Poster group
Details of individual items:
poster
In order to view clearly an object which is at a different distance indepth from the fixation distance, the eyes have to rotate to point at theobject (vergence movement) and the lens has to change shape to focus onthe object (accommodation). In adult humans, when one eye is covered sothat vergence eye movements are not required, a change in accommodation isstill accompanied by a vergence eye movement in the appropriate directionbut of smaller size. In this study, we looked at the development of thelinkage between accommodation and vergence in human infants from birth to1 year of age.133 healthy infants from 2-52 weeks of age were tested in this study.Accommodation and vergence were measured using a remote haploscopicphotorefractor based on the photorefraction technique of Abramov et al,1990. In the design of the remote haploscope, there is a separationbetween the optical path of the camera that views the infant's eyes, andthe target. This allows us to modify the optical pathway between theinfant and the target while the camera has an uninterupted view of theeyes. In this study, we used the remote haploscope to cover (occlude) oneeye thus removing the normal retinal disparity cues which are used by thevergence system in adults to rotate the eyes to point at the target. Blurcues were available to the viewing eye to stimulate accommodation. Anyvergence found when viewing targets at different depths must be driven byblur cues or by information about the proximity of the target. Thislinkage is called the accommodation convergence/accommodation (AC/A) ratioand is defined as the amount of convergence (in metre angles: MA) producedfor each dioptre of accommodative response.Results in normal adults replicated previous findings of a linkage betweenaccommodation and vergence in this situation. When adults were occluded,the AC/A ratio was found to be 0.71 which is in line with previous data.However, although infants showed normal responses for both accommodationand convergence when viewing with both eyes, their response when occludeddid not resemble adult responses at any time within the first year.Instead, both accommodation and vergence responses became flat (Table 1:slopes of the linear regression functions fitted to the accommodation andvergence data). Possible explanations for this are that the target doesnot have a strong enough blur cue to drive accommodation and so there isno drive from accommodation to convergence; or that the infant stopsattending when one eye is covered. However, consideration of thedifferences in the plane of accommodation and vergence rule out theseexplanations. Table 2 gives the intercepts of the linear regressions foraccommodation and vergence for both the unoccluded and the occluded caseat each age measured. It can be seen that when the infants are unoccluded,all but the youngest infants have intercepts close to 0 dioptres foraccommodation and 0 MA for vergence. However, when the infants areoccluded there is a dissociation between accommodation and vergence: Theintercept for accommodation is close to 0.5 dioptres while the convergenceresponse has an intercept of 2.5-4 MA. Thus, accommodation cannot bedriving the vergence response in the age group of infants tested here. Weconclude that infants younger than 1 year of age do not have an adult likelinkage between accommodation and convergence. We are currently testingpre-school children to determine when this link develops. Abramov, I., L. Hainline, Duckman, R.. (1990). Screening infant visionwith paraxial photorefraction. Optometry and Visual Science 67: 538-545. Age (weeks) Unoccluded Occluded Accomm Verg Accomm Verg AC/A 0-8 (n23) 0.52 0.38 0.31 0.06 0.199-12 (n33) 0.75 0.60 0.10 0.15 1.513-16 (n25) 0.73 0.95 0.24 0.38 1.5817-26 (n19) 0.92 0.93 0.15 0.19 1.2727-52 (n33) 0.55 0.75 0.21 0.48 2.29Adults (n13) 0.67 0.95 0.45 0.32 0.71Table 1 - Slope of the linear regression functions fitted to accommodationand vergence data when measured with both eyes viewing (unoccluded) andwhen only one eye could see the target (occluded). The AC/A ratio givesthe ratio of the slope of the vergence to the accommodation response.Age (weeks) Unoccluded Occluded Accomm Verg Accomm Verg0-8 (n23) 1.06 2.38 1.21 3.719-12 (n33) 0.03 0.93 0.80 3.1313-16 (n25) -0.08 -0.28 0.42 2.4117-26 (n19) -0.26 0.13 0.37 2.5727-52 (n33) 0.00 -0.19 0.57 3.51Adults (n13) -0.11 -0.14 0.14 0.46Table 2 - Intercept of the linear regressions fitted to the accommodationand vergence data when measured unoccluded and occluded. Accommodation ismeasured in dioptres and vergence in metre angles (MA).
poster
Infants visually guided head and trunk movements was studied by examiningtheir ability to track an object moving in a circular motion path (40cm/s), vertical presented relatively to the infant. The main question washow do infants organize and control their head movements, and can theypredict this kind of complex motion path, as described from liner tasks. A3-D opto-electronic measurement system (QUALISYS ProReflex MCU) was used tocollect kinematic data of head and trunk movements. Twelve infant's 6- and12 month old were presented with 12 trials each. The object was moving 1.5revolution, either in a clockwise or counter-clockwise direction. Inaddition to the infant studied, a group of six adults was tested, servingto establish an adult criterion for the development. It was found thatadults mainly use eye movements when tracking the object motion path, withjust small, efficient head movements, well timed with the object motion, analmost no trunk movements. All infants were also able to track the movingobject. However, non-of the infants seemed to be able to predict thecircular object motion path, instead, they utilized a strategy of chasingand catching up. The 12-months showed more efficient head movements thandid the 6-month old did. I.e. shorter latency, relatively smaller headmovement and a shorter time lag between the head and the object. However,both infants groups were using compensatory trunk movements relatively totheir head movements, even if the 12-months showed a better posturalcontrol. No overall effect of object direction was found on any of theparameters studied in the infant groups. The preliminary conclusion fromthis study is that development goes from mainly using the trunk and headcontrol, to mainly using eye movements when tracking a complex motion path.Furthermore, the developmental changes of head movement segments andunderlying mechanisms will be discussed in relation to earlier studies.
poster
By 3 months after birth there is a reliable pattern of phasic motor activationthat immediately precedes the spontaneous disengagement of overt visualattention. When attention is interrupted by a distracter, the phasic motoractivation is superimposed on motor quieting which begins before the infantlooks away. Taken together, these findings suggest that central neural activityassociated with the transient motor activation may increase theinterruptibility of attention, facilitating disengagement. In the presentexperiment we examined the coupling between attention and motor activation at 1month, when existing behavioral and neural evidence suggests that thedisengagement of overt visual attention is more difficult. Infants (1 mos, N 3D10) were seated 90 cm in front of a black felt screen on which two brightyellow toy animals were mounted. The stimuli subtended 11 degrees of visualangle and were separated by 24 degrees. In Distracter events, when the infanthad been looking at one stimulus for 2 sec, the other stimulus began to rotate45 back and forth (1.25 cycles/sec) perpendicular to the infant's line ofsight, and continued to rotate for 1 sec after the infant looked at it or until10 sec had elapsed. Control events were identical to Distracter events exceptthat the distracter was never activated. Distracter and Control events wererandomly mixed. Corneal reflections of the stimuli were detected by a videocamera mounted behind the screen midway between the stimuli. Motor activity wasdetected by piezoelectric sensors mounted in the seat and sampled at 60 Hz,synchronized with the camera. All data analysis was based on the videotapedcorneal reflections. For each event, movement sensor output (exceedingthresholds set to exclude respiratory movements and noise) was rectified andaveraged in 100 ms bins before and after the point at which the infant lookedaway from the first stimulus. The first- and second-order rates of change inmotor activity were calculated from the binned data. Results from the multipleevents from each infant were averaged before the group analyses were conducted.In Control events, there were two phasic accelerations in motor activity(d2z/dt2 > 0, p < .05) before the disengagement of overt visual attention. Thefirst occurred approximately 900 ms and the second approximately 300 ms beforeinfants looked away. In Distracter events, no pattern of coupling between motoractivity and the disengagement of overt visual attention was evident in thesample of 10 infants. However, in the 6 infants that provided direct evidencethat the distracter had interrupted attention to the first stimulus (i.e., thelatency to look away was shorter for distracter vs. control events, p < .01),there was a single phasic acceleration in motor activity (d2z/dt2 > 0, p < .05)approximately 700 ms before the disengagement of overt visual attention. Therewas no evidence of motor quieting in response to the onset of the distracter inthe full sample or the subset of 6 infants. Thus at 1 month, as at 3 months,there is a phasic acceleration in motor activity immediately preceding thespontaneous disengagement of overt visual attention. However, at 1 month thereis an additional phasic acceleration nearly a full second before looking away.Furthermore, at 1 month (but not 3 months) the coupling between motor activityand the spontaneous disengagement of overt visual attention is disrupted by anexternal distracting stimulus and there is no motor quieting following theonset of the distracter. The early phasic motor activation (700-900 ms beforelooking away) may be associated with the disruption of attention, and the lateactivation (100-300 ms before looking away) with the successful disengagementof attention. By 3 months after birth, these events may occur simultaneously,permitting faster shifting of attention.
poster
During mother-infant interaction, the infant's focus of attention changeswith age, as described by Trevarthen (1978) : specifically infants becomemore concerned by inanimate objects. However, the way objects are presentedto infants has an impact on the regulation of visual attention : Eppler(1995) investigated how objects are manipulated, and visually explored in 3-to 5-month-old infants. She showed that while manipulation increases withage, the duration of visual exploration does not change. Hand-eyeco-ordination develops around 5 months, and the infant is more interested inwhat his own hands are doing than in inanimate objects. Moreover, infants'orienting of visual attention depends on posture, at least duringmother-infant interaction (Fogel et al., 1992).The present research aims at investigating the dynamic organization ofattention depending on posture and the way objects are presented. The experimental device makes it possible to change the infant's posturewhile spatial relations between visual stimuli and the infant's eyes arekept constant. Each infant is given two 2 minutes trials, one in eachposture while 3 dimensional stimuli are presented in pairs. Visual activityis videorecorded and the following indices are derived : number and durationof fixations on each target, number of shifts from one target to the other(alternations), and number of saccades back to the previous target(repetitions).In a first study infants cannot reach the objects : visual attention is notsustained, in any posture (semi-recline or sitting). In a second study,targets are presented within prehension space, in both postures. Theninfant's visual exploration of objects lasts longer, and shifts betweenobjects are more frequent when infants are in the more erect position thansemi-supine.
poster
no abstract
poster
The developmental path that leads to the achievement of upright independent posture and walking offers a rich landscape for the examination of the development of perception-action coordination. In recent work (Barela et al., 1999), we have developed a paradigm that allows us to systematically examine this coordination. Based on adult work (Jeka et al., 1997) in which light fingertip touch has been shown to provide critical information to adults' postural control system, we created an analog in which infants touch a rigid rounded bar as they stand upright on a short pedestal. Infants from the onset of pull-to-stand until 1.5 months of walking experience were shown to use the contact surface as a source of sensory information to stabilize their posture (Barela et al., 1999). In the current study, we extend this work to infants who have had up to one year of walking experience. Our purpose is to examine the role of somatosensory information in postural control as infants have increasing experience in the upright. Thirteen infants (6 males, 7 females) with 1-12 months of walking experience stood on a small pedestal either lightly touching (< 10 N) a rounded bar or not. Three-dimensional kinematic data were collected on the infants as they stood quietly. The motion of three body segments was digitized using the PEAK motion analysis system including the head, shoulder (C7 vertebral level) and the hip at the approximate center of mass (CoM). Mean sway amplitude was the primary measure of postural sway and cross-correlation analysis was used to examine the motion of the different body segments in relation to one another. Evidence from the mean sway amplitude suggested that, unlike the results found for adults (Jeka & Lackner, 1994), light touch did not attenuate the postural sway of the infants. Examination of the cross-correlation coefficients between the head and CoM, the head and shoulders and the shoulders and the CoM revealed why we did not replicate the adult findings. In the infants, across all ages, the segmental coupling in the touch trials was lower compared to the no-touch trials (p < .01). We interpret these data as evidence that infants are using the sensory information from a contact surface to explore possible means of coordinating the degrees of freedom within their bodies during upright stance. While standing independently (without touch), the infants are constrained by the balance demands of the task, and thus, freeze their segments (i.e., their segmental degrees of freedom) into an inverted pendulum. However, when provided a touch surface, the infants use the added somatosensory information to explore other modes of postural control and coordination. Indeed infants may be exploiting their degrees of freedom in order to explore the relationship between multiple sources of sensory information and the resulting action of their bodies.