Pin T, Eldridge B, Galea MP. A review of
the effects of sleep position, play position,
and equipment use on motor develop-
ment in infants. Dev Med Child Neurol.
2007;49:858 – 867.
68 TerVrugt D, Pederson D. The effect of ver-
tical rocking frequencies on the arousal
level in two-month-old infants. Child Dev.
1973;44:205–209.
69 Orton J, Spittle A, Doyle L, et al. Do early
intervention programmes improve cogni-
tive and motor outcomes for preterm
infants after discharge? A systematic
review. Dev Med Child Neurol. 2009;51:
851– 859.
70 Ketelaar M, Vermeer A, Helders PJM. Func-
tional motor abilities of children with cere-
bral palsy: a systematic literature review of
assessment measures. Clin Rehabil. 1998;
12:369 –380.
Physical Therapy f 947
Copyright of Physical Therapy is the property of American Physical Therapy Association and its content may
not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.