ArticlePDF Available

How Infants Use Vision for Grasping Objects

Authors:

Abstract and Figures

The role of vision was examined as infants prepared to grasp horizontally and vertically oriented rods. Hand orientation was measured prior to contact to determine if infants differentially oriented their hands relative to the object's orientation. Infants reached for rods under different lighting conditions. Three experiments are reported in which (1) sight of the hand was removed (N=12), (2) sight of the object was removed near the end of the reach (N=40, including 10 adults), and (3) sight of the object was removed prior to reach onset (N=9). Infants differentially oriented their hand to a similar extent regardless of lighting condition and similar to control conditions in which they could see the rod and hand throughout the reach. In preparation for reaching, infants may use the current sight of the object's orientation, or the memory of it, to orient the hand for grasping; sight of the hand had no effect on hand orientation.
Content may be subject to copyright.
Child Development, July/August 2001, Volume 72, Number 4, Pages 973– 987
How Infants Use Vision for Grasping Objects
Michael E. McCarty, Rachel K. Clifton, Daniel H. Ashmead, Philip Lee, and Nathalie Goubet
The role of vision was examined as infants prepared to grasp horizontally and vertically oriented rods. Hand
orientation was measured prior to contact to determine if infants differentially oriented their hands relative to
the object’s orientation. Infants reached for rods under different lighting conditions. Three experiments are re-
ported in which (1) sight of the hand was removed (
N
12), (2) sight of the object was removed near the end of
the reach (
N
40, including 10 adults), and (3) sight of the object was removed prior to reach onset (
N
9). In-
fants differentially oriented their hand to a similar extent regardless of lighting condition and similar to control
conditions in which they could see the rod and hand throughout the reach. In preparation for reaching, infants
may use the current sight of the object’s orientation, or the memory of it, to orient the hand for grasping; sight
of the hand had no effect on hand orientation.
INTRODUCTION
Children learn to use vision in a prospective manner
to act effectively within their environment (von Hofsten,
1993). For example, if a child preorients their hand to
fit the orientation of an object before contact, then the
grasp is said to be under prospective control. Efficient
grasping requires preparing the hand, rather than
making adjustments after an object has been awk-
wardly contacted. The focus of this article is on how
and when infants use vision to prepare the hand for
grasping an object. The possibilities range from using
vision throughout the reach to match sight of the
hand with sight of the object, to remembering an ob-
ject’s aspects from a preview and then executing the
entire reach and grasp without further input about
either the object or the hand. In other words, infants
may use sight of the hand as well as the object during
the reach (visually monitoring hand and object), sight
of only the object during the reach (visually monitor-
ing the object), or they may detect the relevant infor-
mation when the object is first localized and before
the reach is initiated (no visual monitoring during the
reach). If infants can prepare their hand based on
memory of the object’s location and orientation, then
they have demonstrated a type of object representa-
tion with respect to the grasping action, which may be
called a motor representation (Jeannerod, 1997).
The role of vision during the reach can be sepa-
rated from its role during the grasp. Accurate reach-
ing entails executing a trajectory that moves the hand
from its initial location to contact with the object,
whereas grasping entails hand preparation that in-
cludes fine motor movements, such as preshaping the
hand to the target’s size and orientation. A series of
studies established that infants do not need to see
their hand to reach and contact a toy (Clifton, Perris,
& Bullinger, 1991; Clifton, Rochat, Litovsky, & Perris,
1991; Perris & Clifton, 1988). In these studies, 6-month-
old infants reached for sounding objects (toys that
made a rattle sound) in the dark. In a procedure used
by Clifton, Rochat, Robin, and Berthier (1994), the
separation of sight of the hand from sight of the target
was accomplished by manipulating lighting in three
conditions: a sounding toy was presented in a nor-
mally lit environment; the same sounding toy,
painted with glow-in-the-dark paint, was presented
in darkness; and a similar, unpainted sounding toy
was presented in darkness. The reach for the object
was completely unaffected by loss of sight of the
hand. Numerous comparisons of kinematic and be-
havioral measures found no difference between
reaching for the toy in the light and reaching for it in
the dark as a glowing object, including a deceleration
in hand speed as the target was approached. Sight of
the target, however, did affect the reach in some unex-
pected ways. While reaching for a sounding object in
complete darkness, infants reached faster, although
they still decelerated just before contact. Although in-
fants did not appear to be affected by losing sight of
their reaching hand, they did appear to be monitoring
the target and modulating the reach with respect to
that input.
In a study designed to require that infants monitor
the target throughout the reach, Robin, Berthier, and
Clifton (1996) presented 5- and 7-month-old infants
with a moving object in both light and dark conditions.
A toy, painted with glow-in-the-dark paint, moved back
and forth in front of the infant so that it was within
reaching range for about 1 s on each pass. Infants
made fewer attempts to obtain the glowing toy in the
© 2001 by the Society for Research in Child Development, Inc.
All rights reserved. 0009-3920/2001/7204-0003
974 Child Development
dark, but when they did reach, the kinematics and the
chance of successful capture were similar in the two
lighting conditions. To catch the moving object, in-
fants had to aim their hands well ahead of the toy’s lo-
cation when the reach was initiated: This they did, in
both light and dark conditions. These results rein-
forced the idea that continual monitoring of a moving
target may be coordinated with proprioception of the
arm and hand, so that a successful trajectory could be
estimated and executed without sight of the limb.
In the studies cited above, the infants’ task was
simply to make contact with the toy in the dark. If the
task had required finer hand adjustments, then con-
tinuous visual monitoring of the hand, the object, or
both might have been beneficial or even necessary. In
the previous studies a reach was considered success-
ful if the infant’s hand contacted the toy; hand adjust-
ments for grasping were not measured. In the three
experiments reported in this article, the questions
asked were (1) whether sight of the hand would affect
the finer motor control reflected in grasp preparation
(Experiment 1), and (2) whether sight of the object re-
moved during the reach (Experiment 2) or before the
reach was initiated (Experiment 3) would affect grasp
preparation. To explore these questions, hand orien-
tation was chosen as the index of grasp preparation.
When a rod is presented vertically or horizontally,
an efficient grasp can only be attained by preorienting
the hand to the rod’s orientation. At 7 to 9 months of
age, infants tested in the light are able to orient their
hand to reflect a rod’s orientation (Lockman, Ashmead,
& Bushnell, 1984; Morrongiello & Rocca, 1989; von Hof-
sten & Fazel-Zandy, 1984; Wentworth, Benson, &
Haith, 2000). Those authors concluded that infants
were able to use visual information to guide manual
action, but most authors did not specify
what
visual
information was used. Morrongiello and Rocca, how-
ever, were more explicit in stating that infants visu-
ally monitored their hand as well as the rod during
the reach. They stressed that even 5-month-old infants
appeared to be aware of “discrepant visual informa-
tion,” referring to a comparison of the hand’s orienta-
tion with the rod’s orientation. They found, however,
that only infants at 7 and 9 months of age were able to
use this information during the reach to adjust hand
orientation more precisely. Because the infants in
these studies were tested in the light and saw their
hand and the object throughout the reach, it is unclear
what information they used to adjust their hand.
What information is being attended to and coordi-
nated with manual action can be determined only by
manipulating visual input. In Experiment 1, hand ori-
entation in 7.5-month-old infants who reached for
rods in the light and glowing rods in the dark was
compared. If visual monitoring of the hand during
the reach is necessary, then hand orientation should
be disrupted in the dark. If hand orientation during
light and dark conditions is similar, this would imply
that proprioception of the hand is sufficiently coordi-
nated with sight of the rod to govern modulation of
the grasp.
EXPERIMENT 1
The effect of visual information of the hand during
transport was evaluated by comparing reaching for a
rod in the light and in darkness when only the glowing
rod could be seen. Hand orientation was measured by
a motion analysis system that provided more precise
and objective measurement than previous studies in
which videotapes were coded by observers who were
not naive to the orientation of the object.
Method
Participants
Twelve 7.5-month-old infants (6 females, 6 males;
M
224 days,
SD
8.9) provided data in all condi-
tions. An additional 11 infants were tested but their
data were eliminated because of experimenter error
(
n
2), the session was not completed (
n
1), or the
child did not reach or had missing motion analysis
system data in one or more of the conditions (
n
8).
Apparatus
The target object was a wooden rod (1 cm
30 cm),
painted with a nontoxic, yellow glow-in-the-dark
paint. The rod was attached at its midpoint to a 1 m
handle that was held by an experiementer. The in-
fant’s hand orientation was recorded using an Op-
trak motion analysis system (Northern Digital, Inc.,
Waterloo, Canada), which detects the precise three-
dimensional position of small, infrared-emitting
markers. Three markers were taped across the back of
the infant’s left hand: Marker 1 was proximal to and
between the third and fourth digits, Marker 2 was be-
tween the second and third digits, and Marker 3 was
between the first and second digits. The position of
each marker was sampled at 100 Hz by three fixed-
position sensors and a computer. The sensors were lo-
cated above and to the left of the infant. An infrared
video camera also recorded the scene, and was time
locked to the motion analysis system by a date-timer
(For-A). The signal that triggered the motion analysis
system to collect data also started the date-timer which
ran until data collection ended for that trial. The date-
timer was superimposed on the video image.
McCarty et al. 975
Procedure
Infants were seated on a parent’s lap. At the start
of each trial, the rod was brought within the infant’s
reach in either a vertical or horizontal orientation
and the motion analysis system was triggered. The
rod remained in position for the duration of the
trial (either 10 or 15 s), then it was withdrawn until
the next presentation. The interval between trials
was approximately 15 s. In the
light
condition, the room
was lit with an incandescent bulb; in the
glowing
condition, the room was dark and only the glow-
in-the-dark rod was visible. Sixteen trials were
randomly presented to each infant in one of four
different random orders. Four vertical and four hor-
izontal rod orientations were presented in each light-
ing condition.
Data Coding and Measures
Videotapes were scored to determine time of reach
onset and time of contact with the rod. Reach onset
was defined as the first frame in which the infant’s
hand started to move toward the rod and continued on
to contact it. Contact was marked on the first frame in
which any part of the rod was touched. von Hofsten
and Ronnqvist (1988) estimated that it takes more than
60 ms (or approximately two video frames) to use haptic
information to adjust the hand. Thus, contact was de-
fined as occurring before haptic adjustments could be
made. To determine interrater reliability, two coders
scored the videotapes of five infants. The coders agreed
(to within 100 ms) on 86% of the reach-onset times and
91% of the contact times. The remaining videotapes
were scored by one coder.
Hand orientation was determined for each trial
from reach onset to contact, and was defined by the
angle formed by two of the hand markers projected
onto the frontal plane with respect to the horizontal
plane. For example, hand orientation was 0
when the
hand was placed palm down parallel to the floor, and
90
when the hand was vertically aligned with thumb
up. Three estimates of hand orientation were com-
puted. The two outer markers (Markers 1 and 3) were
used as the standard in calculating hand orientation.
The other two estimates included the middle marker
(one used Markers 1 and 2, the other used Markers 2
and 3), and hand orientation values were adjusted by
the mean difference between these angles and that
formed by Markers 1 and 3. This was done because
the back of the hand is not flat, which meant that
Marker 2 was not aligned with the other markers and
this affected the computed hand orientation. For ex-
ample, when Markers 1 and 3 formed a 45
angle, the
unadjusted angle formed by Markers 1 and 2 might
have been 55
, and the unadjusted angle formed by
Markers 2 and 3 might have been 35
. When the
angles were adjusted, each would be close to 45
. A
linear spline was implemented on missing sections of
hand orientation data of up to 20 motion analysis
frames (200 ms). The hand orientation estimate with
the smallest percentage of missing data was used for
each trial; when there were no missing data for any
estimate, Markers 1 and 3 were used.
Results
Each of the 12 infants was presented with 16 trials,
for a total of 192 trial presentations. Thirty-two trials
were excluded because infants did not reach for and
contact the rod, and 65 trials were excluded because
hand orientation data were missing. This left 95 trials
for the hand orientation analyses, including 22 light
horizontal trials, 29 light vertical trials, 15 glowing hor-
izontal trials, and 29 glowing vertical trials, with each
child still contributing at least 1 trial for each condition.
The difficulties in collecting and analyzing infant
data using optoelectric systems is one factor in the rel-
atively high level of participant and trial attrition (see
Thelen et al., 1993; and von Hofsten & Ronnqvist,
1988 for a discussion). This problem is even more
acute when hand orientation is measured because
two hand markers must be visible to obtain the mea-
surement (see “Data Coding and Measures” in the
Experiment 1 Method section). A second factor affect-
ing attrition rates is that each infant has to reach in all
four conditions to make within-subjects comparisons.
Given these factors, it is believed that the attrition
rates obtained were typical for research of this type
and that they did not influence the conclusions in this
article about hand orientation.
Hand orientation was reported at five points dur-
ing the reach for each trial as a percentage of total trial
duration: at onset; at 25%, 50%, and 75% of the reach;
and at contact. Mean hand orientation values were
then computed in each condition for each child. If the
child provided more than one trial within a condition,
a mean score was determined for each time point in
that condition. If there was only one usable trial, those
data were used for the condition. Mean hand orienta-
tion values for all infants from reach onset until con-
tact are presented in Figure 1. It should be noted that
infant hand orientation never closely matched the ob-
ject’s orientation, either here or in previous studies
(von Hofsten & Fazel-Zandy, 1984). The data were an-
alyzed with a 2 (lighting condition)
2 (object orien-
tation)
5 (point-in-the-reach) ANOVA.
An effect for object orientation,
F
(1, 11)
10.24,
p
976 Child Development
.008, indicated that hand orientation was different
for horizontal (
M
32
,
SD
27
) and vertical rods
(
M
47
,
SD
20
). There was also a significant ef-
fect for point-in-the-reach,
F
(4, 44)
4.44,
p
.004,
but there were no other statistically significant ef-
fects for lighting condition, any of the two-way in-
teractions, or the three-way interaction. These data
were further explored by testing when, during the
reach, infants started to preorient their hand to
match the object’s orientation as a function of light-
ing condition. These analyses allowed for the ex-
plicit testing of the questions that this experiment
was designed to address. Hence, a 2 (lighting condi-
tion)
2 (object orientation) ANOVA was con-
ducted at each point in the reach.
There were no significant differences in hand ori-
entation at the onset of the reach or at the 25% point
in the reach. When the infants were at the 50% point in
the reach, they had already begun to orient their hand
to the rod’s orientation,
F
(1, 11)
12.33,
p
.005. This
trend continued at 75% through the reach,
F
(1, 11)
5.73,
p
.036. At the 75% point in the reach, there was
also an effect for lighting condition,
F
(1, 11)
7.95,
p
.017, and a Lighting Condition
Orientation in-
teraction,
F
(1, 11)
8.41,
p
.014. In the vertical ori-
entation, infants were more preoriented in the light
condition than in the glowing condition at this junc-
ture of the reach. At contact, hand orientation contin-
ued to be oriented appropriately for horizontal and
vertical rods,
F
(1, 11)
8.56,
p
.014. There was no ef-
fect for lighting condition and no interaction between
lighting condition and orientation, indicating that
hand orientation at contact was similar regardless of
whether infants could see their hand during the reach.
Discussion
To summarize, infants’ hand orientation differed
during reaches for horizontal and vertical rods prior
to contact when the reach occurred in the light and
in the dark for a glowing rod. These 7.5-month-old
infants began to preorient their hands approxi-
mately halfway through the reach, and maintained
the appropriate angle at contact. It did not matter
whether the reach occurred in the light or in the
dark for a glowing rod; the infants preoriented their
hands to a comparable extent in both conditions. It
was concluded that infants did not have to visually
monitor their hand to adapt it to the object. This
conclusion differs from that of other researchers
who found that seeing the hand during the reach is
important for anticipatory hand orientation (Mor-
rongiello & Rocca, 1989; von Hofsten & Fazel-
Zandy, 1984).
EXPERIMENT 2
The added demand of preorienting the hand to match
an object’s orientation did not require sight of the
hand during the reach. In this, the results of Experi-
ment 1 confirm the data from several previous studies
showing that infants are not affected by loss of sight
of their hand when reaching for objects (Clifton et al.,
1994; Robin et al., 1996). An issue still unresolved is
how the loss of sight of the object affects reaching. In
Clifton et al. (1994) reaching for a sounding, invisible
object in the dark elicited faster reaches, with shorter
durations and fewer movement units, than reaching
for the toy in the light or glowing in the dark. Move-
ment units are small accelerations and decelerations
in hand speed during a reach that are thought to re-
flect corrective movements to the trajectory (Berthier,
1996). McCarty and Ashmead (1999) interrupted
sight of the target during the reach: infants were
tested at 5, 7, and 9 months as they reached for an ob-
ject that remained fully visible throughout the reach,
or was darkened while the reach was in progress. As
did Clifton et al., (1994), McCarty and Ashmead
found that reaches for an invisible object in the dark
had shorter durations and fewer movement units
than reaches for a visible target. In 5-month-old in-
fants, the effect of seeing a target was exaggerated;
they slowed their hand dramatically from approxi-
Figure 1 Mean hand orientation values for infants in Exper-
iment 1 in both lighting conditions and for both orientations
at five points during the reach from onset to contact. SEs range
from 4.1 to 10.6. Specific values are available from the authors.
McCarty et al. 977
mately 600 ms to approximately 1,200 ms. Both Clifton
et al. (1994) and McCarty and Ashmead (1999) con-
cluded that although infants did not have to see the
target or their hand to complete a successful reach, if
sight of the target was available, infants took more
time to process this information and they made more
corrective movements.
The target objects in McCarty and Ashmead (1999)
were hollow rods that were lit electrically and could
be turned on and off at any point in the reach. These
rods were presented in vertical and horizontal orien-
tations in light, in darkness with the rod glowing con-
stantly (similar to the conditions of Experiment 1 in
this study), and in darkness with the rod lit at the start
of the reach but darkened once the reach was in
progress. Kinematic information on the reaching hand
is available from McCarty and Ashmead’s article;
only hand orientation data are reported in the present
study. The issue is whether continual visual monitor-
ing of the target throughout the reach affects the finer
motor movements involved in matching hand-to-target
orientation before contact, as well as the subsequent
grasping action.
Method
Participants
Participating infants were from three age groups:
5-month-olds (5 females, 3 males; age:
M
153 days,
SD
6.9); 7-month-olds (5 females, 4 males; age:
M
211 days,
SD
7.0); and 9-month-olds (7 females, 6
males; age:
M
273 days,
SD
7.4). An additional 7,
8, and 3 infants participated from each age group, re-
spectively, but were excluded because they did not
provide usable data in all conditions. Ten young adults
(6 females, 4 males) from the university community
also participated.
Apparatus
The object to be grasped was either a pink or orange,
hollow, plastic rod (2 cm
21 cm) lit from within. The
rod could be darkened either manually by the experi-
menter or when the participant’s hand broke one of
two infrared light beams during the approach to the
object. The infrared light switches were located 10 cm
from the rod in the path that the participant’s hand was
expected to take while reaching for the rod. At the be-
ginning of each trial the rod was placed in a holder that
held the ends of the rod, so that two miniature, 12-volt
light bulbs embedded in each holder could light up the
rod. Participants could easily pull the rod out of the
holder after grasping it. A magnet was embedded in
the center of each rod and a magnetic switch was bro-
ken when a rod was pulled out of the holder.
Participants’ reaching was recorded with an infrared
video camera and an Optotrak motion analysis sys-
tem, both located on the right side of the participant.
Two markers were taped on the right hand (one prox-
imal and between the first and second digits, and the
second proximal and between the third and fourth
digits). Sampling for each trial lasted 12 s, at a sam-
pling rate of 50 Hz. (Sampling rate was reduced from
100 Hz in Experiment 1 to 50 Hz in Experiments 2 and 3.
The different sampling rates did not affect the mea-
sures being reported because only a “snapshot” of the
participant’s hand orientation at five or fewer points
in time during a reach was required.)
Procedure
Infant participants were brought to the laboratory
by a parent and were familiarized with the experi-
mental setting. Infants sat on a parent’s lap, and the
chair height was adjusted so that the infant’s shoul-
der was at or just above the center of the rod. Prior to
each trial, lighting condition and rod orientation
were set, and an experimenter pressed a button to
start data collection and then said “Go.” This cued
the parent to adjust the infant to be within reaching
range. (Infants were kept out of range between trials
so that the infrared beams would not be prematurely
broken.)
Trials were presented randomly in blocks of eight,
which consisted of a horizontal and a vertical trial in
each of four lighting conditions.
In the
light
condition, the room was illuminated
during the trial with the 15-W bulb. In each of the
other conditions, the room was dark during the trial.
In the
glowing
condition, the rod was illuminated
during the trial from the light bulbs at each end of
the rod. In the
darkened-during-reach
condition, the
rod was visible at the start of the trial and at the start
of the reach. The rod darkened during the reach (10
cm from the rod) when a participant’s hand broke
one of the beams of the infrared light switch. In the
darkened manually
condition, the original procedure
had been to darken the rod after it had been local-
ized but before the reach was initiated. This was dif-
ficult to accomplish because infants typically did
not reach for the rod if it darkened before reach on-
set, or infants started to reach before the rod could
be darkened. Thus, the rod was typically darkened
after reach onset in this condition. The room was lit
between trials with a 15-W bulb. Blocks of trials
continued to be presented for as long as the infant
cooperated.
978 Child Development
Data Coding and Measures
Trials in which the rod was picked up, so that the
magnetic switch was broken, were identified from
the Optotrak records. Only those trials were included
in the hand orientation analyses because the interest
was in measuring hand orientation during the reach
and when the rod was picked up. If the rod was not
picked up, it was because there was no reach, the rod
was not contacted, or the rod was contacted but not
picked up during the 12-s trial. A coder viewed a graph
of the motion analysis system data from each grasp
trial on a computer screen. The graph included the
x
-,
y
-, and
z
-axis displacement of the hand; the unsigned
speed and acceleration profiles of the hand; and the
time at which the magnetic switch was broken. Reach
onset was marked at the minimum speed prior to the
movement during which a participant’s hand started
to move toward the rod in the foreaft dimension.
Hand orientation was then computed automatically
throughout the reach and grasp. Pickup occurred
when the magnetic switch was broken, which was the
moment when the rod was removed from the holder.
Other points during the reach were defined with respect
to the distance of the hand markers from the pickup lo-
cation in the fore– aft dimension. Midreach was the
point at which the hand markers first moved to within
10 cm of the pickup location in the fore–aft dimension.
The rod was darkened by this point in the darkened-
during-reach condition. Contact was defined as the
point at which the hand markers first moved to
within 3 cm of the pickup location in the fore– aft di-
mension. This value was determined by scoring con-
tact on videotape for several 9-month-olds, and then
computing the foreaft distance between the hand
markers at contact and at pickup. The furthest dis-
tance between the hand markers at contact and
pickup was 3 cm, which meant that infants did not
contact the rod prior to this point.
Hand orientation was determined at each frame
from reach onset to pickup. It was defined in the same
way as in Experiment 1, with hand orientation equal
to 0
when the hand was parallel to the floor in the
horizontal orientation with the palm down, and 90
when the hand was orthogonal to the floor in the ver-
tical orientation with the thumb up. The hand orien-
tation value was missing for a frame if one of the hand
markers was obscured from the Optotrak system. A
linear spline was computed on missing hand orienta-
tion segments that were less than 200 ms (10 frames).
Adults
The procedures for the 10 adult participants were
similar to those for infants with the following excep-
tions: The beam from the infrared light switch was lo-
cated 17 cm from the rod rather than 10 cm because
adults have longer arms than infants. Thus, midreach
occurred at the point at which the hand moved to
within 17 cm of grasp location in the fore– aft axis.
Likewise, contact was defined as occurring at the
point at which the hand first moved within 6 cm of
the rod in the foreaft axis. This distance was deter-
mined by scoring contact from video for several
adults, and finding the furthest distance from contact
to pickup in the foreaft dimension. Each adult
reached for 40 trials. The hand orientation analyses
were virtually identical for the adults as for the in-
fants. One minor difference was that reach onset was
determined automatically as the first frame in which
hand acceleration was greater than 150 mm/s
2
and
also remained greater than that for 160 ms (eight frames).
The second part of this definition eliminated small
movements of the hand that may have occurred prior
to the actual onset of the reach.
Results
Because of the difficulties in getting infants to ini-
tiate a reach in complete darkness, the number of
lighting conditions was reduced to two. The light and
glowing conditions were combined to form an object-
visible condition on the basis of preliminary analyses
that found no differences between the light and glow-
ing conditions. In addition, the results from Experi-
ment 1 showed that infants were not using sight of
their hand to preorient it to a visible rod. For infants,
the two darkened conditions were combined to form
an object-darkened condition, in which the rod dark-
ened when the hand was at least 10 cm from the rod
in the fore aft orientation. (The videotapes were
viewed to ensure that any trial in which the rod dark-
ened after the infrared switches were broken was
discarded.) For adults, the two darkened conditions
were also combined to form an object-darkened
condition, in which the rod darkened either before
or during the reach. Table 1 shows the total number
of trials collected and analyzed for each age group
in each lighting condition. Note that each partici-
pant contributed at least one trial with usable hand
orientation data to each Lighting Condition
Object
Orientation cell.
Mean Hand Orientation
For each trial, hand orientation was determined at
four points during the reach: at reach onset, midreach
(at which point the rod was darkened in the object-
darkened condition), contact, and when the rod was
McCarty et al. 979
picked up. Mean hand orientation values were com-
puted for each participant in each Lighting Condition
Object Orientation cell. The age-group means are pre-
sented in Figure 2. A 2 (lighting condition)
2 (orien-
tation)
4 (point-in-the-reach) ANOVA was con-
ducted for each age group. Follow-up comparisons of
significant within-subjects effects were made using
paired-sample
t
tests, adjusting for the number of
comparisons made with Dunn’s test and
.05
(Howell, 1987, p. 590). The critical point in the reach
for hand orientation differences was at contact, before
adjustments could be made from haptic information.
Preorientation at this point in the reach required that
infants anticipated the rod’s orientation and adjusted
their hands accordingly.
Five-month-olds.
There was a significant interac-
tion between object orientation and point-in-the-reach,
F
(3, 21)
12.66, p .001, but no other significant dif-
ferences for lighting condition, object orientation,
point-in-reach, or any other interactions. To follow up
on the significant interaction, lighting condition was
collapsed and hand orientation was compared for
horizontal and vertical trials at all four points in the
reach. Not until the rod was picked up did hand ori-
entation differ as a function of object orientation, t(7)
5.49, p .001. These results demonstrated that 5-month-
old infants could rotate their hands to either the hori-
zontal or vertical orientation after contacting an object.
However, the means were in the correct direction at
contact, and in absolute values these means were sim-
ilar to those reported in earlier studies (Lockman et al.,
1984; von Hofsten & Fazel-Zandy, 1984). Statistically,
the results were in agreement with those of Lockman
et al., who did not find a significant difference in hand
orientation at grasp among 5-month-olds, but were
not in agreement with the results of von Hofsten and
Fazel-Zandy, who did find a significant difference in
hand orientation at grasp among 5-month-olds. There
was a concern that this negative result may have been
a Type II error caused by low power from using data
Figure 2 Mean hand orientation values for infants in Exper-
iment 2 for both lighting conditions and for both orientations
at four points during the reach and grasp. SEs range from 3.9 to
14.1 for 5-month-olds, from 2.2 to 11.8 for 7-month-olds, from
2.1 to 7.6 for 9-month-olds, and from 1.7 to 4.8 for adults. Spe-
cific values are available from the authors.
Table 1 Number of Trials Collected, Grasp Trials, and Grasp
Trials in Which Hand Orientation Was Analyzed in Experiment 2
Age Group and
Lighting Condition
Number of Trials
Collected Grasped Analyzed
5 Months (n 8)
Object darkened 130 60 34
Object visible 130 94 57
7 Months (n 9)
Object darkened 151 99 68
Object visible 153 132 99
9 Months (n 13)
Object darkened 237 191 149
Object visible 238 209 158
Adults (n 10)
Object darkened 200 200 197
Object visible 200 200 197
980 Child Development
from only 8 infants in the analysis. The data were re-
analyzed and each trial was considered independent
in order to include data from all of the trials in this
analysis. A total of 91 trials were analyzed, and hand
orientation at contact with the horizontal rod (M 53,
SD 31) was not statistically different from that with
the vertical rod (M 58, SD 34), F(1, 89) 1. In
summary, 5-month-old infants were not affected by
losing sight of the object during the reach. They did
not preorient their hand before contact with the rod,
but did adjust after contact when completing the grasp.
Seven-month-olds. Hand orientation was found to
be different between horizontal and vertical rods,
F(1, 8) 29.00, p .001, with no effect found for light-
ing condition. As with the 5-month-olds, there was a
significant Object Orientation Point-in-the-Reach
interaction, F(3, 24) 15.86, p .001. To follow up on
the significant interaction, lighting condition was col-
lapsed and hand orientation was compared for hori-
zontal and vertical trials at all four points in the reach.
Hand orientation differed as a function of object ori-
entation at contact, t(8) 3.90, p .005, and at pickup,
t(8) 5.42, p .001. Thus, 7-month-olds preoriented
their hand, both with and without sight of the object.
There was a timing difference between these results
and the findings from Experiment 1, in which 7.5-
month-olds had begun to differentially orient their
hand midway through the reach. One explanation
may be that children from Experiment 1 were 2
weeks older than the children in Experiment 2; an-
other possibility is that the midpoint of the reach
was defined differently in the two experiments. The
midpoint in Experiment 2 was set at 10 cm from the
object for all infants, whereas the midpoint in Exper-
iment 1 was defined as half-way through each reach.
Nevertheless, both experiments showed that infants
in the seventh month do begin to preorient their
hand to match an object they are reaching for prior
to contacting it.
Nine-month-olds. Hand orientation was found to
be different between horizontal and vertical rods,
F(1, 12) 609.02, p .001, with no effect found for
lighting condition. Again, there was a significant
Object Orientation Point-in-the-Reach interaction,
F(3, 36) 101.64, p .001. To follow up on the sig-
nificant interaction, lighting condition was col-
lapsed and hand orientation was compared for hor-
izontal and vertical trials at all four points in the
reach. Hand orientation differed as a function of ob-
ject orientation at the midpoint, t(12) 7.66; at contact,
t(12) 18.95; and at pickup, t(12) 31.12, all ps
.001. Figure 2 illustrates that preshaping of the hand
was present at midreach and continued to increase
until grasp for the 9-month-old infants.
Adults. The findings for the adults were similar to
those for the 9-month-old infants: hand orientation
differed with rod orientation, F(1, 9) 238.27, p
.001, and with point in the reach, F(3, 27) 6.36, p
.002, with the relation evidenced through a signifi-
cant Object Orientation Point-in-the-Reach inter-
action, F(3, 27) 157.40, p .001. To follow up on
the significant interaction, lighting condition was
collapsed and hand orientation was compared for
horizontal and vertical trials at all four points in the
reach. Hand orientation differed as a function of ob-
ject orientation at the midpoint, t(9) 11.17; at con-
tact, t(9) 17.73; and at pickup, t(9) 14.20, all ps
.001. There were no main effects for lighting condi-
tion nor any interactions involving lighting condi-
tion during the reach or grasp.
In summary, there was a steady progression in the
development of anticipatory hand orientation— from
5-month-olds not preorienting their hand, to 7-
month-olds beginning to preorient the hand to an ob-
ject by contact, to 9-month-olds starting to orient their
hand early in the reach, to adults being almost com-
pletely oriented to the target by contact time. These
age trends in hand orientation are presented in Figure
2. There were no effects at any age for hand orienta-
tion when the object was darkened during the reach.
It was concluded that infants and adults incorporated
object-orientation information into their reach-and-
grasp plan either when an object was initially local-
ized or early in the reach, and that they did not have
to visually monitor the object throughout the reach to
preorient their hand.
Hand Orientation on Individual Trials
When data are averaged and only a few points are
presented as in Figures 1 and 2, a clear sense of what
the data from an individual trial looks like is not ob-
tained. For example, the data in Figure 2 depict mean
hand orientation as remaining relatively stable,
slowly separating as infants approach horizontally
and vertically oriented rods, and separating progres-
sively earlier in the reach in older infants.
To explore whether this pattern held during in-
dividual trials, graphs of individual trials were ex-
amined. Hand orientation graphs from one indi-
vidual in each age group are presented in Figure 3.
Each graph shows data from all of the trials for one
participant in one lighting condition and in one object
orientation. Each line represents hand orientation
from reach onset (or 2 s before contact for reaches
longer than 2 s) until contact on a single trial. The
shape of the data are considered representative of
other individuals from the same age group.
McCarty et al. 981
Although the general age trend for hand orienta-
tion to differ with object orientation is apparent in
these individual trial data, what is most striking is
variability between trials and within trials for the in-
fants. When an infant begins to reach, the adjust-
ment in hand orientation is not smooth. The hand
wobbles back and forth on its way to the rod. This
variability is probably due to lack of motor control
over the arm and hand, rather than to a change in
goal during the reach. The goal is to adjust the hand
to the rod, and by 7 months of age infants are fairly
consistent in achieving this to some extent. Because
control over their arm and hand increases with age,
infants learn to match the orientation of the rod
more closely and with less variability, although
there is considerable improvement in stability be-
tween 9 months and adulthood.
Discussion
The developmental sequence that emerged from
this experiment is that 7- and 9-month-old infants
began to orient their hand to the orientation of an ob-
ject during the reach, but that 5-month-olds did not.
Figure 3 Hand orientation during the reach for all trials of one participant in each age group. Each line represents the data from
a single trial.
982 Child Development
The data therefore agree more with those of Lockman
et al. (1984) than with Morrongiello and Rocca (1989)
and von Hofsten and Fazel-Zandy (1984), who report
some degree of preorienting in 5-month-olds. It is
not clear whether 5-month-olds cannot fully control
their hand while moving their arm toward the location
of the target, or whether they have not yet learned
about the advantages of preorienting their hand to
the target.
The extent to which infants preoriented their hand
to the orientation of an object was not affected by
darkening the object during the reach, which left the
infant to complete the reach and grasp in complete
darkness. Although hand orientation was not altered
by loss of sight of the object, other kinematic measures
were affected. McCarty and Ashmead (1999) found
that movements toward the object took longer and
involved more corrections if infants saw the target
throughout the reach. Control over hand orientation
seems immune to sight of the object during the reach.
However, the object was visible at the start of the
reach and during part of the trajectory. The question
remains whether hand orientation is controlled on the
basis of target information before motor action be-
gins. In Experiment 3 all visual information was re-
moved before the reach was initiated.
EXPERIMENT 3
The purpose of this experiment was to assess whether
infants can rely on their memory of an object’s orienta-
tion to guide reaching and grasping. Only 9-month-old
infants were tested because only their hand orienta-
tion data in Experiment 2 reflected the rod’s orientation
from midreach on through contact. To test whether
infants could match hand orientation with their mem-
ory of the rod’s orientation, all visual information was
removed at the start of the trial.
Method
Participants
Nine 9-month-olds (4 females, 5 males; age: M
277 days, SD 3.2) provided kinematic data in all
of the trials (i.e., both orientations and both light-
ing conditions). Nine other infants also participated,
but their data were not included because of an equip-
ment problem (n 1), refusal to leave the markers
taped to the hand (n 1), no reaching in either
lighting condition (n 1), no reaching in the dark
(n 2), and no kinematic data obtained for one of
the orientations in one of the lighting conditions
(n4).
Apparatus
The apparatus from Experiment 2 was modified in
two ways for this experiment. First, the apparatus
was placed on drawer slides so that the rod could be
easily moved within or beyond an infant’s reaching
space. Second, a loudspeaker was placed directly be-
hind the rod at the center of rotation. The speaker was
connected to a tape recorder, and a tape of either bells
or rattles was played during each trial. The purpose
of this modification was to give infants an auditory
cue to the distance and location of the rod for reach-
ing, but no cue to the orientation of the rod for grasp-
ing.1 Infants at 6 months of age have been shown to
use these same sounds to localize objects in the azi-
muth and in depth (Clifton, Perris, & Bullinger, 1991;
Litovsky & Clifton, 1992).
As in Experiment 2, two hand markers from the
Optotrak motion analysis system were placed on
the infants’ right hand. Each trial lasted 15 s, and the
sampling rate was 50 Hz. Each session was also
videotaped under infrared light and was time locked
to the motion analysis system with a date-timer as in
Experiment 1.
Procedure
All trials were conducted in a darkened room, with
the lighted rod as the only source of illumination. Al-
though the rod was not illuminated, the room was to-
tally dark. During intertrial intervals, a room light
was turned on. The rod was withdrawn beyond the
infant’s reach and placed in either the horizontal or
vertical orientation prior to the start of a trial. The in-
fant’s attention was drawn to the apparatus, the tape
recording was started, the apparatus lights were
turned on to light the rod, the room lights were dark-
ened, and data collection began. On glowing trials
the visible rod was brought within the infant’s
reaching space. On dark trials after the infant had
seen the orientation of the rod, the rod was dark-
ened and then brought within the infant’s reaching
space. The experimenter sometimes flashed the ap-
paratus lights during the trial if an infant had not
initiated a reach and the experimenter thought that
the infant was off-task (e.g., had turned toward the
1Five adults were tested on this apparatus to determine if
there were any cues to the orientation of the rod when the lights
were darkened and the audio tape was playing. The lighted rod
was in an oblique orientation in a darkened room when the
audio tape was started. The rod was then darkened, rotated to
either the vertical or horizontal orientation, and pushed forward
to be grasped. The adults were not able to preorient their hand
to the orientation of the rod; thus, it was concluded that there
were no nonvisual cues to the rod’s orientation.
McCarty et al. 983
parent). The experimenter flashed the lights in an
attempt to reorient the infant to the task. At times,
however, an infant initiated a reach during the times
the rod was flashed, and these trials had to be ex-
cluded. Trials continued to be presented for as long as
the infant cooperated.
Data Coding and Measures
Reach onset, contact, and pickup times were coded
from videotape, as well as the time when the object
was last darkened if it had been flashed after trial on-
set. Two coders scored tapes of 4 infants. For contact
times, the coders agreed to within 100 ms on 80% of
trials, and to within 200 ms on 92% of trials.
Results and Discussion
The purpose of this study was to determine when
the reach was initiated in darkness, whether infants
began to preorient their hand to the remembered ori-
entation of the rod. Thus, the focus of the analyses
was on whether, at contact, hand orientation differed
between horizontal and vertical rods in the different
lighting conditions. A total of 119 glowing and 127
dark trials were presented to the infants, who initi-
ated a reach on 112 of the glowing trials and on 106 of
the dark trials. For 50 of the dark trials, however, the
reach was initiated before the rod was darkened or
when the rod was flashed. Therefore, these trials were
excluded from the analyses. Also, there were missing
motion analysis system data for 2 glowing trials and
these trials were also excluded. Thus, data from 110
glowing trials (47 horizontal and 63 vertical) and 56
dark trials (26 horizontal and 30 vertical) were analyzed.
Each child’s mean hand orientation at contact was
determined for horizontal and vertical trials in both
lighting conditions. When the rod was in the horizon-
tal orientation, mean hand orientation at contact was
17 (SD 10) when the rod was lit, and 22 (SD 19)
when it was dark. When the rod was vertical, mean
hand orientation at contact was 65 (SD 17) in the
glowing condition, and 56 (SD 26) in the dark
condition. These results indicate that infants preori-
ented their hand in both lighting conditions. Further-
more, all 9 infants had higher mean hand orientations
at contact for the vertical rod than for the horizontal
rod, both in the glowing condition and in the dark
condition. A 2 (lighting condition) 2 (orientation)
ANOVA was performed on these data. Infants differ-
entially oriented their hand as a function of the rod’s
orientation, F(1, 8) 46.40, p .001, with no effect
found for lighting condition nor a Lighting Condition
Orientation interaction. Infants demonstrated that
they remembered and utilized information about the
orientation of an object while executing a reach in
darkness. One implication of this result is that infants
can encode aspects of an object during a preview so
that they do not have to monitor this information dur-
ing execution of the reach.
GENERAL DISCUSSION
The focus of these experiments was on the role of vi-
sion as infants prepared to grasp an object that varied
in orientation and in lighting condition. Several
studies have documented that infants appropriately
orient their hands before contact with vertically and
horizontally oriented rods (Lockman et al., 1984;
Morrongiello & Rocca, 1989; von Hofsten & Fazel-
Zandy, 1984; Wentworth et al., 2000). In those experi-
ments infants could see both the rod and their hand,
and may have been visually matching their hand to
the rod’s orientation throughout the reach. In the
present three experiments sight of the target was
variedfrom fully visible throughout the reach (Ex-
periment 1), to visible in the early stages of the reach
(Experiment 2), to invisible throughout the entire
reach (Experiment 3). Sight of the hand was occluded
by darkness in all three experiments.
In Experiment 1, 7.5-month-old infants oriented their
hands correctly to a glowing rod in both light and dark
conditions. Because the infants could not see their hand
in the dark, these findings indicate that they were not
relying on sight of their hand in making the adjust-
ments. In Experiment 2, age differences were demon-
strated: At 7 and 9 months of age, infants began to ori-
ent their hand to the rod’s orientation while the reach
was in progress (prior to contact), both when the rod
was visible and when it was darkened. At 5 months,
however, infants did not anticipate the rod’s orientation
in either lighting condition. When infants have enough
motor control (i.e., at 7 months) to preorient their hands
to an object, they do not have to monitor the object
throughout the entire reach. In Experiment 3, 9-month-
old infants had a preview of the rod’s orientation be-
fore they initiated their reach, but no sight of the rod
at the start of or during the reach. Again, infants ap-
propriately oriented their hand for horizontally and
vertically oriented rods prior to contact. The infants
remembered the rod’s orientation, then prepared and
carried out the reach and grasp accordingly.
The findings of each experiment contribute to the
understanding of early perceptual-motor and cogni-
tive development. The results of Experiment 1 lead
one to distinguish between the importance of seeing
the hand during the reach and the importance of see-
984 Child Development
ing the hand at other times (e.g., during exploratory
or non-goal-directed movements). The results of Ex-
periment 2 lead one to consider whether early grasp-
ing in 5-month-olds is visually or haptically con-
trolled, and the developmental changes that occur over
the next 2 to 4 months. The results of Experiment 3
have implications for whether infants have motor rep-
resentations. Each of these ideas is discussed in turn.
Sight of the Hand
A striking finding that emerged in all three experi-
ments was that lighting condition did not affect hand
orientation at any age. It was concluded that as soon
as infants developed the necessary motor control to
engage in preorientation of their hand, sight of the
hand was not important. How does one control hand
orientation without looking at it? Two senses provide
information on a hand’s location in space: vision and
proprioception, the latter being defined as the posi-
tion sense of the movable parts of the body. Sacks
(1985) provides an excellent description of the conse-
quences of losing the sense of proprioception as an
adult in a chapter entitled “The Disembodied Lady.”
He describes how a woman was forced to rely solely
on vision to regulate her movements, and even to
know the position and placement of her limbs. The
present research indicates that infants have adequate
proprioceptive control of their arm and hand by 7
months of age to preorient their hand appropriately
without vision. This conclusion leads to consider-
ation of how proprioception develops.
The available evidence suggests that the develop-
ment of proprioception is facilitated by sight of the
limbs; several findings lead to this conclusion. First,
proprioception may not develop as fully or as quickly
when one is blind; Warren (1984) described several
interventions designed to help children who were vi-
sually impaired develop the sense of proprioception.
Of course, their visual impairment prevented sight of
the environment as well as their own limbs. This
leads to a second line of research in support of the ar-
gument that sight of the limbs facilitates the develop-
ment of proprioception. When monkeys are reared
with normal vision but without sight of their limbs,
they are impaired in tasks that involve proprioception
of the unseen limbs. Held and Bauer (1974) reared
monkeys in this way and found that they were delayed
in the development of reaching, but that they quickly
learned to reach for objects once they were allowed
sight of their limbs. Furthermore, once they had seen
their limbs and started to reach, the monkeys readily
reached for objects even when tested without sight of
their limbs (i.e., when proprioception was required).
This area of research provides strong evidence that
seeing one’s limbs enables an organism to use and in-
terpret proprioceptive information from those limbs.
Recent findings indicate that newborn infants at-
tend to the sight of their moving limbs. Newborns
moved a seen hand more often than an unseen one,
even when they were seeing the hand by looking in a
monitor placed on the opposite side of their body (ver
der Meer, van der Weel, & Lee, 1995, 1996). This ten-
dency to move the seen hand may enable newborns
to start to link the location of the visible hand with the
location of the felt hand because movement provides
proprioceptive feedback. By 3 to 5 months, infants
have enough experience observing their limb move-
ments to recognize an anomalous view. Infants in this
age range can discriminate between a normal view of
their own legs and a view that reversed the leftright
location of their legs (Morgan & Rochat, 1997), which
indicates that they have already learned some of the
visual consequences of their own movements. Fur-
ther evidence for the early development of proprio-
ception in humans comes from a longitudinal study
of the onset of reaching (Clifton, Muir, Ashmead, &
Clarkson, 1993). Infants were tested weekly from be-
fore they had demonstrated reaching behavior until
they reached for and grasped a toy reliably. The in-
fants were able to reach for and grasp glowing objects
in total darkness at the same age as they first accom-
plished this task in the light. From the very onset of
reaching behavior, the infants appeared to sense the
position of their arm and hand without actually see-
ing the limb. In interpreting these findings, Clifton et
al. (1993) stressed the importance of infants’ early vi-
sual experience of their limbs and the environment
prior to the onset of reaching behavior. To summarize,
it should be emphasized that there is a distinction be-
tween using sight of the hand while executing a
reach, which does not seem to be necessary during in-
fancy, and looking at the hand during the early months
of life, which is probably critical for the development
of proprioception.
Furthermore, seeing the hand, even when not reach-
ing with it, continues to be important for accurate
reaching throughout the life span. Jeannerod (1988)
suggests that, outside of the context of reaching, sight
of the hand is used by adults to match the propriocep-
tive map (which encodes limb position relative to the
body) with the visual map (which encodes limb posi-
tion relative to the target).
Grasping Control among 5-Month-Olds
Whether grasping should be considered haptically
or visually controlled early in development is a con-
McCarty et al. 985
troversial issue (Piaget, 1936/1952; Pieraut-Le Bon-
niec, 1990; Twitchell, 1970; von Hofsten, 1990, 1993). If
grasping is haptically controlled, infants would not
preorient their hand to the orientation of the object on
the basis of vision. Rather, they would contact an ob-
ject first, and then adjust their hand to grasp it. If
grasping is visually controlled, infants may be able to
preorient their hand to the orientation of the object
during the reach. Advocates of the first perspective
have argued that early grasping is haptically con-
trolled, but that vision takes over the control of grasp-
ing at around 6 months of age (Piaget, 1936/1952;
Pieraut-Le Bonniec, 1990; Twitchell, 1970). The transi-
tion may occur after the position and location of the
felt hand becomes coordinated with the position and
location of the visible hand and target. Hence, this hy-
pothesis involves learning and coordination. Advo-
cates of the second perspective argue that purposeful
movements are under prospective control, and that
some aspects of early grasping should be considered
anticipatory in nature (von Hofsten, 1990, 1993).
The findings in Experiment 2 are consistent with a
learning and coordination explanation. Specifically,
the 5-month-olds did not differentially preorient their
hand to the orientation of the target; the 7-month-olds
did; and the 9-month-olds and adults continued to pre-
orient their hand to an even greater extent, to match
the orientation of the target.2 It is unclear, however,
how to attribute the failure of 5-month-olds to preori-
ent their hand to the target. One possibility is that
they do not know that it is desirable to orient their
hand to the orientation of the object, and that they
must learn this knowledge from manipulating ob-
jects. Another possibility is that they do know to ori-
ent their hand to match the orientation of an object,
but that poor motor control does not allow them to
demonstrate this knowledge.
Whatever the reason, the statement that purpose-
ful movements are under prospective control may
need to be qualified. Novice performers in most tasks
may be goal directed, but their actions are usually
controlled by feedback rather than by prospective
means (McCarty, Clifton, & Collard, 1999). One could
argue that learners are adapted to detect relevant in-
formation in the environment and motivated to act
prospectively on that information. Then purposive
movements could be said to be under prospective
control, with the novice performer defined as one
who has not yet learned to detect or act prospectively
on the relevant information. Most 5-month-olds are
just learning to grasp objects of various sizes and
shapes. Within the next 2 to 4 months, however, in-
fants show prospective control by preadapting their
hand to the object’s orientation as well as to its size
(von Hofsten & Ronnqvist, 1988).
Motor Representations
In Experiment 3, the 9-month-old infants remem-
bered the orientation of an object and acted on that
memory to preorient their hand in preparation for
grasping the object. Stated differently, even with no
visual support from either hand or object, infants were
able to differentially preorient their hand to the object.
It is suggested that these infants guided their actions
with a motor representation that included the relation
between the orientation of their hand and the orienta-
tion of the object (Jeannerod, 1997; see also Arbib,
1981). In Experiments 1 and 2, infants demonstrated
knowledge of their unseen hand’s orientation, which
may be considered a motor representation, but this
evidence was less strong than in Experiment 3, in
which infants also remembered the orientation of the
object. The following discussion speculates on how
this representation fits into the adult cognitive neuro-
science literature on representational systems.
Milner and Goodale (1995) distinguished two sys-
tems of object knowledge: an action system that is used to
control goal-directed actions, and a perceptual system that
is used to form representations of the enduring charac-
teristics of objects (i.e., categorize objects). Whereas the
action system involves the rapid transformation of sen-
sory input into motor commands, the perceptual system
refers to the use of perceptual and cognitive cues that
may also be used to generate actions. In adults, the ac-
tion system appears to remain active only for about 2 s
after an object is no longer visible (Milner & Goodale;
see also Elliott & Madalena, 1987). When action is de-
layed for longer than 2 s, adults must rely on represen-
tations in the perceptual system to guide their actions.
The limited time duration of the action system is clearly
seen in the behavior of people with visual agnosia, who
have an intact action system but a deficient perceptual
system (Milner & Goodale, 1995). People with visual ag-
nosia who perform some action on an object immedi-
ately or after a very brief delay act appropriately toward
the object. Their actions, however, are no longer appro-
2Note that even the adults did not fully preorient their hand
to the orientation of the target. At contact, the mean hand orien-
tation for adults across the two lighting conditions was 13.4
for the horizontal rod and 65.5 for the vertical rod, which
did not match the orientation of either rod (0 and 90, respec-
tively). Mean hand orientation for adults at pickup, however,
was 13.2 for the horizontal rods and 70.0 for the vertical
rods; therefore their hand was almost fully preoriented at
contact. Flexibility in the fingers may allow infants and adults
to make a power grip without fully aligning their hand to the
target.
986 Child Development
priate when delays exceed 2 s. For example, when
reaching for different-sized objects, maximum aperture
between thumb and index finger during reach was
scaled to the size of the object following a short delay,
but it was no longer scaled to object size following a
longer delay (Goodale, Jacobson, & Keillor, 1994).
The present study was not designed to distinguish
between these representational systems, but it is ap-
propriate to speculate about their role in this research.
When the object darkened after the onset of the reach
in Experiment 2, infants could probably rely on the
action system representations to continue with a
reach that was already in progress. When the object
darkened before the reach in Experiment 3, however,
infants probably had to rely on representations in the
perceptual system to initiate and complete the reach.
Because the infants were able to preorient their hand
even when the object was not visible at the onset of
the reach, the perceptual system appeared to be func-
tional in 9-month-old infants.
There is further evidence that representations in
the perceptual system can be used to generate actions
in infants. Goubet and Clifton (1998) presented 7-
month-old infants with a task in which a ball fell nois-
ily through a tube and then came to rest silently at one
of two locations that had been specified earlier in the
sequence by distinctive auditory information. In total
darkness, infants initiated a reach to the inferred rest-
ing location of the ball, and they averaged initiating
the first reach in the dark more than 4 s after the ball
was silent. This delay is well beyond the 2-s threshold
for when action system representations fade in
adults. More generally, the development of these two
representational systems is a fitting topic for future
research in psychology (e.g., Bertenthal, 1996). In-
deed, the distinction may fit into an historical and on-
going debate concerning the development of knowl-
edge (e.g., Hume, 1740/1987; Kant, 1781/1965). One
side argues that knowledge is built from actions (e.g.,
Bushnell & Boudreau, 1993; Gibson, 1988; Piaget,
1936/1952; Thelen & Smith, 1994 ). This can be recast
as arguing that perceptual system representations
emerge from action system representations. The other
side argues that knowledge in infants is native or can be
built from observing the world (e.g., Baillargeon, 1999;
Spelke, 1998; Wynn, 1992). From this perspective, per-
ceptual system representations may precede action
system representations, which may develop slowly be-
cause of the slow motor development of human infants.
In conclusion, by systematically reducing the vi-
sual information available to infants as they reached
for horizontally and vertically oriented rods, infor-
mation was obtained about their perceptual-motor
development, leading to speculations about issues in
cognitive development. It was found that infants dif-
ferentially preoriented their hand to the object they
were reaching for under a variety of conditions with
reduced visual information. It was concluded that
proprioceptive control of hand and arm movements
develops early in infancy, and that this development
is probably facilitated by infants watching their limbs
move; that the early control of hand orientation may
not be controlled in a prospective manner, but that it
quickly comes to be prospectively controlled; and
that infants have sufficient knowledge of a target and
their arm to initiate and complete a reach without vi-
sual support, and they may use a motor representa-
tion of the event to accomplish this feat.
ACKNOWLEDGMENTS
This research was supported by grants HD27714 and
HD23191 from the National Institutes of Child Health
and Human Development, by a Research Scientist
Award (MH00332) from the National Institute of Men-
tal Health, and by a National Science Foundation grad-
uate fellowship. The authors thank Linda Lucas and
Virginia Hitch for their help in collecting and scoring
the data from Experiment 2, and Roselee Brown for her
help in scoring the data from Experiments 2 and 3.
ADDRESSES AND AFFILIATIONS
Corresponding authors: Michael E. McCarty or
Rachel K. Clifton, Department of Psychology, Univer-
sity of Massachusetts, Tobin Hall, 135 Hicks Way,
Amherst, MA 01003-9271; e-mail: mccarty@psych.
umass.edu or rachel@psych.umass.edu. Daniel H.
Ashmead is at Vanderbilt University, Nashville, TN;
Philip Lee and Nathalie Goubet are also at the Uni-
versity of Massachusetts.
REFERENCES
Arbib, M. A. (1981). Perceptual structures and distributed
motor control. In V. B. Brooks (Ed.), Handbook of physiol-
ogy, Section 1: The nervous system, Vol. 2. Motor control (pp.
1449– 1480). Baltimore: Williams & Wilkins.
Baillargeon, R. (1999). Young infants’ expectations about
hidden objects: A reply to three challenges. Developmen-
tal Science, 2, 115–132.
Bertenthal, B. I. (1996). Origins and early development of
perception, action, and representation. Annual Review of
Psychology, 47, 431–459.
Berthier, N. E. (1996). Learning to reach: A mathematical
model. Developmental Psychology, 32, 811–823.
Bushnell, E. W., & Boudreau, J. P. (1993). Motor develop-
ment and the mind: The potential role of motor abilities
as a determinant of aspects of perceptual development.
Child Development, 64, 1005– 1021.
McCarty et al. 987
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, 1099– 1110.
Clifton, R., Perris, E., & Bullinger, A. (1991). Infants’ perception
of auditory space. Developmental Psychology, 27, 187–197.
Clifton, R. K., Rochat, P., Litovsky, R. Y., & Perris, E. E.
(1991). Object representation guides infants’ reaching in
the dark. Journal of Experimental Psychology: Human Per-
ception and Performance, 17, 323–329.
Clifton, R., Rochat, P., Robin, D. J., & Berthier, N. E. (1994).
Multimodal perception in the control of infant reaching.
Journal of Experimental Psychology: Human Perception and
Performance, 20, 876–886.
Elliott, D., & Madalena, J. (1987). The influence of premove-
ment visual information on manual aiming. Quarterly
Journal of Experimental Psychology, 39A, 541–559.
Gibson, E. J. (1988). Exploratory behavior in the develop-
ment of perceiving, acting, and the acquiring of knowl-
edge. Annual Review of Psychology, 39, 1–41.
Goodale, M. A., Jacobson, L. S., & Keillor, J. M. (1994). Dif-
ferences in the visual control of pantomimed and natural
grasping movements. Neuropsychologia, 32, 1159–1178.
Goubet, N., & Clifton, R. K. (1998). Object and event repre-
sentation in 6.5-month-old infants. Developmental Psy-
chology, 34, 63–76.
Held, R., & Bauer, J. A., Jr. (1974). Development of sensori-
ally-guided reaching in infant monkeys. Brain Research,
71, 265– 271.
Howell, D. C. (1987). Statistical methods for psychology (2nd
ed.). Boston: Duxbury.
Hume, D. (1987). A treatise on human nature. In E. C. Mossner,
(Ed.). New York: Penguin. (Original work published 1740)
Jeannerod, M. (1988). The neural and behavioural organization
of goal-directed movements. New York: Oxford University
Press.
Jeannerod, M. (1997). The cognitive neuroscience of action.
Cambridge, MA: Blackwell.
Kant, I. (1965). Critique of pure reason. (N. K. Smith, Trans.).
New York: St. Martin’s Press. (Original work published
1781)
Litovsky, R. Y., & Clifton, R. K. (1992). Use of sound pres-
sure in auditory distance discrimination by 6-month-old
infants and adults. Journal of Acoustical Society of America,
92, 794– 802.
Lockman, J. J., Ashmead, D. H., & Bushnell, E. W. (1984).
The development of anticipatory hand orientation dur-
ing infancy. Journal of Experimental Child Psychology, 37,
176– 186.
McCarty, M. E., & Ashmead, D. H. (1999). Visual control of
reaching and grasping in infants. Developmental Psychol-
ogy, 35, 620–631.
McCarty, M. E., Clifton, R. K., & Collard, R. R. (1999). Prob-
lem solving in infancy: The emergence of an action plan.
Developmental Psychology, 35, 1091–1101.
Milner, A. D., & Goodale, M. A. (1995). The visual brain in ac-
tion. New York: Oxford University Press.
Morgan, R., & Rochat, P. (1997). Intermodal calibration of
the body in early infancy. Ecological Psychology, 9, 1–23.
Morrongiello, B. A., & Rocca, P. T. (1989). Visual feedback
and anticipatory hand orientation during infants’ reach-
ing. Perceptual and Motor Skills, 69, 787– 802.
Perris, E. E., & Clifton, R. K. (1988). Reaching in the dark to-
ward sound as a measure of auditory localization in in-
fants. Infant Behavior and Development, 11, 473–491.
Piaget, J. (1952). The origins of intelligence in children (M.
Cook, Trans.). New York: Norton. (Original work pub-
lished 1936)
Pieraut-Le Bonniec, G. (1990). Reaching and hand adjusting
to the target properties. In H. Bloch & B. I. Bertenthal
(Eds.), Sensory-motor organizations and development in in-
fancy and early childhood (pp. 301–314). Dordrecht, The
Netherlands: Kluwer Academic Publishers.
Robin, D. J., Berthier, N. E., & Clifton, R. K. (1996). Infants’
predictive reaching for moving objects in the dark. De-
velopmental Psychology, 32, 824–835.
Sacks, O. (1985). The man who mistook his wife for a hat and
other clinical tales. New York: Summit Books.
Spelke, E. S. (1998). Nativism, empiricism, and the origins of
knowledge. Infant Behavior and Development, 21, 181– 200.
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 Devel-
opment, 64, 1058– 1098.
Thelen, E., & Smith, L. B. (1994). A dynamic systems approach
to the development of cognition and action. Cambridge, MA:
MIT Press.
Twitchell, T. E. (1970). Reflex mechanisms and the develop-
ment of prehension. In K. Connolly (Ed.), Mechanisms of
motor skill development (pp. 25–38). New York: Academic.
van der Meer, A. L. H., van der Weel, F. R., & Lee, D. N.
(1995). The functional significance of arm movements in
neonates. Science, 267, 693–695.
van der Meer, A. L. H., van der Weel, F. R., & Lee, D. N. (1996).
Lifting weights in neonates: Developing visual control of
reaching. Scandinavian Journal of Psychology, 37, 424–436.
von Hofsten, C. (1990). Early development of grasping an ob-
ject in space-time. In M. A. Goodale (Ed.), Vision and action:
The control of grasping (pp. 65–79). Norwood, NJ: Ablex.
von Hofsten, C. (1993). Prospective control: A basic aspect
of development. Human Development, 36, 253– 270.
von Hofsten, C., & Fazel-Zandy, S. (1984). Development of
visually guided hand orientation in reaching. Journal of
Experimental Child Psychology, 38, 208–219.
von Hofsten, C., & Ronnqvist, L. (1988). Preparation for
grasping an object: A developmental study. Journal of Ex-
perimental Psychology: Human Perception and Performance,
14, 610– 621.
Warren, D. H. (1984). Blindness and early childhood develop-
ment (2nd ed., rev.). New York: American Foundation for
the Blind.
Wentworth, N., Benson, J. B., & Haith, M. M. (2000). The de-
velopment of infants’ reaches for stationary and moving
objects. Child Development, 71, 576– 601.
Wynn, K. (1992). Addition and subtraction by human in-
fants. Nature, 358, 749– 750.
... Infants and toddlers gather object-related information through visual attention and use this information to inform specific goal-directed actions. For example, visually perceiving the orientation of an object, an occluded path, or that an object is large, informs how to effectively reach for the object (Barrett et al. 2008;McCarty et al. 2001a). Visual attention also allows infants to update their strategic reaches when object orientation changes, or employ visual memory when objects are obscured (McCarty et al. 2001a). ...
... For example, visually perceiving the orientation of an object, an occluded path, or that an object is large, informs how to effectively reach for the object (Barrett et al. 2008;McCarty et al. 2001a). Visual attention also allows infants to update their strategic reaches when object orientation changes, or employ visual memory when objects are obscured (McCarty et al. 2001a). Children with DS, on the other hand, misallocate attention from objects towards social engagement (Fidler et al. 2005b;Kasari and Freeman 2001), which may suggest an impairment the use of visual attention that is critical for supporting goal-directed behavior. ...
Article
Full-text available
Objectives Early goal-directed behavior is an important area of cognitive development for long-term outcomes in typical development yet is substantially disrupted in Down syndrome (DS). While motor abilities, visual attention, and haptic perception are established associated mechanisms of early goal-directed behavior in typical development, there has been minimal investigation into these as mechanisms of goal-directed behavior in children with DS. The aim of the present study was to examine fine motor, gross motor, visual attention, and haptic perception as associated mechanisms of two types of goal-directed behavior in DS. Methods Participants included 38 infants and toddlers with DS (mean CA=24.84 months). Measures included standardized assessments of fine and gross motor, and behavioral assessments of visual attention, tactile integration, problem solving, and functional object use. Results Results identified specificity in associated mechanisms across different types of goal-directed behavior—problem solving, and functional object use. Haptic perception was identified as the most salient predictor of goal-directed behavior. Conclusions Collectively, findings indicate that haptic perception and fine and gross motor abilities are essential mechanisms of goal-directed behavior in DS and therefore may serve as potential targets for intervention.
... The present considerations are limited to examining grip selection during object manipulation in ESC tasks. Some intriguing observations by Keen and colleagues suggest that children show advanced planning skills for object manipulation at much earlier ages (e.g., Clifton, Rochat, Litovsky, & Perris, 1991;McCarty et al., 1999;McCarty, Clifton, Ashmead, Lee, & Goubet, 2001). For example, Clifton et al. (1991) presented children either a small or a large object in the dark (after a familiarization phase in the light), which afforded either a one-handed or a two-handed grip configuration. ...
... That is, they selected a one-handed grip for small objects and a two-handed grip for large objects. In another study by McCarty, Clifton, Ashmead et al. (2001), children were presented a horizontally or vertically oriented rod in the light. When the light was switched off, 9-month-old children already oriented their grasp in a way to adapt to the rod's orientation. ...
... In contrast, at 4 mo, infants are unable to grasp objects, showing immaturity of the networks that control hand movements and hand-object interaction. Interest in graspable objects increases during the first year of life (52,53), as infants develop grasping skills (57). Consistent with this trajectory, we found that by 10 mo, the preference for big over small objects had disappeared. ...
Article
Full-text available
Significance Categorization is the basis of thinking and reasoning. Through the analysis of infants’ gaze, we describe the trajectory through which visual object representations in infancy incrementally match categorical object representations as mapped onto adults’ visual cortex. Using a methodological approach that allows for a comparison of findings obtained with behavioral and brain measures in infants and adults, we identify the transition from visual exploration guided by perceptual salience to an organization of objects by categories, which begins with the animate–inanimate distinction in the first months of life and continues with a spurt of biologically relevant categories (human bodies, nonhuman bodies, nonhuman faces, small natural objects) through the second year of life.
... Some studies indicate that directed actions might already be available at birth, but the movement is so jerky and irregular that it is difficult to tease apart the effects of inaccurate target specification and faulty execution [137]. At seven months infants adjust grasping movements (e.g., hand aperture) correctly, and they do it even when the object is not visible [138]. This finding suggests that they use a mental representation of object size to adapt their grasp. ...
Article
Full-text available
The aim of this article is to discuss the logic and assumptions behind the concept of neural reuse, to explore its biological advantages and to discuss the implications for the cognition of a brain that reuses existing circuits and resources. We first address the requirements that must be fulfilled for neural reuse to be a biologically plausible mechanism. Neural reuse theories generally take a developmental approach and model the brain as a dynamic system composed of highly flexible neural networks. They often argue against domain-specificity and for a distributed, embodied representation of knowledge, which sets them apart from modular theories of mental processes. We provide an example of reuse by proposing how a phylogenetically more modern mental capacity (mental rotation) may appear through the reuse and recombination of existing resources from an older capacity (motor planning). We conclude by putting arguments into context regarding functional modularity, embodied representation, and the current ontology of mental processes.
... Hand pre-shaping is the orienting of the hand and fingers during the movement of the arm toward the target, configuring them for the final grasp action, independently of the size, shape, and position of the target object. This movement arises several weeks after the start of reaching and by nine months infants can correctly orient their hand toward the target once they have previewed it [188]. [189], [190] address development in the sense of learning the association of motor babbling actions with hand tracking and object recognition, without the morphological changes we have defined in section V. Caligiore at al. [191] reference to the strategy of actuating over the joint properties (acting over the J Ƥ matrix) of the robot by locking DOFs of the arm, to facilitate learning. ...
Preprint
Full-text available
Humans and animals undergo morphological development processes from infancy to adulthood that have been shown to facilitate learning. However, most of the work on developmental robotics considers fixed morphologies, addressing only the development of the cognitive system of the robots. This paper aims to provide a survey of the work that is being carried out within the relatively new field of morphological development in robots. In particular, it contemplates morphological development as the changes that occur in the properties of the joints, links and sensors of a robot during its lifetime and focuses on the work carried out by different authors to try to determine their influence on robot learning. To this end, walking, reaching, grasping and vocalization have been identified as the four most representative tasks addressed in the field, clustering the work of the different authors around them. The approach followed is multidisciplinary, discussing the relationships among developmental robotics, embodied artificial intelligence and developmental psychology in humans in general, as well as for each of the tasks, and providing an overview of the many avenues of research that are still open in this field.
... In contrast, at 4 months, infants are unable to grasp objects, showing immaturity of the networks that control hand movements and hand-object interaction. Interest in graspable objects increases during the first year of life 45,46 , as infants develop grasping skills 49 . Consistent with this trajectory, we found that by 10 months, the preference for big over small objects had disappeared. ...
Preprint
Full-text available
Humans make sense of the world by organizing things into categories. When and how does this process begin? We investigated whether real-world object categories that spontaneously emerge in the first months of life match categorical representations of objects in the human visual cortex. Taking infants' looking times as a measure of similarity, we defined a representational space where each object was defined in relation to others of the same or different categories. This space was compared with hypothesis-based and fMRI-based models of visual-object categorization in the adults' visual cortex. Analyses across different age groups revealed an incremental process with two milestones. Between 4 and 10 months, visual exploration guided by saliency gives way to an organization according to the animate-inanimate distinction. Between 10 and 19 months, a category spurt leads towards a mature organization. We propose that these changes underlie the coupling between seeing and thinking in the developing mind.
... In contrast, at 4 mo, infants are unable to grasp objects, showing immaturity of the networks that control hand movements and hand-object interaction. Interest in graspable objects increases during the first year of life (52,53), as infants develop grasping skills (57). Consistent with this trajectory, we found that by 10 mo, the preference for big over small objects had disappeared. ...
... In a developing human, learning the visuomotor task of reaching for an object goes hand-in-hand with the visual tasks of localizing and identifying the object in the visual field [32]. On a neurocognitive level, these tasks share an important component, namely the early visual system. ...
Article
Full-text available
We present a follow-up study on our unified visuomotor neural model for the robotic tasks of identifying, localizing, and grasping a target object in a scene with multiple objects. Our Retinanet-based model enables end-to-end training of visuomotor abilities in a biologically inspired developmental approach. In our initial implementation, a neural model was able to grasp selected objects from a planar surface. We embodied the model on the NICO humanoid robot. In this follow-up study, we expand the task and the model to reaching for objects in a three-dimensional space with a novel dataset based on augmented reality and a simulation environment. We evaluate the influence of training with auxiliary tasks, i.e., if learning of the primary visuomotor task is supported by learning to classify and locate different objects. We show that the proposed visuomotor model can learn to reach for objects in a three-dimensional space. We analyze the results for biologically-plausible biases based on object locations or properties. We show that the primary visuomotor task can be successfully trained simultaneously with one of the two auxiliary tasks. This is enabled by a complex neurocognitive model with shared and task-specific components, similar to models found in biological systems.
Article
Humans and animals undergo morphological development processes from infancy to adulthood that have been shown to facilitate learning. However, most of the work on developmental robotics considers fixed morphologies, addressing only the development of the cognitive system of the robots. This paper aims to provide a survey of the work that is being carried out within the relatively new field of morphological development in robots. In particular, it contemplates morphological development as the changes that occur in the properties of the joints, links and sensors of a robot during its lifetime and focuses on the work carried out by different authors to try to determine their influence on robot learning. To this end, walking, reaching, grasping and vocalization have been identified as the four most representative tasks addressed in the field, clustering the work of the different authors around them. The approach followed is multidisciplinary, discussing the relationships among developmental robotics, embodied artificial intelligence and developmental psychology in humans in general, as well as for each of the tasks, and providing an overview of the many avenues of research that are still open in this field.
Article
Full-text available
First published in 1995, this book presents a model for understanding the visual processing underlying perception and action, proposing a broad distinction within the brain between two kinds of vision: conscious perception and unconscious 'online' vision. It argues that each kind of vision can occur quasi-independently of the other, and is separately handled by a quite different processing system. For this new edition, the text from the original edition has been left untouched, standing as a coherent statement of the authors' position. However, a very substantial epilogue has been added to the book, which reviews some of the key developments that support or challenge the views that were put forward in the first edition. The new chapter summarizes developments in various relevant areas of psychology, neuroscience, and behaviour. It supplements the main text by updating the reader on the contributions that have emerged from the use of functional neuroimaging, which was in its infancy when the first edition was written. Neuroimaging, and functional MRI in particular, has revolutionized the field by allowing investigators to plot in detail the patterns of activity within the visual brains of behaving and perceiving humans. The authors show how its use now allows scientists to test and confirm their proposals, based largely on evidence accrued from primate neuroscience in conjunction with studies of neurological patients.
Article
Early childhood development is a robust and vibrant focus of study in multiple disciplines, from economics and education to psychology and neuroscience. Abundant research from these disciplines has established that early childhood is critical for the development of cognitive abilities, language, and psychosocial skills, all of which turn, in large measure, on the parent-child relationship. And because early childhood relationships and experiences have a deep and lasting impact on a child's life trajectory, disadvantages during early childhood replicate inequality. Working together, scholars in these disciplines are actively engaged in a national policy debate about reducing inequality through early childhood interventions. Despite the vital importance of this period, the law and legal scholars have been largely indifferent to the dynamics of early childhood development. Doctrine and legislation are rarely developmentally sensitive, lumping children into an undifferentiated category regardless of age. The legal system thus misses key opportunities to combat inequality and foster healthy development for all children. And most legal scholars do not engage with the wealth of interdisciplinary research on early childhood, nor are they part of the interdisciplinary dialogue and policy debates. As a result, that conversation does not include the voices of lawyers and legal scholars, who are uniquely positioned to add critical insights. Remedying this stark disconnect requires doing for law what scholars have done in other disciplines: creating a distinctive field. Accordingly, this Article proposes a subdiscipline of early childhood development and the law. The new field crystallizes a distinctive interest that the legal system must attend to and charts a path for legal scholars to follow for years to come. As with the dawning of fields such as juvenile justice, domestic violence, and elder law, early childhood development and the law will be a focal point for research within the legal academy, a vital bridge to scholars in other disciplines, and an important means for bringing lawyers and legal scholars to the heart of emerging policy debates.
Chapter
This paper deals with the mechanisms responsible for hand adjusting as they develop in the infant’s prehensive action. The questions are: Does seeing an object involve anticipating its haptic properties and is anticipating haptic properties sufficient for ajusting the hand to these object properties? The findings of several studies show that hand ajustment does not mature parallel with the accurate movement of the arm in the direction of the target; practice with handling objects is the most condusive to developing the skill of finger grip formation.
Article
Infants were presented with a moving object under 2 lighting conditions to investigate the role of vision in early reaching. The motion of the target object also allowed for an analysis of the infants' ability to use a predictive style of reaching. Infants were tested twice, at 5 and 7.5 months of age, with a moving object in the light and the same object painted with luminescent paint in the dark. Infants successfully contacted the glowing object on about half of their attempts at both ages, although 7.5-month-olds reached more often. Infants also took into account the motion of the target object by aiming their reaches ahead of the object and by reaching with their contralateral hand. These results suggest that proprioceptive feedback and sight of the target allow for successful reaching with limited visual information, even in relatively complex reaching tasks. The infants' success also demonstrates their ability to adapt their movements and reaching strategy to the speed and trajectory of the target object in order to reach predictively.
Article
Three experiments are reported addressing the issue of what kind of visual-proprioceptive information is picked up by young infants when engaging in the visual exploration of self-produced movements of their legs projected onto a video display. Results of the first experiment indicate that 3- to 5-month old infants show discrimination between a normal view of their own legs and a view that reverses the left-right location of both legs on the screen. Results of the second experiment show that there is no such discrimination when the overall featural configuration of the legs is occluded. In relation to Experiment 1, these results suggest that infants are sensitive to the invariant featural characteristics of the legs, in particular, how they relate to the rest of the body. Finally, results of the third experiment replicate infants' pattern of visual preference toward a view that reverses the seen and felt overall movement directionality, using a successive instead of a simultaneous visual presentation paradigm. These results provide further evidence of the early development of an intermodal calibration of the body. This calibration is interpreted as the perceptual foundation of the body schema and the early expression of an ecological self.
Article
Infants were presented with a moving object under 2 lighting conditions to investigate the role of vision in early reaching. The motion of the target object also allowed for an analysis of the infants' ability to use a predictive style of reaching. Infants were tested twice, at 5 and 7.5 months of age, with a moving object in the light and the same object painted with luminescent paint in the dark. Infants successfully contacted the glowing object on about half of their attempts at both ages, although 7.5 month-olds reached more often. Infants also took into account the motion of the target object by aiming their reaches ahead of the object and by reaching with their contralateral hand. These results suggest that proprioceptive feedback and sight of the target allow for successful reaching with limited visual information, even in relatively complex reaching tasks. The infants' success also demonstrates their ability to adapt their movements and reaching strategy to the speed and trajectory of the target object in order to reach predictively. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Infants between 26 and 32 wks of age were presented with sounding objects in the dark at 4 locations, 30° left and right, within reaching distance (15 cm) and out of reach (60 cm) at the same orientation. Infants' ability to localize the sound was assessed by measuring reaches of either hand into the space occupied by the objects when within reach. In 2 experiments infants reached more often into the correct location of the sounding object when placed within reach than when placed beyond reach. In the 2nd experiment the objects were presented at midline in the light and off midline in the dark. Infants reached correctly in the dark and showed no inclination to reach toward the midline position where they had handled the object in the light. By 7 mo of age infants have at least a dichotomous discrimination of auditory space: within and beyond reach. The implications of these behaviors are discussed in relation to object permanence and cognitive development. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Presents a mathematical model of the development of reaching that assumes that the major problem facing infants is their lack of lower level motor control and that infants learn to adjust their reaching strategies as a consequence of their previous experience and to match their current level of control. The model hypothesizes that infant reaches are a series of submovements, with the goal being to get the hand to the target in the face of errors in executed submovements. To relate actual infant reaches to this model, reaching data were decomposed into submovements, using a polynomial fitting algorithm that assumed minimum-jerk submovements. The model makes quantitative predictions about the course of development that are supported by existing results. The validity of the model's underlying assumptions was assessed by comparing the directional variability of the submovements with the variability assumed in the model. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
This book [reviews] cognitive neuroscience studies of the representation for actions. The fundamental question addressed concerns the nature and role of different representations in the planning and execution of movements. Adopting a cognitive neuroscience approach to this question generates a new perspective and some challenging hypotheses. The book explores in detail the contribution of the brain structures, particularly the cerebral cortex, to the various aspects of movement preparation and execution. In so doing, the author discusses a wide range of evidence, including the study of anatomical connections between areas, the recording of single neuron activity in animals, and brain stimulation and imaging studies in human Ss. This neuroscience evidence is related to both behavioral experiments in normal Ss and clinical observations in brain-lesioned Ss, resulting in provocative hypotheses about the cognitive structure of central representations and processes which subserve actions. (PsycINFO Database Record (c) 2012 APA, all rights reserved)