Poster group
Details of individual items:
poster
The improvement in the ability to avoid the distractionaction of the immediate sensory environment and inhibitinterference is supposed to be critical for thedevelopmental progress in attention control in infantduring the second half-year of life. The current study isdesigned to investigate the role of cortical inhibitoryprocesses indexed by alpha synchronisation in thesustaining anticipatory attention in infants. EEG wasregistered in 60 infants aged 8-11 months under twoexperimental conditions: attention to an object in thevisual field (baseline) and anticipation of the person inthe peek-a-boo game (anticipatory attention). The mainresult of the study is the specific relation betweenposterior alpha synchronisation and maintenance ofanticipatory attention. The infants who maintained theanticipatory attention for longer time had higher alphasynchronisation over the posterior parietal region duringthis state. This effect was found selectively at 6.8 Hzfrequency, which is close to the peak frequency ofposterior alpha rhythm during this age period. Thesefindings support the hypothesis that cortical inhibitoryprocesses that are reflected by selective alpha parietalsynchronisation are important for maintaininganticipatory attention. Given the neuropsychologicalevidence on the role of posterior parietal cortex in theattention to periphery of visual field, we hypothesisethat the activity of posterior parietal networks engagedin peripheral attention should be suppressed in order toprevent shifts of attention to the surrounding objectsduring focused anticipatory attention.
poster
A new neurobiological model of infant visual development is proposed,building on Atkinson's (1984) model of the developing interaction ofsubcortical and cortical streams of visual information processing. The newmodel (Atkinson, in press) incorporates (a) the distinction between ventral (object recognition) and dorsal(spatial relations and control of action) streams established by adultneuropsychology, functional imaging, and non-human primate studies;(b) the idea of discrete visuo-motor modules within the dorsal stream,linking parietal and frontal lobes and associated with different actionsystems such as eye/head, reach/grasp, and locomotion(c) the role of spatial selective attention, involving both cortical andsubcortical processing, in modulating and integrating these visual actionsystems.(d) the requirement to integrate different action and attentional controlsystems to achieve the major changes in functional behavioural capabilitiesin the first years of life.This model will be applied to three groups with abnormal development ininfancy: infants with perinatal brain lesions identified by serialstructural MRI (N>100), young hemiplegic children (N22) , and youngWilliams Syndrome children (N>100). All groups show greater deficits indorsal stream functioning (tested in terms of motion processing) than inventral-stream form processing. This deficit may resemble that reported ina milder form in some dyslexic children. Tests that are specific to dorsal stream function and to specificvisuo-motor modules may be valuable in specifying not only the effects ofabnormal development, but also in identifying at what stage development hasdeviated from its normal course. Such information about developmentalcourse may also allow us to specify more precisely the time windows ofopportunity for intervention and treatment.References: Atkinson, J .(1984). Human visual development over the first six months oflife. A review and a hypothesis. Human Neurobiology, 3: 61-74.Atkinson, J. (in press) The Developing Visual Brain. Oxford UPSupported by MRC programme grant G7908507
poster
Neuroimaging and neuropsychological studies have shown that there areparticular areas within the human temporal cortex that are activated byfaces (the fusiform face area; FFA) (Kanwisher 1997). The general questionwe are investigating is how such specialised neural processing developsduring infancy. Electrophysiological studies (ERPs) have revealed a face -sensitive component that peaks between 120 96 200ms after stimulus onset (theN170). The N170 is thought to reflect activation of neurons within the FFAand/or surrounding temporal cortex. Using ERPs Bentin et al (1996) showedthat inverting or scrambling face stimuli produced components of slowerlatency and different amplitude. In terms of stimulus complexity, uprightand inverted faces pose identical demands on the visual system, thus this istaken as further evidence for the existence of a specialised neural systemfor face processing.Previous high density event-related potential (HD-ERP) studies from ourlaboratory have shown that there is an effect of inverting human faces withinfants and adults, however, the latency and morphology of this effectchanges with age. In adults, the N170 over occipito-temporal sensors was oflarger amplitude and peaked more slowly for inverted than for upright faces.12 month olds also show a larger negative component to inverted faces but ata later latency (around 290ms). In 6-month olds upright faces elicited agreater positivity over occipital electrodes peaking at approximately 350ms.In order to ascertain the specification of these responses to human faces wehave also examined responses to monkey faces in adults and 6 month olds.These studies have shown that in adults there are no effects of invertingmonkey faces on theN170, indicating that there is specific processing ofupright human faces. In contrast, 6 month olds show a similar effect withmonkey faces as they showed with human faces, suggesting that their faceprocessing is more broadly tuned than that in adults and is not yet speciesspecific. Therefore, the present study sought to explore the ERP responsesof 12 month olds to upright and inverted monkey faces to establish whethertheir processing of faces has become as specialised as that observed in adults. The results to date (N3D7) indicate that 6/7 subjects show greater negativityaround 290ms at right occipito-temporal sites for inverted compared withupright faces. Since this work is still in progress it is difficult to drawany firm conclusions at present. However, with a larger sample size it willbe possible to state whether by 12 months of age infants have developedspecialised face processing systems that are more similar to that of adultsor whether they still operate on a more broadly tuned system like infants ofyounger ages.
poster
Visual orienting, driven by subcortical structures, is one of the fewrelatively mature motor systems available to the newborn. Nevertheless,the control of eye-movements goes through further development during thefirst year of life. One aspect of this development is thought to be aprogressive corticalization of control, during which the eye-movementcentres in the cerebral cortex gain more and more influence over visual orienting.Several recent studies, including experiments which recordedsaccade-locked event-related potentials (ERPs), have demonstrated thatthe parietal areas of eye-movement control are not fully functioning at6 months of age. A direct indication of the weak parietal involvement at6 months of age is the complete lack of the characteristic parietalspike potential (SP) which precedes the execution of saccades in adults.In the present study we aimed to look for the SP in 12-month-oldinfants, the age when eye-movements are more likely to be controlled bythe parietal cortex.We recorded high-density ERPs in ten 12-month-old infants while theywere making horizontal saccades to target stimuli. The fixation stimuluseither went off 200 millisecond prior to target presentation ( gap trials) or stayed on during the eye-movement ( overlap trials). Asexpected, the infants showed a sizeable gap-effect : their saccadicreaction times in the gap and overlap trials were 226 msec and 310 msec,respectively. More importantly, the saccade-locked ERPs included a smallbut reliable spike potential peaking at about 20 msec before saccadeexecution. Furthermore, there is preliminary evidence for more corticalactivity prior to eye-movements in the overlap trials, a result which issimilar to the adult ERPs recorded during the same task.These results suggest that over the second half of the first year theparietal cortex goes through a considerable development and by 12 monthsof age it plays a similar role in eye-movement control as it does in adults.
poster
Many studies have investigated the role that motion cues play ininfants segmentation the visual world. Much less is known about howstatic cues, such as subjective contours, help the infants to findobjects in the visual scene when faced with incomplete information.While some studies suggest that 3- and 4-month-olds perceive thesubjective contours of a Kanizsa square (Ghim, 1990), others found thatit is not until 7 months of age that infants are sensitive to thisillusion (Bertenthal et al., 1980). Here we report the first steps of anattempt to approach this question by recording brain electricalresponses to subjective contour figures in adults and infants.Twelve adult participants watched passively briefly presented Kanizsasquares and several control figures while we recorded high-densityevent-related potentials (ERPs) from their scalp. The Kanizsa squareselicited an enhanced N1 response about 150 msec post-stimulus over theoccipital region, which suggests the involvement of the visual cortex inthe generation of the subjective figure. The Kanizsa squares alsoproduced a stronger P2 (200-240 msec) over the temporo-occipital cortexand this difference was more enhanced on the right side. Comparison withthe ERPs to control stimuli showed that these differences cannot beattributed to other sensory factors, such as spatial frequency distribution.We have been conducting a similar study in 8-month-old infants. So far 4infants completed the experiment successfully. Preliminary analyses haveshown ERP effects similar to adults in that the Kanizsa squares eliciteda positivity beyond 200 msec post-stimuls, and this difference was morepronounced over the right hemisphere. The presence or absence of theearlier effect (N1) is not yet clear. These results suggest that atleast partly similar neural structures contribute to the perception ofsubjective contours in infants as in adults.We plan to extend the age range of our participants to younger infantsto investigate the development of these neural structures. We also planto apply time-frequency analyses to our data which have been shown to beuseful in uncovering the neural correlates of construction of objectrepresentations (Tallon-Baudry et al., 1999).
poster
While there is strong evidence for frontal lobe involvement in studies ofadult attentional abilities and memory processes, only recently have webegun to consider this cortical area as an essential element of infantdevelopment. This study was associated with a program of research aimed atexplaining individual differences in the roles of the frontal lobe ininfant cognitive development, the precursor to higher order cognitivefunctioning in children. The purpose of this specific study was to examine individual differencesin the development of performance on two versions of an object permanencescale. One version was assessed using the reaching modality and the otherthe looking modality. The hypothesis was that performance on the lookingversion of the task would exceed performance on the reaching version from 5to 7 months, but that after 8 months of age there would be 3 groups ofinfants. One group would perform at a higher level on the looking versionfrom 8 to 10 months. The 2nd group would perform at a higher level on thereaching version and the 3rd group would perform at the same level on thereaching and looking versions from 8 to 10 months. Each of these 3patterns of performance was hypothesized to be associated with a differentpattern of brain electrical activity and heart rate activity. Specificallyfor the EEG, frontal-occipital EEG power and coherence values werehypothesized to be related to exceptional reaching performance, whilefrontal-parietal EEG power and coherence values were hypothesized to berelated to exceptional looking performance. Fourteen infants (8 male) were participants in this longitudinal study.Infants were seen in the research lab monthly from 5 to 10 months of ageand were assessed on both looking and reaching versions of an objectpermanence task. Assessment resulted in each infant receiving ordinalscale scores. EEG and ECG were recorded during baseline and looking taskassessment. Artifact associated with gross motor movements during thereaching task precluded EEG recording during the reaching version of thetask. As predicted, performance on the looking version of the task and thereaching version of the task was comparable from 5 to 7 months of age.Between 8 and 10 months of age, the 3 predicted patterns of performancewere observed. Six (42%) infants performed at a higher level on thelooking task relative to the reaching task. Five (37%) infants performedat a higher level on the reaching task relative to the looking task and 3(21%) infants had the same performance on both versions. Associationsamong task performance, brain electrical activity, and heart rate activityfor the exceptional reaching and exceptional looking performance groupswill be discussed. Parallels will be drawn between these data and otherdata sets that have demonstrated associations between frontal brainelectrical activity and object permanence task performance.
poster
Individuals can focus processing resources on a particular location inspace by controlling visual attention. Moreover, attention can act as aself-regulative tool, allowing individuals to deal with distressingstimuli by disengaging from the unpleasant and shifting attention topleasant stimuli. Individuals who are can effectively control attentionmay be better able to handle novel situations, including those of a socialnature. By the same token, children who can not engage in thegive-and-take of a social encounter may tend to withdraw and appearsocially reticent. Other studies of social reticence and withdrawal havefocused on physiology, namely frontal lobe activation. In particular,individuals with right frontal asymmetries tend to express more withdrawalemotions than individuals with left frontal asymmetry. This studyexplored the relationship between frontal asymmetry and attention and itssubsequent effect on social reticence. In the attention task, nine-month-old infants were presented with afixation point. A distracter was then introduced at regular intervals.The data were coded for total time spent in task, proportion of time spentfocused on fixation, proportion at distracter, and the number of visualshifts from fixation to distracter. Subjects were also seen at 14-months,24-months, and 4-years. Children with stable right frontal asymmetries (seen in at least three ofthe four visits) showed less control of visual attention than childrenwith stable left frontal asymmetries. Both groups spent significantlymore time focused on the fixation point than the distracter. However, thegap was significantly larger for the Right Frontal infants, since theyspent a greater proportion of the testing period focused on the fixationpoint. Left Frontals were able to remain on task longer while RightFrontal infants often became distressed and were unable to continue. The raw number of visual shifts from distracter to fixation most clearlyreflects the infants' use and control of visual attention. At ninemonths, Right Frontals were significantly less likely to shift attentionback-and-forth. The number of shifts were rank ordered and median split.We then looked to see how the children were sorted in a two (Right vs.Left Asymmetry) by two (Low vs. High frequency of shifts) matrix. Of theten Right Frontals, nine shifted attention infrequently. Among thethirty-eight Left Frontals, however, twenty-two were high shifters andsixteen were low shifters. This produced a significant chi-squaredistribution. At the 4-year visit, subjects were observed in a play group and weremonitored for signs of reticence. Reticence scores were compared usingthe two by two tables described above. Among the children who shiftedvisual attention infrequently at nine months, the Right Frontals hadsignificantly higher reticence scores at age 4. Because of the numericalimbalance between Right and Left Frontals in the frequent shifters, asimilar analysis could not be run. In summary, there appears to be arelationship between frontal asymmetry, attention shifting ability, andsocial reticence so that right frontal asymmetry, when combined with lessfluid control of attention, produces greater social withdrawal.
poster
Singing and infant-directed (or musical) speech are essential tools incommunicating with preverbal infants. For example, infant-directed singing aidsmothers in controlling the emotional state of their infants (Rock, Trainor, &Addison, 1999). As well, infants respond differently to positive versus negativeintonational messages in infant-directed speech (Fernald, 1993) and children asyoung as age three can reliably associate musical excerpts and emotions (e.g.,Cunnningham & Sterling, 1988; Trainor & Trehub, 1992). The strength ofassociation between music and emotion and the involvement of music and/ormusical speech in early emotional communication suggests that a comprehensiveunderstanding of how emotional information is processed in the brain will not becomplete until responses to music are considered. The purpose of the presentstudy was to examine cortical responses in infants to emotion elicited by music. We recorded regional brain electrical activity (EEG) using a lycraelectrocap and SA Bioamplifiers in 52 healthy 9 month-old infants during thepresentation of musical stimuli which were known to vary in emotional valence(pleasant vs. unpleasant) and intensity (intense vs. calm) based upon adult ratings. Three musical excerpts were used and each presented for 30 sec. These excerptscomprised intense-unpleasant emotion (e.g., Peter and the Wolf by Prokofiev,wolf excerpt); calm-unpleasant emotion (e.g., Adigo by Barber); and calm-pleasant (e.g., Spring by Vivaldi). The EEG data were recorded from the left andright frontal (F3, F4), parietal (P3, P4), and occipital (O1, O2) brain regions. We found that the pattern of frontal brain activity changed betweenbaseline and the presentation of the musical excerpts. Infants exhibited greaterrelative left frontal EEG activity during baseline and then shifted to a pattern ofgreater relative right frontal activity during the presentation of the musicalexcerpts, regardless of the affective valence of the music. Findings are consistentwith the substantial literature extant noting increased right hemisphereinvolvement during the processing of emotion in humans. These findings suggestthat the pattern of asymmetric brain activity during the processing of emotion seenin response to other stimulus modalities in humans is present in the second half ofthe first year of life in infants using auditory stimuli.* Supported by a research grant from the Social Sciences and HumanitiesResearch Council of Canada
poster
Infant visual attention has been studied extensively using cognitivemeasures, such as habituation and paired-comparison procedures, butless work has examined how infant attention operates in socialcontexts. The still-face procedure is a widely used procedure thatmeasures infants' responses to violation of social expectancy. In thisparadigm, mothers first engage in 'normal' interaction with theirinfants, which is immediately followed by a 'still-face' episode, inwhich the mother provides no feedback and sits still with a neutralfacial expression. Following that phase, the mother resumes normalinteraction. Although several studies have examined infants' emotionaland behavioral responses to the still-face procedure, little work hasbeen done on infant attentional responses to this violation of socialexpectancy. The recent application of physiological measures to the study ofinfant attention has added a great deal of theoretical andmethodological specificity to the ways that infant attention ismeasured, understood, and interpreted. Researchers such as Porges(e.g., 1992) and Richards (e.g., 1997) have examined infants'physiological responses during visual attention, and have developedtheories to describe the interplay between behavioral andphysiological measures of attention. Although little work to date hasexamined physiological measures of attention during socialinteraction, physiological measures of attention may prove to beparticularly valuable in social contexts, which are inherently dynamicand evolving. The current study examined individual differences and developmentaltrends in both behavioral and physiological measures of infantattention, and their relation to infants' performance in a socialinteraction task in which the mother's feedback was temporarilyaltered. Fifty infants were tested cross-sectionally at 4 and 6months of age (25 infants at each age). First, infants engaged in a'cognitive' attention task, in which habituation and reaction timewere measured in a peripheral stimulus localization procedure. Following this, mothers participated with the infants in thestill-face procedure. Infant affect, gaze, and self-comfortingbehaviors were measured in all phases of the social interaction. Also,mothers' vocalizations, physical activity toward the infant, andaffect were coded in order to assess infant and maternal synchrony.Infant heart rate and respiration (for calculation of respiratorysinus arrhythmia) were recorded throughout the entire experimentalsession via electrodes and a respiration bellows (expandable cuff)attached to the infant's chest.Preliminary analyses indicate that infant heart rate increased duringthe still-face procedure relative to the baseline play period. Further analyses will examine whether individual differences inheart-rate and respiratory sinus arrhythmia are related to infantattentional and emotional responses during the still-face procedure. In addition, we will examine developmental changes in infantattentional responses during the still-face procedure, and therelations between attentional measures taken during the still-face andduring the baseline habituation and reaction time tasks. Results willbe discussed in terms of the stability of attention across cognitiveand social tasks, and the usefulness of physiological measures ofattention for the study of mother-infant interaction.
poster
The ability to recognize familiar objects is an important part of infantcognitive development. Infants' favorite toys are familiar companions intheir everyday lives, and are often important vehicles by which infantslearn about their worlds. Experimenters of infant cognitive developmenthave used event-related potentials (ERPs) to investigate the brain basisof recognition memory in young children and infants. De Haan and Nelson(1999) measured infants' brain responses to 2-dimensional photographs offamiliar and novel faces (the infant's mother and an unfamiliar woman) andobjects (the infant's toy and an unfamiliar toy). Although such studiesmay provide important information about the development of infants'ability to recognize stimuli, infants do not typically encounter2-dimensional representations in their worlds. Instead, infants encounterand must recognize real, 3-dimensional objects. In the present study,18-month-olds' recognition memory was tested by measuring ERPs in responseto the child's actual 3-dimensional toy and an unfamiliar toy. Theunfamiliar toy was selected from a set that had been favorites for infantsof the same age in previous experiments. The toys were matched forperceptual features such as approximate size, volume, shape, and color.The toys were placed inside of an opaque box that became transparent whena light inside the box was illuminated. The toys were placed in aspecially designed device such that the child could only see one object ata time. Each toy was presented for 50 trials of 500 msec duration each,with a variable intertrial interval. ERP data were collected using theGeodesic Sensor Net system (Electrical Geodesics, Inc.). Results indicateevidence of recognition memory. A number of components were observed thatdifferentiated responses to familiar and novel toys. A middle-latencynegative component (the Nc) was observed over anterior midline electrodesights, suggesting attention to and memory for the stimuli. In addition,late onset slow wave activity, which previously has been hypothesized toreflect novelty detection and memory updating was observed. Finally, amiddle-latency positive component was observed over posterior midlineelectrodes sights. This component may reflect a precursor of the P300,which indicates detection of task relevant stimuli. The results suggestthat it is feasible and beneficial to test brain correlates of recognitionmemory for real world 3-dimensional objects.
poster
This study examined the relation between attention engagement and ERPindices of the recognition of briefly presented visual stimuli. One goal ofthe study was to determine if ongoing attention state affected therecognition of briefly presented visual stimuli in infants at 6 and 7.5months of age. The second goal was to examine if developmental changes inthe ERP responses indicating recognition were affected by ongoing attention.Infants at age 6 and 7.5 months (16 per age) were tested in an 'oddball'procedure in which one stimulus was presented frequently and other stimuliwere presented infrequently. The infant's attention was elicited andmaintained by presenting an interesting visual stimulus (Sesame Streetmovie, 'Follow that Bird'). Ongoing attention status was monitored withheart rate, where a significant heart rate deceleration and maintainedlowered heart rate defined 'sustained attention', and a lack of heart ratedeceleration, or a return of heart rate to its prestimulus level,represented inattention. The recognition memory procedure included thepresentation of a stimulus that was familiar to the infant that waspresented 60% of the time (FF, frequent familiar), a stimulus that wasfamiliar to the infant that was presented 20% of the time (IF, infrequentfamiliar), and a series of novel stimuli that were presented 20% of the time(IN, infrequent novel). The EEG was measured from 20 channels (19 channelsof the 10-20 system, and Oz) and event-related-potential (ERP) averages wereconstructed from the onset of the stimulus for 2 seconds. The firstnegative ERP component occurring about 400-800 ms after stimulus onset('Nc') and the slow wave portion of the ERP response occurring about 800 to1500 ms following stimulus onset were quantified.First, the Nc component was larger when attention was occurring compared towhen the infants were inattentive. The Nc was not different for the FF, IF,and IN presentations. This confirms the idea that the Nc componentrepresents an 'attention-alerting' mechanism that occurs in response to anyvisual stimulus. The Nc component was approximately the same size in the 6and the 7.5 month old infants. Second, the slow wave components differedfor the three stimulus presentation conditions. For the 6-month-old infants,during sustained attention there was a large positive slow wave for the IFstimuli, a small negative slow wave for the IN stimuli, and no slow waveactivity for the FF stimuli. The 7.5-month-old infants also showed thispattern for both inattentive and attentive states, but the differencebetween the ERP amplitude for the three presentation conditions was muchlarger when attention was engaged. These findings suggest that the Nccomponent represents an attention response to any stimulus and the laterslow waves represent both recognition memory and stimulus probability.Developmental changes in the ERP measures of recognition of brieflypresented stimuli may be due to older infants showing recognition memoryover a wide variety of attentional states, whereas younger infants onlyexhibit recognition memory when attention is sufficiently engaged.