Poster group
Details of individual items:
poster
46etuses of all vertebrate species show spontaneous movement before birth.Although the motor activity of animal and human fetuses subjectivelyappears random, quantitative analysis has revealed that fetal movements areorganized in time and space. Moreover, the temporal and spatialorganization evident in fetal motor activity continues to be expressed byanimal and human infants after birth. One form of temporal patterning thatwe recently have reported is interlimb synchrony: the tendency for two ormore limbs to engage in nearly simultaneous movement. Rat fetuses, forinstance, show progressive improvement in the temporal coupling ofmovements involving different limbs during the last five days of gestation(Stansfield and Robinson, 1998, Infant Behavior & Development, 21:698).In the present study, we compared the prenatal development of spontaneouslimb activity in three species: Norway rats (Rattus norvegicus), spiny mice(Acomys cahirinus), and preterm human infants. Rats bear altricial young,which are born relatively immature, whereas spiny mice bear precocialoffspring. Both rodent species were studied using procedures that involvesurgical preparation of the pregnant mother to facilitate directvisualization of fetal behavior. To provide information during a comparablewindow of early development, preterm human infants born between 26-29 weekspost-conception were videotaped at weekly intervals in an NICU environment.Spontaneous motor activity was quantified by scoring individual limbmovements (B1 0.1 s) during playback of video recordings. These samples offetal and infant movement then were analyzed to characterize patterns ofinterlimb synchrony across a range of prenatal ages (gestational days 17-21in rats; days 26-37 in spiny mice; weeks 27-35 in preterm infants).All three species showed a significant tendency for limb movements to beorganized into discrete multilimb bouts. Bouts were defined as the set ofall consecutive limb movements in which the interval between successivemovements was 0.2 s or less. Clear bout structure was evident across agesin fetuses and infants. Analysis of bout structure suggested that movementof any limb significantly increased the probability of movement of anotherlimb, with each limb moving only once in each bout. This striking patternof temporal organization may be mediated by a relatively simple neuralmechanism. Evidence from animal research has demonstrated that isolatedregions of the fetal spinal cord exhibit spontaneous bursts of activity,with motor neurons projecting collateral fibers to other segments of thespinal cord. To test the hypothesis that bout organization is generated bya spinal mechanism, rat fetuses were prepared on E20 of gestation with acervical spinal transection, thereby isolating the spinal cord from thebrain. Transected fetuses continued to show spontaneous limb movements thatwere organized into multilimb bouts, providing support for the spinalmediation hypothesis. These comparative findings indicate that thedevelopment of spontaneous motor activity may exhibit very similar patternsof organization, and by inference, similar underlying mechanisms, in fetalrodents and preterm human infants.This research was supported by NIH grant HD 33862 and a CASSPR grant fromthe University of Iowa to SRR.
poster
Recent studies of spontaneous movements of infants have revised the traditional theory of motor development based on the newborn reflex and its inhibition through the cortical maturation. However, patterns of the spontaneous movements have not been characterized well in newborns and young infants and their functional roles in the motor development are still unclear. The purpose of our study was to elucidate the principle of developmental changes in the pattern of spontaneous movements by measuring and quantifying their complexities in terms of nonlinear dynamics. Subjects were 10 infants; 7 normal full-term infants, twin infants born pre-term, one of who was normal but the other was diagnosed as periventricular leukomalacia (PVL), and 1 infant who had midcerebral artery thrombosis. We observed spontaneous movements of each infant every month from 1 until at least 4 months of age (corrected age based on post-conceptual age in case of pre-term infants). Two-dimensional positions of four reflective markers, which were taped on each of wrists and ankles, were measured using a video camera and a computer with software for digitizing and processing of video images. We finally obtained epochs of spontaneous movements for 150 sec for each observation. As a probe of complexity of motor patterns, we calculated predictability of trajectories in a phase space that was constructed by embedding of the original time series. Statistical significance of nonlinearity was also examined using the method of surrogate data processing. We found evidence that the spontaneous movements of normal subjects were generated by nonlinear dynamics, which can be distinguished from linear processes and correlated noises. We also analyzed developmental trends in the motor pattern changes and detected U-shape changes in the complexity around 2 months of age for 5 infants out of 8 normal infants. Furthermore, movements of the 2 abnormal infants were characterized by loss of complexity; one showed too rhythmic pattern and the other showed a random one. From these findings, early motor development may be accounted by a mechanism of dynamic freezing and freeing degrees of freedom of the neuro-musculo-skeletal system with nonlinearity. To confirm these findings, three-dimensional measurement of motion of entire body is in progress.
poster
The sample included 90 infant twins participating in geneticproject aged from 10.2 to 11.5 months. Informationconcerning birth weight and gestational age was collectedfrom birth records. Additionally, from birth records indexof medical risk was derived. The index represents a singlescore that summarize overall medical risk taking intoaccount respiratory distress, feeding difficulty, prenatalhypoxia, asphyxia, wrong fetal presentation, low one-minuteApgar score, multiple embryogenetic stigmata. All infantswere tested using Bayley Scales of Mental and PsychomotorDevelopment.The infant's motor behavior was tested during objectretrieval task performance (Diamond, 1990) which is shown todepend on the maturity of prefrontal cortex. Then it wasanalyzed off-line using videotape records. Infants of nearlythe same age substantially varied in their performance ofmotor control task and were divided into 3 groups:'successful', 'intermediate' and 'unsuccessful'.ANOVA results revealed that although infants of three groupsdid not differ in chronological age corrected for the periodof gestation those who failed to accomplish the OR task hadsignificantly lower birth weight comparing to their moresuccessful peers. There were also highly significant groupdifferences in gestational age. The inspection of relativemeans showed that the period of gestation was significantlyshorter in unsuccessful group comparing to both successfuland intermediate groups.The intrauterine variables associated with lower birthweight may be the intrauterine malnutrition or some otherprenatal complications. In fact, unsuccessful infants may bemore compromised then the other newborn twins and thisinfluence their subsequent development of motor controlfunctions. To check this suggestion indexes of medical riskat birth of all groups were compared. ANOVA results revealedthat the group factor was significant and unsuccessfulinfants were actually at higher risk for medicalcomplication during pre- and perinatal developmental period.Thus, despite the selection of healthy twins for our study,seemingly minor but cumulative medical complicationsappeared responsible for differences in maturity ofexecutive motor control in 11-month old twins. However, thefactor of subtle developmental abnormalities explains onlythe relative immaturity of motor control functions inunsuccessful infants but is of no importance for differencesbetween other two groups.The ANOVA for PDI and MDI scores showed that factor 'group'was significant and successful infants displayed higher MDIcomparing to both intermediate and unsuccessful infants.Despite the similar tendency in these two groups toward thelower mean values on the psychomotor scale in contrast tobetter performers, the effect of group on PDI did not reachsignificance level. Thus, advanced maturity of motor controldomain in successful infants was associated with higherlevel of psychomotor development. This data suggests thatgood performers of motor control task tend to have a fasterrate of an overall psychomotor development.The results of study point the two independent sources ofindividual variations in motor control task performanceamong 11-month old infants. Differences in the maturity ofmotor executive functions may result from subtledevelopmental abnormalities or may be linked to differencesin a maturational rate.
poster
Actions in everyday life are performed with linkages between perception andaction systems, as between vision and hand movement. The main question is tounderstand how they develop together or how the relations change across time.It has been shown however that vision influes on this movement : von Hofsten(1982) demonstrated that the gaze on the object trained a more proximalstep ofthe hand. Then Ennouri and Bloch (1996) showed, in neonates, that the units ofmovements, which begin under visual fixation, are clearly more directed towardthe object than units initiated without fixation.This study aims to estimate how the relation between visual fixations andmanual behaviours (approach, pre-reaching and reaching) develops from2-to-6-months-old infants.Parameters of oculomotor functioning and manual movement are considered fromtwo video records. A videocomputer interfacing software permits to codingmanual movement organization (as arm position and type of hand movements).The results shown that the sequence fixation followed by reaching appearedfrom4 months and increased with age. At 5 months, temporal organization betweenvisual fixation and reaching appeared closer as adults.
poster
Goal-directed movements are performed after a latency, called Reaction Time(RT). This time is supposed to be used simultaneously for processingperceptual(generally visual) information and for planing the movement. This hypothesisbetween perception and action suggests that the longer the RT, the shorter theMovement Time (MT). Such a correlation was examined in infants' normal actiontoward a distal object. Data come from a transversal study, conducted withinfants from 4 to 7 months. RT and MT were first considered separately, thenthe RT/MT correlation was calculated for reaching movements with respect tothehand used. The expected negative correlation was found for the right handreaching, in the youngest infants. On the contrary, the correlation for theleft hand reaching is positive (the longer the RT, the longer the MT). Aninterpretation of these changes is presented in the final discussion.
poster
Infants' arm movements before reaching emerges are occasionally spatiallyrelated to a nearby, fixated object (von Hofsten & Ronnqvist, 1993; Ennouri& Bloch, 1996). Two important questions raised by these results are: whatare the spatial characteristics of pre-reaching arm movements, and arethese movements related to reaching? Our objective was to determinewhether, and if so, how, infants' arm movements with respect to a nearbyobject change as reaching nears.Methods. Infants' (N21) hand positions and eye movements were recorded at8, 12, 16 wks then every 2 wks until reaching was observed. While seatedin a custom chair reclined 30 degrees from vertical. Each infant waspresented with 1 Toy, 2 Toys or No Toy. The small (5-7cm), illuminatedtoys were presented midline, at chin height, 12-16 cm from the baby. Ineach four, 4-min trial, No Toy or 1 Toy was presented for 1 min, thenanother toy was added for the remaining 3 min. Individual infants werepresented with the same trial sequence at every visit.Arm movements with respect to the toy were measured by changes in handposition using a passive reflective marker. The marker was tracked in 3D(30 Hz) by a custom-built optical tracking system. The corneal reflectiontechnique was used (frame-by-frame) to determine fixation on and off thetoy.Analysis. Data from 7 infants at two ages before reaching (Age 1, 6-12 wksbefore; Age 2, 1-4 wks) were available in 3 fixation conditions (for 1Toy): visual fixation ON the toy, fixation OFF the toy, and No Toy.Movement (right hand) measures were: average distance between the infant'shand and the toy; minimum distance between hand and toy; variability of thedistance between the hand and toy; number of times the hand moved closer tothe toy; and orientation of hand movements with respect to toy positionwhen the hand moved closer to the toy.Results. (1) At Age 1, infants' minimum hand-toy distance was greaterduring OFF than during No Toy (p.037). (2) At Age 2, the variability inhand-toy distance was greater during ON than during OFF (p.047). (3)Correcting for the amount of movement, at Age 1, infants' hands movedcloser to the toy more often during OFF than ON (p.02). At Age 2,however, infants' hands moved closer to the toy more often during ON thanOFF (p.03). (4) At Age 2, when the hand moved closer to the toy, thedirection of movement was more oriented to the toy position during ON thanduring OFF (p.046).Conclusions and Speculations. Differences in pre-reaching arm movementsduring infants' fixation on versus off a nearby object may reveal gains inarm control. Results suggest that infant's directional arm control mayprecede distance control, and that the spatial coordination observed inpre-reaching infants may be discontinuous with reaching. Furthermore, thespatial variability between a pre-reaching infant's arm movements and anearby object during sustained attention may promote motor learning.
poster
The control of movement by infants has become a central topic in the study of perceptuo-motor development. In recent work, the infants' abilities to correct and anticipate the parameters of their movements have been questioned. In this study, we examine the development of the ability to control the force in a tracking task. With this aim, we used a display which consists of a connection between a computer, a balance, a big doll with lights and sound and a ring. If the child wants to switch on the doll's lights and the music, he or she has to pull from the ring that is in front of the doll hanging from the ceiling. The experimenter can manipulate the force the child has to pull with to switch on the doll. This is done through the computer and the balance. The computer also records the force exerted by the child every second in terms of its equivalent weight in grams. The experimental design consisted of a progressive reduction of the range of force needed to activate the doll. We used four experimental conditions: 50-330 grs, 80-300 grs., 110-270 grs. and 140-240 grs. Participants were 30 children in three age groups: 12, 17 and 21 months. The direct measure was the weight in grams equivalent to the force applied. From this measure, we got derived measures as success in the task, amplitude of movement, search of the thresholds, and so on. Data were analysed through a GEE (Generalised Estimating Equation) analysis. Results show that, although children from all age groups were able to control the force to switch on the doll, the way this control was performed was different according to the child's age. Young children show a stereotyped, rigid pattern of movement in which they tend to maintain their parameters once they got their goal. When the range changed, they would stay in the previous range for a while before changing their performance. They could also stay for long periods without switching on the doll, showing difficulties to correct their movements. In contrast, the oldest children in this experiment show movements with a wide amplitude and they tend to vary along all the range of force. They explore the upper and lower thresholds switching on and off the doll at their will. When they made errors, they would correct them quickly. Then, it seems that even when children are able to control a movement to perform a task successfully, the way this control is exerted changes with age and with other abilities implied in the task.
poster
PurposeWe measured the effect of changing a property of the target object immediately relevant to the online control of action (object size) and a property irrelevant to action control (visual pattern) upon infants' looking preferences, reaching preferences and movement kinematics. The aim was to examine how far the systems controlling infants' looking can be manipulated independently of the visual information processing which determines reaching.MethodsWe presented infants aged 5-17 months with pairs of objects - a small 'graspable' cylinder paired with a cylinder too large for the child to grasp. In experiment 1, the large object was plain. In experiment 2, the large object was either plain, or had a schematic face drawn on it. The direction of the first look and the object first touched were recorded, and the transport of the hand tracked with an Elite motion tracking system. Four infants were also studied longitudinally, between the ages of 5 and 12 months.Results In experiment 1, younger infants (5-8 1/2 months) show a looking preference for the larger object of the pair but no reaching preference is observed at this age. The looking preference for larger objects is not present in infants over 8 1/2 months. However, a reaching preference for smaller objects develops around 8 1/2 months, indicating a developing ability to use visual information to predict the 'graspability' of objects. Kinematic data suggest that the onset of object-oriented action requires a slowing down of the reach, and an extended 'homing-in' phase.In experiment 2 a looking preference for the face object over the small object was maintained across all age groups, whereas a looking preference for the large plain object over the small object was only observed in younger infants (as in experiment 1). Children over 30 weeks showed a reaching preference for the small (graspable) object, as in experiment 1. However, this preference was reduced when the face object was presented with the small object. The longitudinal studies have a similar pattern of results.DiscussionThere is an early preference for looking at the largest object in the visual field, which falls away around 8 1/2 months. A preference for reaching to graspable objects develops between 7-8 1/2 months. Looking and reaching preferences diverge during the development of reaching, particularly when a non-preferred object for reaching is made visually attractive. This divergence may reflect a dissociation during development of visual processing streams subserving object-related action from those related to visual orienting .Supported by Medical Research Council programme grant PG7908507 and a BBSRC studentship
poster
There has been an increased interest in the study of spontaneous movementsof infants (Heriza 1991, Piek 1994, Thelen 1993, Turvey & Fitzpatrick1993). A reinforcement protocol has been used to look at how infantsmodify their spontaneous leg movements to obtain the movement and sound ofan overhead mobile (Rovee & Rovee 1969, Thelen 1994). Unfortunately, fewresearchers have focused on the influence of arm movements in a task thatrequires leg movements. The goal of this study was to examine therelationship between arm and leg movement frequency and level ofperformance in infants at 3 months of age. We hypothesize that infantswho produce more spontaneous leg movements will gain a larger number ofreinforcements than infants who have more and better controlled armmovements when given a leg movement task to move the mobile. Eightinfants (89-109 days of age) were placed in a supine position under amotorized mobile. Sensors were placed on the infants' knees, hips, andleft shoulder to measure their joint movements. The testing sessionlasted 12 min: 2 min in baseline (no reinforcements), 8 min in acquisition(where the mobile moved each time the infant produced a left knee or hipextension), and 2 min in extinction (no reinforcements). Frequency of armand leg movement units (MU) was determined by customized software.Developmental level of arm control was established using 25 arm-relateditems from the Bayley, Peabody, and MAI scales. Arm MU, leg MU, andreinforcement frequency were plotted against time for each infant and as agroup. Both arm and leg MU increased their frequency with time. However,only leg MU maintained a high frequency after the reinforcement wasremoved in the extinction period. Arm control scores and arm and leg MU inthe baseline were separately compared to the maximum reinforcementfrequency in the acquisition phase. As expected, leg MU at baseline showeda direct relationship with peak performance. Contrary to our hypothesis,there was also a positive correlation between the number of reinforcementsin the peak acquisition and the arm control scores and arm MU in baseline.These results indicate that at this age upper and lower extremities workin a synergistic fashion and therefore, the overall spontaneous movementin either upper or lower extremities is a good predictor of performance.Overall, these data suggest that infants who spontaneously explore themost are the ones performing better.
poster
Sensori-motor development, such as the emergence of reaching, becomes againa central topic in developmental psychology. Various theories were advancedto explain motor development by a maturational or a cognitivist approach.Motor abilities would emerge in infants either only from the maturation ofthe neuro-muscular system -this maturation being genetically programmed-either from the development of mental structures.Currently, the non-linear system theory, represented by Thelen indevelopmental psychology, disputes the relevance of these two theories toaccount for the development of motor behaviour. According to Thelen, theseprescriptive theories do not explain the origin of motor abilities. Thedynamic approach postulates that emergence of sensori-motor pattern isdetermined mainly by intrinsic properties (biomechanical, energetic -such asthe muscular force of the limb, its proportions...-) and extrinsic to theorganism (i.e environmental). The interaction of these components leads tothe emergence of stable behavioural pattern (in dynamics, a stable state) bya process of auto-organisation.In order to complete this theory, the behaviourist approach is applied tothe development of motor abilities. This approach is based on a principle oforganism-environment interaction. It postulates that the environmentalconsequences select behaviours that organism emits. Motor development isconsidered under the control of environment: some events of the environment,which are reinforcing, would control the acquisition and the maintenance ofmotor behaviour by an operant conditioning mechanism. Our study aims atdetermining the environment's role on the acquisition of motor behaviours.The emergence of reaching in neonates was selected to test this assumption.Infants from 3 to 5 months, for which reaching behaviour is not acquiredyet, are used as subjects. Each baby was placed in an experimental device inwhich the localisation of one hand at a precise point of space wasreinforced by the arrival of a sound stimulus. Infant sat in semi-tiltedposition. A precise point of space was defined, in front of the baby, at adistance equal to the length of his arm. It corresponds to theÔreinforcement point': when the hand of the baby passed by this point, musicappeared and lasted as a long as the baby's hand remained at the same place.A pre-test shows that the music reinforces the babies' arm movements:response rate or duration of these behaviours at the Ôreinforcement point'increases. A reinforcing role of the environmental stimulus on theacquisition of reaching, and generally on the development of motorabilities, may exist. Results, considering other indicators (such as thevelocity of the hand, and the kinetics of the movements) are currentlyprocessed in order to test thoroughly our assumptions.
poster
INTRODUCTION: In adult reaching, joints are coordinated such that thehand's path is relatively straight with a bell-shaped velocity profile,and little or no overshoot. Shoulder and elbow excursions are alsotypically smooth, direct and accurate. In contrast, the infant's firstreaches at 3-5 months of age are neither direct nor smooth, but show awinding hand path with many velocity changes. Over the first year, thepath becomes markedly more straight, smooth and accurate. There is littleknown, however, about how the joints are coordinated to produce theseimprovements in path.The purpose of this study was to characterize the pattern of directness,smoothness and accuracy at the shoulder, elbow and wrist across the firstyear. We addressed two sets of questions. First, what is theorganization of joint movements in infant reaching and how does thisorganization change over the first year? As infants fashion goal-directedreaching from earlier, non-reaching movements, the second questionwas--what is the organization of joint movements before infants learn toreach, and how do these patterns change when reaching emerges?METHODS: The position of the shoulder, elbow and wrist were recorded fromsix infants (seated with trunk secured) during reaches to a toy placed inthe midline at chest height. Infants were seen weekly from 3 to 30 weeksof age and on alternate weeks from 32 to 52 weeks. For the reachingportion of the year, we determined three variables at each joint: 1)directness, the number of joint reversals, 2) smoothness, the number ofjoint velocity peaks, and 3) accuracy, a ratio of the maximum jointexcursion to the excursion from initial to final position. To comparereaches and pre reaching movements, we determined the number ofreversals/sec and velocity peaks/sec for the pre reaching and reachingportions of the first year.RESULTS: Infant reaches displayed a consistent proximal to distal patternin which the shoulder was more direct, smooth and accurate than the elbow,and the elbow more than the wrist. This pattern was maintained as infantsimproved control in all three joints. As such, infants produced adultlevels of shoulder control well within the first year. Interestingly,infants displayed this pattern well before the onset of reaching.Moreover, the transition from pre-reaching movements to goal-directedreaches was marked by a sharp decrease in reversals/sec and peaks/sec.CONCLUSIONS: 1) Infants reaching with their trunk stabilized showdifferential characteristics at the three joints in both pre reaching andreaching movements. 2) The more massive and proximal shoulder appearsbetter controlled of the three joints during reaching, and the most stableduring pre reaching movements. One likely reason is that the elbow andwrist are more affected by intersegmental dynamics than the shoulder(Galloway and Koshland 1997). Thus, as the shoulder joint moves, itsforces are reflected in the more distal joints. 3) It appears as thoughinfants are initially unable to stabilize these distal joints againstshoulder movements. Improvements in reaching skill result from learningthese and other features of the intersegmental dynamics of the moving arm.4) Goal-directed reaching does not arise de novo, but is fashioned withinthe biomechanical constraints of the multisegmented limb.