University of Groningen. Through the Eyes of an Infant Hunnius, S.

Transcription

1 University of Groningen Through the Eyes of an Infant Hunnius, S. IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record Publication date: 2005 Link to publication in University of Groningen/UMCG research database Citation for published version (APA): Hunnius, S. (2005). Through the Eyes of an Infant: The Early Development of Visual Scanning and Disengagement of Attention [S.l.]: s.n. Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons). Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Downloaded from the University of Groningen/UMCG research database (Pure): For technical reasons the number of authors shown on this cover page is limited to 10 maximum. Download date:

2 Through the Eyes of an Infant The Early Development of Visual Scanning and Disengagement of Attention

3 2004, Sabine Hunnius All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means without prior permission by the author. Printed by: Drukkerij Alba, Groningen ISBN

4 RIJKSUNIVERSITEIT GRONINGEN Through the Eyes of an Infant The Early Development of Visual Scanning and Disengagement of Attention Proefschrift ter verkrijging van het doctoraat in de Psychologische, Pedagogische en Sociologische Wetenschappen aan de Rijksuniversiteit Groningen op gezag van de Rector Magnificus, dr. F. Zwarts, in het openbaar te verdedigen op maandag 10 januari 2005 om uur door Sabine Hunnius geboren op 19 juni 1974 te Bonn, Duitsland

5 Promotores: Copromotor: Beoordelingscommissie: Prof. dr. P. L. C. van Geert Prof. dr. J. M. Bouma Dr. R. H. Geuze Prof. dr. R. N. Aslin Prof. dr. A. Johnson Prof. dr. G. J. P. Savelsbergh

6 Wir sehen in der Natur nicht Wörter, sondern immer nur Anfangsbuchstaben von Wörtern, und wenn wir alsdann lesen wollen, so finden wir, daß die neuen sogenannten Wörter wiederum bloß Anfangsbuchstaben von andren sind. Georg Christoph Lichtenberg

7

8

9 CONTENTS Chapter 1 Introduction 9 Chapter 2 Developmental Changes in Visual Scanning 31 of Dynamic Faces and Abstract Stimuli in Infants Chapter 3 Gaze Shifting in Infancy: 57 A Longitudinal Study Using Dynamic Faces and Abstract Stimuli Chapter 4 Associations between the Developmental Trajectories 79 of Visual Scanning and Disengagement of Attention in Infants Chapter 5 A Longitudinal Study of Shifts of Attention and Gaze 99 in Preterm and Full-term Infants Chapter 6 Summary and General Discussion 127 Samenvatting 147 Zusammenfassung 155 Acknowledgements 165 Curriculum Vitae 169

10 Chapter 1 Introduction

11 Chapter 1 Anybody who has ever observed a baby of a few weeks of age looking around and examining his environment has undoubtedly noticed how different his visual behavior is from that of an adult. A young infant tends to move his eyes slowly, and often it seems as if he gazes vacantly in front of him. He likes to look at patterns with high contrast, so the locations that interest him can be the dark frame of a painting on a light colored wall or the knob of a drawer. From time to time, it seems as if the baby s gaze gets stuck at one location for 10 to 20 seconds, and the infant may keep on staring there, although there may be other equally salient things to look at in his environment, for example a colorful toy. If a moving object catches his eye, he may follow it, as long as it is not going too fast. Were we to observe the same infant 3 or 4 months later, we would see that his visual behavior has changed dramatically. The infant now examines objects by scanning them quickly and systematically. Complex, colorful, and moving stimuli attract his attention particularly easily. The baby tends to alternate his intense inspections with brief looks away. His eye movements are fast, and he tracks moving objects or persons easily. Vision plays a crucial role in the daily life of an infant. As young infants are unable to move around or to grasp objects easily, they explore their environment and learn about the world by looking. Looking is also one of the most important ways in which infants communicate with their caretakers. Face-to-face interaction forms the beginning of social communication and plays an important role in the bonding of infant and caretaker. The ability to carry out fast, accurate shifts of gaze serves as the basis for visual information processing and communication in early infancy. Accordingly, the developmental changes in the visual system sketched above are of great importance. The question which factors enable or hinder its functional development therefore deserves precise investigation. This dissertation deals with the developmental changes in attention and gaze shifting behavior which occur during the first few months of life. It reports the results of an intensive longitudinal study on visual exploration behavior and the disengagement of attention and gaze from a fixated stimulus. The purpose of the study was to examine these issues using dynamic stimuli and to investigate the influence of different stimuli on young children s visual behavior. Abstract and socially relevant stimuli were contrasted. In addition to examining a group of healthy full-term infants, it was the goal of this study to describe the nature and extent of differences in the development of these attentional processes between full-term infants and infants born prematurely. A sequence of six measurements during a period of five months, in which rapid development of the visual system was expected, was supposed to provide exact information on the timing and tempo of development. Precise techniques to measure visual fixations and eye movements were employed. This introductory chapter begins with an overview on vision, attention, and eye movements in early infancy. Different methods of measuring eye movements are discussed. Then, the biological mechanisms which are thought to underlie shifts of 10

12 Introduction gaze and attention and their development are expounded. Next, different aspects of the role of visual input are examined: The impact of exposure to visual stimulation is considered, and the question is raised whether (extra) visual experience influences the early visual and attentional development. Then, the relation between the sort of visual input and its processing is investigated. In this context, the importance of using ecologically valid stimuli in experiments is discussed. The chapter ends with a description of the goals of this study and an outline of the dissertation. Attention Shifts and Eye Movements in Infancy Shifts of gaze and shifts of attention are tightly associated, although they do not necessarily occur conjointly (Stelmach, Campsall, & Herdman, 1997). Whereas overt shifts of attention involve saccadic movements of the eyes in order to align the fovea of the retina, the focus of attention can also be shifted covertly in absence of eye or head-movements (Wright & Ward, 1998). It has been argued that covert attention shifts occur as premotor activity in the preparation of eye movements but without an actual gaze shift (Rizzolatti, Riggio, Dascola, & Umiltà, 1987). In daily life, however, attention and gaze shifting are usually very closely joined. Gaze and attention shifting has been studied extensively in diverse research fields. For example, topics as diverse as stimulus discrimination, memory, or concept formation have been investigated using the habituation paradigm. This procedure is based on the idea that the way infants look at and away from a stimulus signifies their processing of information about the stimulus (Bornstein, 1985). In addition, variables measured during infant habituation studies as duration of the first fixation, the total fixation time, or the duration of the longest fixation have been shown to predict later cognitive abilities (for reviews, see Colombo, 1993; Slater, 1995). Also, gaze shifting behavior has been analyzed in studies on social development. Eye contact has been studied as one of the earliest forms of social interaction and as an important factor for the attachment of mother and infant and for the quality of their relationship (Keller & Gauda, 1987; Schölmerich, Leyendecker, & Keller, 1995). Research on face-to-face interaction between mother and infant has examined infants looking to or away from their mothers face and has revealed that infants tend to regulate the intensity of an interaction by shifting their gaze (Field, 1979; Stifter & Moyer, 1991). Furthermore, it has been shown that children who are able to disengage attention and move gaze more easily, are less susceptible to distress and are more soothable (Johnson, Posner, & Rothbart, 1991). These various examples emphasize the importance of shifts of attention and gaze in infancy. They also illustrate that attention and gaze shifting behavior has been studied to gain insight into very different areas of infant psychology, such as visual exploration, early communication, or arousal modulation. Also diagnostic instruments which assess infants developmental status during the neonatal or the infant period are among other tasks based on items which require visual orientation to animate and inanimate stimuli or shifting gaze between 11

13 Chapter 1 two objects (Neonatal Behavioral Assessment Scale, NBAS, Brazelton & Nugent, 1995; Bayley Scales of Infant Development, BSID-II, Bayley, 1993). Development of Vision For a long time, it was commonly accepted that newborn and young infants perceive their environment in an extremely impoverished way (see e.g., Dewey, 1935; Pratt, 1954; see also Stone, Smith, & Murphy, 1973, p. 3-4 for more examples). Also Wilhelm Preyer, whose book Die Seele des Kindes (1882) is often considered as the beginning of infant psychology, reported the relative immaturity of the visual system of a human newborn and, as one of the first, sought to examine a young infant s reactions to visual stimuli. William James described the early experiences of an infant as one great blooming, buzzing confusion (James, 1890, Vol. 1, p. 488), and also Jean Piaget (1936, 1937) emphasized the strong sensory limitations during infancy. It was not before the 1960s that researchers started to specify what exactly the perceptual limitations of early infancy are, and concentrated on describing what young infants are able to see. It is clear that young infants vision falls far short of adult standards. The optical state of the eye, though, seems to be quite good in the newborn. Infants can accommodate on targets which are relatively close already during the first days of life (Braddick, Atkinson, French, & Howland, 1979; Haynes, White, & Held, 1965). Visual acuity in newborns, however, has been shown to be about 1/30 of the acuity of adult levels and to have improved by a factor of 3 to 4 by 5 months of age (Fantz, Ordy, & Udelf, 1962). However, it has been argued that unsharp vision does not handicap the infants, as they can still perceive relevant features at a proper distance, but might instead even help them to prevent excessive visual stimulation (Maurer & Maurer, 1988). Even very young infants seem to have some form of color vision, and it is probably very similar to that of adults. However, infants of 1 month of age have been shown to be unable to discriminate certain colors (e.g., green and yellow), but by 2 months they succeed in making most color discriminations (Clavadetscher, Brown, Ankrum, & Teller, 1988). In a complex colorful stimulus, young infants thus probably do not distinguish as many different colors as adults would, and colors presumably appear less intense to them than they do to adults. When infants are 3 to 4 months old, their color vision is at a relatively adult level (Teller & Bornstein, 1987). To sum up, although infants vision is not yet fully mature during the first few months after birth, they see well enough to respond appropriately to relevant aspects of the environment and function effectively in their roles as infants (Hainline, 1998; Hainline & Abramov, 1992). Developmental Changes in Visual Attention and Eye Movements Eye movements emerge already during prenatal development (Prechtl, 1984). They have been observed in 16 to 18 weeks gestation, but they are naturally not associated 12

14 Introduction with visual stimuli until exposure to light after birth. When infants are born, they have been demonstrated to prefer stimuli and objects with simple patterns and high contrast (Fantz & Yeh, 1979). They are able to localize a visual target, although in a rather inaccurate and unreliable way (Atkinson, 1992). Once they look at a stimulus, they scan it actively, but tend to fixate only limited parts of it (Haith, 1980; Bronson, 1990) and to ignore other stimuli in their visual field (Bronson, 1996; Haith, 1980; Salapatek, 1975). During the first weeks of life, infants have also been shown to look especially at edges or outer contours of a stimulus pattern ( contour salience effect, Bronson, 1991) and not to attend to stationary inner parts of a stimulus ( externality effect, Salapatek, 1975; Milewski, 1976). However, if the internal elements of a pattern are moving, they are more likely to be fixated even by very young infants (Bushnell, 1979; Girton, 1979). Newborns are able to track moving stimuli (Tauber & Koffler, 1966), although they tend to do it not in a smooth, but in a saccadic, step-wise way and tend to lag behind the movement of the stimulus (Aslin, 1981). Around 2 months of age then, infants have been observed to follow a moving visual target smoothly (von Hofsten & Rosander, 1996), but it is not before about 3 months that infants predict the movement of the stimulus in an anticipatory way and do not lag behind anymore (Aslin, 1981). Between approximately 1 and 3 months of age, infants have trouble looking away from a stimulus, once their attention has been engaged, and they may exhibit long periods of staring. This phenomenon of disengagement difficulty has been called sticky fixation (Hood, 1995) or obligatory attention (Stechler & Latz, 1966). It can be observed in a laboratory context (Hood, Murray, King, Hooper, Atkinson, & Braddick, 1996; Aslin & Salapatek, 1975), in free looking situations (Stechler & Latz, 1966), during social interaction (Hopkins & van Wulfften Palthe, 1985), and is also reported frequently by mothers and other caretakers as an every-day experience. As infants grow older, disengaging attention and shifting gaze away from a stimulus becomes increasingly efficient. By 4 months of age, infants are able to shift gaze easily and rapidly, and staring behavior becomes rare (Hood & Atkinson, 1993). When infants become more capable of achieving a balance between engaging and shifting attention, also their scanning behavior changes. Infants of about 3 months have been described to explore the stimulus under examination more consistently and more extensively (Bronson, 1996). They exhibit more brief fixations and scan more rapidly over the entire array of stimulus figures. Salient parts of a stimulus still attract the infants gaze, but they have gained volitional, strategic control over their scanning behavior (Bronson, 1994). From 3 months on, infants not only start anticipating stimulus movement during visual tracking, but also begin to form expectations about the locations of upcoming stimuli and may even initiate an eye movement in advance (Haith, Hazan, & Goodman, 1988; Canfield & Haith, 1991). The increasing intentional control over their eye movements enables infants of 4 to 5 months to examine their environment in an efficient and flexible way. They can shift their gaze fast and reliably between and within 13

15 Chapter 1 visual stimuli and are able to direct their gaze to relevant locations. Eye movements are now generated in accordance with the strategic demands of ongoing information processing. When scanning familiar stimuli, recursive scanning patterns can be observed (Bronson, 1982). During face-to-face interaction, infants of 3 months and older tend to shift their gaze away more often, either in order to regulate arousal (Stifter & Moyer, 1991) or to explore other locations which are becoming increasingly more interesting to them (Kaye & Fogel, 1980). Measuring Eye Movements in Infants Vision and human viewing patterns have fascinated researchers for a long time (see e.g., Müller, 1826; Preyer, 1882). Eye movements and fixations in infants have been observed in order to address very different topics, such as visual scanning (Haith, 1980; Bronson, 1982) and visual processing (Bronson, 1991), the acquisition of object knowledge (Johnson & Johnson, 2000; Johnson, Slemmer, & Amso, 2004), or the formation of categories (McMurray & Aslin, 2004). Probably the most common method of studying eye movements is simple observation of gaze. There are two more precise methods of measuring eye movements and fixations: electro-oculography (EOG) and corneal-reflection photography. EOG is based on measuring the change in electrical potential which accompanies the rotation of the eye. However, this method has several limitations, especially when used with young infants (Aslin & McMurray, 2004). It is particularly sensitive to artifacts and requires the application of electrodes on the subject s face. Furthermore, EOG provides only data on the relative displacement of the eye and no information about where on the stimulus the subject is looking. For corneal reflection eye-tracking, an (infrared) light source is used to create a reflection off the front surface of the eyeball, while the eye is recorded on video. The reflection is displaced when the subject moves fixation, and the information about the relative position of the corneal reflection with respect to the center of the pupil and its change is used to determine whether an eye movement took place. However, to gather information about the location of fixation, the corneal reflection eye-tracking system has to be individually calibrated before the measurement in order to map the output data onto the field the subject is looking at (Bronson, 1983; Harris, Hainline, & Abramov, 1981). The technique of infrared corneal photography has been applied to human infants first in the 1960s (Salapatek & Kessen, 1966; Haith, 1969) and has been improved in many respects since then. The sampling rate has been increased from 2-6 Hz to Hz (Aslin, 1981; Hainline, 1981; Bronson, 1982) and even to 120 and 250 Hz. The increased temporal resolution has greatly amended the accuracy in determining fixation durations. Another recent improvement is to have the camera mounted on a motor-driven base which moves in order to maintain the image of the eye in the camera s field-ofview and compensate for head-movements of the infant. However, large or rapid head- 14

16 Introduction movements still formed a problem for a corneal photography eye-tracking system. This resulted in experimental setups in which the infant s head had to be restrained. As this is a quite unnatural situation and can be distressing for the infants, the implementation of a head-tracker, a position-sensing system that monitors head movements and communicates this information to the eye-tracking system, means an important innovation. Today, it is thus possible to examine how infants look at different stimuli in a more precise and at the same time more natural way than ever before. However, there are still some problems, which form a challenge for the researcher: The accuracy of the measurements of the location of fixations depends largely on the quality of the calibration carried out. For working with infants as young as 6 weeks of age, custombuilt calibration procedures have to be developed in order to make the infants fixate a sequence of (preferably many) calibration points. Furthermore, young infants tend to have poor postural control, which requires several accommodations of the experimental setup, including the position of the infrared camera and the stimulus display. To sum up, measuring eye movements in infants remains a challenging task for the researcher as it is tried to apply a complex and highly sensitive technique to delicate subjects, who are indifferent to instructions. Neurophysiological Models of Eye Movement Generation in Adults After this overview on vision and eye movements in infancy and on early visual and attentional development, the question remains which neurophysiological processes underlie the described functions and developmental changes. The following two paragraphs are dedicated to this question. Eye movements are controlled by different cortical and subcortical structures. Many neurobiological models proceed on the assumption of two visual systems, a phylogenetically older retinotectal system and a newer geniculostriate system. Early anatomical studies had already identified two distinct streams from the retina through the brain, but the functional distinction arose from studies in the 1950s en 1960s (see e.g., Sprague & Meikle, 1965). The first models featuring the distinction between two routes were proposed by Trevarthen (1968) and Schneider (1969). Trevarthen (1968) described an ambient system for movement control and a focal system for object vision. Schneider (1969) suggested that the tectal system defines where an object is located and initiates an orienting response and the newer cortical mechanisms are about what there is to see precisely in the selected location. In 1982, Ungerleider and Mishkin revised the dichotomy into a ventral versus a dorsal cortical stream. According to this model, the ventral stream, concerned with the what -aspects of an object as color, form, or face recognition, is assigned to the inferotemporal cortex. The identification of spatial location, on the other hand, is thought to be subserved by the dorsal stream ( where ), which is anchored by the posterior parietal cortex. The dorsal versus ventral distinction has been associated with a division earlier in the visual pathway, namely between the parallel magnocellular and parvocellular sys- 15

17 Chapter 1 tems (Livingstone & Hubel, 1988; Shapley & Perry, 1986; Van Essen & Maunsell, 1983). These two systems are anatomically segregated at the retina and the lateral geniculate nucleus and project to different parts of the primary visual cortex. The parvocellular based system subserves form and color vision, whereas the magno cells are specialized in movement perception and some aspects of stereoscopic vision. However, the simple distinction into these two parallel systems has been questioned recently, as there are many interactions in visual processing between the magnocellular and the parvocellular stream (see e.g., Cowey, 1994; Merigan & Maunsell, 1993 for reviews). One of the currently predominant models of eye movement generation is the one developed by Peter Schiller (Schiller, 1985, 1998). His recent model (Schiller, 1998) is based on adult primate electrophysiological and lesion data and also distinguishes between two different, but partly overlapping, neural systems of eye movement control: the anterior and the posterior eye movement control system. The anterior system is responsible for saccades that are voluntary or planned (as e.g., scanning behavior), whereas the posterior system generates fast, reflex-like eye movements and orienting responses, as they occur, for example after the sudden appearance of a salient stimulus in the periphery. The streams of the anterior system originate in retinal ganglia, which are specialized for the analysis of fine detail and color (Richards & Hunter, 1998). They project through the lateral geniculate nucleus to the striate cortex, and from there they run through the temporal or the parietal lobe to the frontal eye fields. Then they project via the basal ganglia and the superior colliculus to the eye movement centers of the brain stem. However, these brain stem structures also receive direct input from the frontal eye fields within the anterior eye movement control system. The posterior eye movement control system receives the majority of its input from the retinal ganglia, which are located in the peripheral retina and are specialized for the detection of sudden changes (Richards & Hunter, 1998). Its pathways run via the lateral geniculate nucleus to the striate cortex. Then, they project partially via the parietal lobe through the basal ganglia to the superior colliculus. The activity of the superior colliculus thus is controlled via both systems, the anterior as well as the posterior eye movement control system. Their excitatory or inhibitory input plays an important role in the generation of eye movements to interesting location and at the same time in the inhibition of reflexive eye movements in order to ensure a well-organized input of visual information. However, within the anterior eye movement control system, there is also a pathway which bypasses the superior colliculus and hereby enables the generation or inhibition of eye movements independently from collicular control. Neuropsychological Models of Attentional Development in Infants The visual and attentional behavior of an infant is largely determined by the developmental state of the brain structures which form the visual system. Changes 16

18 Introduction observed in behavior and neural correlate can be due to maturation, but can as well be a response to experience (Greenough, Black, & Wallace, 1987). Also, there are several examples of neural and behavioral changes that are the result of an interaction between intrinsic factors and environmental aspects (Johnson & Morton, 1991; Greenough, Black, & Wallace, 1987). Anatomical (Conel, ) and PET scan studies (Chugani, 1994) have demonstrated that, generally, subcortical brain structures are more mature at birth than cortical visual mechanisms. The superior colliculus is one of the most mature structures involved and is thought to play a crucial role in the generation of eye movements during early infancy. Gordon Bronson was one of the first to propose a model which applied findings from research on adult neurological systems (e.g., Schneider, 1969; Trevarthen, 1968) to infant visual behavior (Bronson, 1974). According to his model, the early development of visual attention can be viewed as a shift from subcortical to cortical processing. Visual behavior in the newborn thus is mainly controlled by means of a phylogenetically older visual system. It is only by 2 to 3 months of age that the locus of control switches to the primary visual system and its mainly cortical pathways. However, the original model proposed by Bronson (1974) based on the two visual systems (Schneider, 1969) and a subcortical-cortical dichotomy has been criticized as being too simplistic and incomplete (Atkinson, 1984; Johnson, 1990). Also, the early presence of certain perceptual abilities has given rise to the notion that there is at least some degree of cortical functioning at birth (e.g., Slater, Morison, & Somers, 1988). It is now known that several comparatively independent cortical streams of visual processing exist (see e.g., Van Essen, 1985) and that they undergo various forms of developmental changes, such as myelination, synaptic generation, neural innervation, synaptic pruning, and neurotransmitter development (see e.g., de Haan & Johnson, 2003). Bronson s most recent model (Bronson, 1994, 1996) is based on two pathways the striate and the poststriate networks (Bronson, 1996) which are similar to the posterior and anterior eye movement control system proposed by Schiller (1998, 1985) and the assumption that the changes observed in early visual behavior can be explained by reference to the maturational state of these pathways. During the first few weeks of life, eye movements are mainly controlled by the striate networks. These areas are highly responsive to stimulus salience, accordingly, young infants visual behavior tends to be mainly salience-guided. Once the fovea is aligned with an area of high salience, fixations are often concentrated around this area because due to the anatomical structure of the retina close salient areas produce higher striate activity than comparable areas further away. As highly salient stimuli produce long lasting activity, fixations tend to be long in young infants. From about 6 weeks of age on, the poststriate networks with their pathways through the parietal and frontal cortex become increasingly effective. This system comprises areas that are able to encode the location and the form of visual stimuli. These pathways project to the superior 17

19 Chapter 1 colliculus and to the brain stem centers which directly generate eye movements. Older infants thus can draw on these poststriate capacities to override salience effects and move their eyes intentionally to locations of interest. Also the model of visual and attentional development by Mark Johnson (1990, 1995; Johnson, Gilmore, & Csibra, 1998) refers to the eye movement control system by Schiller (1985), particularly to four distinct pathways. Johnson assumes that the characteristics of visually guided behavior mirror the degree of functionality of the four pathways three cortical and one subcortical and that the developmental state of the primary visual cortex determines which of theses pathways is functional. In correspondence with the inside-out pattern of postnatal development in the cerebral cortex (see e.g., Nowakowski, 1987; Rakic, 1988), he hypothesizes that the lower layers tend to be more capable than more superficial ones. In newborn infants, only the deeper layers of the primary visual cortex are functional, and the visually guided behavior is controlled predominantly by the subcortical pathway. According to Johnson, the saccadic pursuit tracking observed in young infants (Aslin, 1981) and the phenomenon that young infants do not attend to a stationary pattern within a larger frame or pattern ( externality effect ) are characteristic of visual behavior that is controlled subcortically. Other behaviors, such as early pattern recognition (Slater, Morison, & Rose, 1982) or orientation discrimination (Atkinson, Hood, Wattam-Bell, Anker, & Tricklebank, 1988) suggest at least some cortical functioning also in the newborn infant. During the first month, the nigral pathway, which is an inhibitory input to the superior colliculus from several deeper layers of the primary visual cortex, becomes increasingly functional. According to Johnson, this unregulated tonic inhibition has as a transient consequence the infants disengagement difficulties, known as sticky fixation or obligatory attention. Around 2 months of age, infants begin to show smooth visual tracking. According to Johnson, the onset of this behavior coincidences with the functioning of the middle temporal area pathway. During the third and fourth month then, the pathways involving the frontal eye fields become functional, as the upper layers of the primary visual cortex mature. This leads to a more differentiated regulation of collicular activity and ends the tonic inhibition, which caused the staring behavior. Infants are now able to move their gaze intentionally from fixation to other locations of interest and to generate anticipatory eye-movements. The models by Johnson and Bronson succeed in explaining the most important developmental changes of visual behavior in infancy, but they differ in the underlying processes which they assume to be involved. However, both models have strongly influenced the models proposed by other researchers (see e.g., Atkinson & Braddick, 2003; Richards & Hunter, 1998). Premature Birth and the Role of Early Visual Input Development emerges from the interaction of many different factors. As described earlier, maturation of the eye and of the brain play a crucial role in the development 18

20 Introduction of visual and attentional skills. However, other factors might be equally important. The next two sections address different aspects of the role of environmental factors. First, early exposure to visual input caused by premature birth is discussed. The next paragraph focuses on the impact of the nature of visual stimulus material on young infants visual and attentional performance. When infants are born prematurely, they are confronted with a very different environment than they experience in utero much earlier than their full-term agemates. The preterm birth also puts them at risk for severe medical complications such as breathing problems, infections, and brain damage. In neonatal intensive care units (NICUs), many efforts are being made to create the optimal conditions according to the infants physiological needs (e.g., temperature and nutrition) and psychological requirements (e.g., stimulation and contact). A large amount of research on the optimal treatment of preterm infants has been carried out to date (see e.g., Holditch-Davis & Black, 2003; Wolke, 1987). One important difference between full- and preterm infants is that infants born prematurely are confronted with visual input earlier in life than full-term infants. There are contrasting theories about the impact of this extra experience on the visual development of the infant. One theory implies that healthy preterms benefit from their early exposure to the visual world (Fielder, Foreman, Moseley, & Robinson, 1993). This account is supported by Hunt and Rhodes (1977), who found that during the early months preterm infants have higher scores on the mental scale (MDI) of the Bayley Scales of Infant Development (Bayley, 1969), a scale which when very young infants are tested relies mainly on the infants visual responses. Also, superiority of visual acuity (Sokol & Jones, 1979; Mohn & van Hof-van Duin, 1986) and more mature focusing and tracking of moving stimuli (Dubowitz, Dubowitz, Morante, & Verghote, 1980; Bloch, 1983) have been described in preterm infants compared to full-terms of the same (corrected) age. On the other hand, it has been suggested that the immature visual system might suffer from early exposure to visual stimulation (Friedman, Jacobs, & Werthmann, 1981; Turkewitz & Kenny, 1985). In accord with this, there are several studies reporting that preterm infants have longer look durations in habituation studies (Rose, Feldman, McCarton, & Wolfson, 1988; Spungen, Kurtzberg, & Vaughan, 1985) as well as difficulties localizing new stimuli (Landry, Leslie, Fletcher, & Francis, 1985). In free play, they also pay less attention to toys (Landry & Chapieski, 1988). It has been shown that preterm infants problems concerning visual processing and recognition memory persist throughout the first year of life (Rose, Feldman, & Jankowski, 2001; Rose, 1983). In later childhood these infants tend to have lower scores on attention tests (Taylor, Hack, & Klein, 1998) and cognitive scales (Wolke & Meyer, 1999). Even in early adolescence, they have been shown to be at risk concerning their intellectual functioning (Botting, Powls, Cooke, & Marlow, 1998). To summarize, there are two major reasons to examine preterm infants develop- 19

21 Chapter 1 ment of vision and attention: As described above, preterm infants tend to perform more poorly on tasks requiring attentional and processing skills, even later in childhood. Sorting out precisely how the development of vision and attention is different in preterm infants is crucial in order to understand their eventual deficits and to develop suitable and effective interventions. On the other hand, examining healthy preterm and full-term infants is of great scientific interest, as it offers the possibility to learn more about the roles of maturation and experience in visual and attentional development. The Role of the Nature of the Stimuli In 1977, Urie Bronfenbrenner provokingly postulated that much of contemporary developmental psychology is the science of the strange behavior of children in strange situations for the briefest possible periods of time (Bronfenbrenner, 1977, p. 513). Still, more than 25 years after Bronfenbrenner s well-known criticism and about 60 years after Egon Brunswik introduced the concept of ecological validity (Brunswik, 1943), developmental psychologists struggle with the demand of producing research that allows generalizing to real world phenomena. A prominent topic in the debate around ecological validity of experimental studies is the choice of stimuli (Schmuckler, 2001). The stimulus material used for instance in studies on cognitive processes has been criticized as abstract, discontinuous and marginally real, and results of these kind of studies have been suspected to be irrelevant to the phenomenon that one would like to explain (Neisser, 1976, p. 33, 34). Lewkowicz (2001) correctly remarks that setting up an experiment with ecologically valid stimulus material does not necessarily mean creating conditions which are as naturalistic as possible. Instead, it is essential to identify the relevant aspects of the natural environment that control the infant s responsiveness and to capture them in the stimuli used. However, when examining the current research on infants development of perception and attention, one can still question the ecological validity of a large number of studies. For example, much we know regarding attention, perceptual responsiveness, and information processing during infancy is based on experiments using unimodal stimuli, although it has been shown that multimodal stimulation can elicit enhanced responsiveness (Bahrick, 1992, 1994). Another example are studies on the perception and scanning of faces. Most of these studies have been carried out with schematic drawings (e.g., Maurer & Maurer, 1988; Caron, Caron, Caldwell, & Weiss, 1973) or photographs of faces (e.g., Hainline, 1978) or manikins (e.g., Carpenter, 1974). When real faces have been used, they often were still faces (e.g., Maurer & Salapatek, 1976; Bronson, 1982). The generalizability of the findings from these studies is unknown. At the same time, there are several reasons to assume that infants reactions to a naturally moving, smiling face might be considerably different. First, it has been demonstrated that moving stimuli both 20

22 Introduction faces (Wilcox & Clayton, 1968; Haith, Bergman, & Moore, 1977) and non-face stimuli (Tronick, 1972) attract more attention in infants. Further, there are indications that moving stimuli are regarded differently than static displays (Bronson, 1990; Girton, 1979; Johnson & Johnson, 2000). Examples like these emphasize the importance of complying with the demands of ecological validity, especially when investigating young infants skills and competences. Issues of Further Investigation and Goals of the Study The developmental changes which occur in the visual attentional behavior of young infants have been studied in detail. However, nearly all of those studies have used only unnatural, often abstract stimulus material. Information on the early development of attention and eye movements in the context of natural stimuli is largely missing (but see e.g., Bornstein & Ludemann, 1989). Consequently, the first aim of the current study was to fill in this gap by examining the development of two important attentional skills disengagement of attention and visual scanning using two carefully selected stimulus types. As an ecologically valid stimulus, a video recording of the face of each infant s mother was used. Further, it was chosen to present the mother s face in a natural way. Thus, for the video recording she was filmed moving and smiling as she would do during a face-to-face interaction with her baby. In order to test the influence of different sorts of stimuli, it was also chosen to use a second, abstract stimulus which matched the actual mother video. The majority of the studies on the development of gaze and attention shifting and visual scanning presents only cross-sectional data with broad age intervals. Studies which provide longitudinal data with several dense measurement occasions are scarce (but see e.g., Butcher, Kalverboer, & Geuze, 2000). However, only the latter type of research can provide detailed information on developmental trajectories, the timing and tempo of developmental change and inter-infant differences concerning this change. In this study, an intensive, longitudinal design was used. The intervals between measurements were kept short, because rapid development was expected during the measurement period. Using this design avoids the random variance that arises when different groups of infants are compared at different ages and allows studying and comparing the developmental trajectories of the two attentional mechanisms chosen as well as the interindividual variance of this development. As described earlier, infants born prematurely are exposed to visual stimulation earlier in their development and, at the same time, have been shown to have an increased risk for later attentional problems. In early infancy both inferior as well as enhanced visual functioning has been described and theoretically underpinned. Further research on the early development of fundamental visual attentional mechanisms is needed. A third goal of this study was thus to compare the development of attention and gaze shifting observed in a group of healthy infants 21

23 Chapter 1 to the developmental trajectory found in a group of preterm infants. Only in the recent years, a technique to measure eye movements in very young infants reliably and non-intrusively has become available. In order to describe scanning patterns properly, it is necessary to rest upon precise and solid data on shifts of gaze as well as locations of fixation points. Therefore, the current study makes use of the latest eye-tracking techniques. Outline of the Dissertation This thesis is divided into six chapters. Following this 1st introductory chapter, the development of visual scanning is addressed in Chapter 2. The eye movements and fixations while scanning a naturally moving face and a matched abstract stimulus are registered throughout the first few months of infancy. The development of attention and gaze shifting between the two types of stimuli is examined in Chapter 3. In Chapter 4, the association between the development of these skills visual scanning and gaze shifting is explored. While the Chapters 2, 3, and 4 are devoted to the development of full-term infants with no history of medical complications, Chapter 5 deals with the comparison of a group of preterm infants and a group of full-terms. Again, the development of gaze shifting between faces and abstract stimuli is analyzed. The last chapter, Chapter 6, presents summaries of the preceding chapters and ends with a discussion of the general conclusions to be drawn from this study, its limitations and its implications for further research. REFERENCES Aslin, R. N. (1981). Development of smooth pursuit in human infants. In D. F. Fisher, R. A. Monty, & J. W. Senders (Eds.), Eye movements: Cognition and visual perception (pp ). Hillsdale, NJ: Erlbaum Associates, Inc. Aslin, R. N., & McMurray, B. (2004). Automated corneal-reflection eye-tracking in infancy: Methodological developments and applications to cognition. Infancy, 6, Aslin, R. N., & Salapatek, P. (1975). Saccadic localization of visual targets by the very young human infant. Perception and Psychophysics, 17, Atkinson, J. (1984). Human visual development over the first six months of life. Human Neurobiology, 3, Atkinson, J. (1992). Early visual development: Differential functioning of parvocellular and magnocellular pathways. Eye, 6, Atkinson, J., & Braddick, O. (2003). Neurobiological models of normal and abnormal visual development. In M. de Haan & M. H. Johnson (Eds.), The cognitive neuroscience of development (pp ). Hove, UK: Psychology Press. Atkinson, J., Hood, B. M., Wattam-Bell, J., Anker, S., & Tricklebank, J. (1988). Development of orientation discrimination in infancy. Perception, 17,

24 Introduction Bahrick, L. E. (1992). Infants perceptual differentiation of amodal and modality specific audio-visual relations. Journal of Experimental Child Psychology, 53, Bahrick, L. E. (1994). The development of infants sensitivity to arbitrary intermodal relations. Ecological Psychology, 6, Bayley, N. (1969). Manual for the Bayley Scales of Infant Development. New York: Psychological Corporation. Bayley, N. (1993). Bayley Scales of Infant Development (Second Edition). San Antonio, TX: The Psychological Corporation. Bloch, H. (1983). La poursuite visuelle chez le nouveau-né à terme et chez le prématuré. Enfance, 1, Bornstein, M. H. (1985). Habituation of attention as a measure of visual processing in human infants: Summary, systematization, and synthesis. In G. Gottlieb & N. A. Krasnegor (Eds.), Measurement of audition and vision in the first year of postnatal life: A methodological overview (pp ). Norwood, NJ: Ablex. Bornstein, M. H., & Ludemann, P. M. (1989). Habituation at home. Infant Behavior and Development, 12, Botting, N., Powls, A., Cooke, R. W., & Marlow, N. (1998). Cognitive and educational outcome of very-low-birthweight children in early adolescence. Developmental Medicine and Child Neurology, 40, Braddick, O., Atkinson, J., French, J., & Howland, H. C. (1979). A photorefractive study of infant accommodation. Vision Research, 19, Brazelton, T. B., & Nugent, J. K. (1995). The Neonatal Behavioral Assessment Scale. London: Mac Keith Press. Bronfenbrenner, U. (1977). Toward an experimental ecology of human development. American Psychologist, 32, Bronson, G. (1974). The postnatal growth of visual capacity. Child Development, 45, Bronson, G. W. (1982). The scanning patterns of human infants: Implications for visual learning. Monographs on Infancy No. 2. Norwood, NJ: Ablex. Bronson, G. W. (1983). Potential sources of error when applying a corneal reflex eyemonitoring technique to infant subjects. Behavior Research Methods and Instrumentation, 15, Bronson, G. W. (1990). Changes in infants visual scanning across the 2- to 14-week age period. Journal of Experimental Child Psychology, 49, Bronson, G. W. (1991). Infant differences in rate of visual encoding. Child Development, 62, Bronson, G. W. (1994). Infants transitions towards adult-like scanning. Child Development, 65, Bronson, G. W. (1996). The growth of visual capacity: Evidence from infant scanning patterns. In C. Rovee-Collier & L. P. Lipsitt (Eds.), Advances in infancy research (Vol. 11, pp ). Norwood, NJ: Ablex. 23

25 Chapter 1 Brunswik, E. (1943). Organismic achievement and environmental probability. The Psychological Review, 50, Bushnell, I. W. R. (1979). Modification of the externality effect in young infants. Journal of Experimental Child Psychology, 28, Butcher, P. R., Kalverboer, A. F., & Geuze, R. H. (2000). Infants shifts of gaze from a central to a peripheral stimulus: A longitudinal study of development between 6 and 26 weeks. Infant Behavior and Development, 23, Canfield, R. L., & Haith, M. M. (1991). Young infants visual expectations for symmetric and asymmetric stimulus sequences. Developmental Psychology, 27, Caron, A. J., Caron, R. F., Caldwell, R. C., & Weiss, S. J. (1973). Infant perception of the structural properties of the face. Developmental Psychology, 9, Carpenter, G. C. (1974). Visual regard of moving and stationary faces in early infancy. Merrill-Palmer Quarterly, 20, Chugani, H. T. (1994). Development of regional brain glucose metabolism in relation to behavior and plasticity. In G. Dawson & K. Fischer (Eds.), Human behavior and the developing brain (pp ). New York: Guilford. Clavadetscher, J. E., Brown, A. M., Ankrum, C., & Teller, D. Y. (1988). Spectral sensitivity and chromatic discriminations in 3- and 7-week-old infants. Journal of the Optical Society of America A, 5, Colombo, J. (1993). Infant cognition: Predicting later intellectual functioning. Newbury Park, CA: Sage. Conel, J. L. ( ). The postnatal development of the human cerebral cortex (Vol. 1-8). Cambridge, MA: Harvard University Press. Cowey, A. (1994). Cortical visual areas and the neurobiology of higher visual processes. In M. J. Farah & G. Radcliff (Eds.), The neuropsychology of high-level vision (pp. 3-31). Hillsdale, NJ: Erlbaum Associates, Inc. de Haan, M., & Johnson, M. H. (2003). Mechanisms and theories of brain development. In M. de Haan & M. H. Johnson (Eds.), The cognitive neuroscience of development (pp. 1-18). Hove, UK: Psychology Press. Dewey, E. (1935). Behavior development in infants. New York: Columbia University Press. Dubowitz, L. M., Dubowitz, V., Morante, A., & Verghote, M. (1980). Visual function in the preterm and fullterm newborn infant. Developmental Medicine and Child Neurology, 22, Fantz, R. L., Ordy, J. M., & Udelf, M. S. (1962). Maturation and pattern vision in infants during the first six months. Journal of Comparative and Physiological Psychology, 55, Fantz, R. L., & Yeh, J. (1979). Configurational selectivities: Critical for development of visual perception and attention. Canadian Journal of Psychology, 33, Field, T. M. (1979). Visual and cardiac responses to animate and inanimate faces by young term and preterm infants. Child Development, 50,

26 Introduction Fielder, A. R., Foreman, N., Moseley, M. J., & Robinson, J. (1993). Prematurity and visual development. In K. Simons (Ed.), Early visual development, normal and abnormal (pp ). New York: Oxford University Press. Friedman, S. L., Jacobs, B. S., & Werthmann, M. W. (1981). Sensory processing in pre-and full-term infants in the neonatal period. In S. L. Friedman & M. Sigman (Eds.), Preterm birth and psychological development (pp ). New York: Academic Press. Girton, M. R. (1979). Infants attention to intrastimulus motion. Journal of Experimental Child Psychology, 28, Greenough, W. T., Black, J. E., & Wallace, C. S. (1987). Experience and brain development. Child Development, 58, Hainline, L. (1978). Developmental changes in visual scanning of face and nonface patterns by infants. Journal of Experimental Child Psychology, 25, Hainline, L. (1981). An automated eye movement recording system for use with human infants. Behavior Research Methods and Instrumentation, 13, Hainline, L. (1998). The development of basic visual abilities. In A. Slater (Ed.), Perceptual development. Visual, auditory, and speech perception in infancy (pp. 5-50). Hove, UK: Psychology Press. Hainline, L., & Abramov, I. (1992). Assessing visual development: Is infant vision good enough? In C. Rovee-Collier & L. P. Lipsitt (Eds.), Advances in infancy research (Vol. 7, pp ). Norwood, NJ: Ablex. Haith, M. M. (1969). Infrared television recording and measurement of ocular behavior in the human infant. American Psychologist, 24, Haith, M. M. (1980). Rules that babies look by. Hillsdale, NJ: Erlbaum Associates, Inc. Haith, M. M., Bergman, T., & Moore, M. J. (1977). Eye contact and face scanning in early infancy. Science, 198, Haith, M. M., Hazan, C., & Goodman, G. S. (1988). Expectation and anticipation of dynamic visual events by 3.5-month-old babies. Child Development, 59, Harris, C. M., Hainline, L., & Abramov, I. (1981). A method for calibrating an eye-monitoring system for use with human infants. Behavior Research Methods and Instrumentation, 13, Haynes, H., White, B. L., & Held, R. (1965). Visual accommodation in human infants. Science, 148, Holditch-Davis, D., & Black, B. (2003). Care of preterm infants: Programs of research and their relationship to developmental science. In J. J. Fitzpatrick (Series Ed.), M. Miles & D. Holditch-Davis (Vol. Eds.), Annual review of nursing research: Vol. 2. Research on child health and pediatric issues (pp ). New York: Springer. Hood, B. M. (1995). Shifts of visual attention in the human infant: A neuroscientific approach. In C. Rovee-Collier & L. P. Lipsitt (Eds.), Advances in infancy research (Vol. 9, pp ). Norwood, NJ: Ablex. Hood, B. M., & Atkinson, J. (1993). Disengaging visual attention in the infant and adult. Infant Behavior and Development, 16,

27 Chapter 1 Hood, B. M., Murray, L., King, F., Hooper, R., Atkinson, J., & Braddick, O. (1996). Habituation changes in early infancy: Longitudinal measures from birth to 6 months. Journal of Reproductive and Infant Psychology, 14, Holditch-Davis, D., & Black, B. P. (2003). Care of preterm infants: programs of research and their relationship to developmental science. Annual Review of Nursing Research, 21, Hopkins, B., & van Wulfften Palthe, T. (1985). Staring in infancy. Early Human Development, 12, Hunt, J. V., & Rhodes, L. (1977). Mental development of preterm infants during the first year. Child Development, 48, James, W. (1890). Principles of psychology. New York: Holt. Johnson, M. H. (1990). Cortical maturation and the development of visual attention in early infancy. Journal of Cognitive Neuroscience, 2, Johnson, M. H. (1995). The development of visual attention: A cognitive neuroscience perspective. In M. S. Gazzaniga (Ed.), The cognitive neurosciences (pp ). Cambridge, MA: MIT Press. Johnson, M. H., Gilmore, R. O., & Csibra, G. (1998). Toward a computational model of the development of saccade planning. In J. E. Richards (Ed.), Cognitive neuroscience of attention: A developmental perspective (pp ). Hillsdale, NJ: Erlbaum Associates, Inc. Johnson, M. H., & Morton, J. (1991). Biology and cognitive development: The case of face recognition. Oxford, UK: Blackwell. Johnson, M. H., Posner, M. I., & Rothbart, M. K. (1991). Components of visual orienting in early infancy: Contingency learning, anticipatory looking, and disengaging. Journal of Cognitive Neuroscience, 3, Johnson, S. P., & Johnson, K. L. (2000). Early perception-action coupling: Eye movements and the development of object perception. Infant Behavior and Development, 23, Johnson, S. P., Slemmer, J. A., & Amso, D. (2004). Where infants look determines how they see: Eye movements and development of object perception. Infancy, 6, Kaye, K., & Fogel, A. (1980). The temporal structure of face-to-face communication between mothers and infants. Developmental Psychology, 16, Keller, H., & Gauda, G. (1987). Eye contact in the first months of life and its developmental consequences. In H. Rauh & H.-Ch. Steinhausen (Eds.), Advances in Psychology, No. 46. Psychobiology and Early Development (pp ). Amsterdam: Elsevier. Landry, S. H., Leslie, N. A., Fletcher, J. M., & Francis, D. J. (1985). Visual attention skills of premature infants with and without intraventricular hemorrhage. Infant Behavior and Development, 8, Landry, S. H., & Chapieski, M. L. (1988). Visual attention during toy exploration in preterm infants: Effects of medical risk and maternal interaction. Infant Behavior and Development, 11,

28 Introduction Lewkowicz, D. J. (2001). The concept of ecological validity: What are its limitations and is it bad to be invalid? Infancy, 2, Livingstone, M., & Hubel, D. H. (1988). Segregation of form, color, movement and depth: anatomy, physiology and perception. Science, 240, Maurer, D., & Maurer, C. (1988). The world of the newborn. New York: Basic Books. Maurer, D., & Salapatek, P. (1976). Developmental changes in the scanning of faces by young infants. Child Development, 47, McMurray, B., & Aslin, R. N. (2004). Anticipatory eye movements as a window on infants auditory and visual categories. Infancy, 6, Merigan, W. H., & J. H. R. Maunsell (1993). How parallel are the primate visual pathways? Annual Review of Neuroscience, 16, Milewski, A. (1976). Infant s discrimination of internal and external pattern elements. Journal of Experimental Child Psychology, 22, Mohn, G., & van Hof-van Duin, J. (1986). Development of the binocular and monocular visual fields of human infants during the first year of life. Clinical Vision Science, 1, Müller, J. (1826). Zur vergleichenden Physiologie des Gesichtssinnes des Menschen und der Thiere nebst einem Versuch über die Bewegungen der Augen und über den menschlichen Blick. Leipzig, Germany: C. Cnobloch. Neisser, U. (1976). Cognition and reality: Principles and implications of cognitive psychology. San Francisco: Freeman. Nowakowski, R. S. (1987). Basic concepts of CNS development. Child Development, 58, Piaget, J. (1936). La naissance de l intelligence chez l enfant. Neuchâtel, Switzerland: Delachaux et Niestlé. Piaget, J. (1937). La construction du réel chez l enfant. Neuchâtel, Switzerland: Delachaux et Niestlé. Pratt, K. C. (1954). The neonate. In L. Carmichael (Ed.), Manual of child psychology (2nd ed., pp ). New York: Wiley. Prechtl, H. F. R. (1984). Continuity and change in early neural development. In H. F. R. Prechtl (Ed.), Continuity of neural functions from prenatal to postnatal life. Clinics in Developmental Medicine No. 94 (pp 1-15). Oxford, UK: Blackwell. Preyer, W. (1882). Die Seele des Kindes. Beobachtungen über die geistige Entwicklung des Menschen in den ersten Lebensjahren. Leipzig, Germany: Th. Grieben s Verlag. Rakic, P. (1988). Specification of cerebral cortical area. Science, 241, Richards, J. E., & Hunter, S. K. (1998). Attention and eye movement in young infants: Neural control and development. In J. E. Richards (Ed.), Cognitive neuroscience of attention: A developmental perspective (pp ). Mahwah, NJ: Erlbaum Associates, Inc. Rizzolatti, G., Riggio, L., Dascola, I., & Umiltà, C. (1987). Reorienting attention across the horizontal and vertical meridians: Evidence in favor of a premotor theory of attention. Neuropsychologia, 25,

29 Chapter 1 Rose, S. A. (1983). Differential rates of visual information processing in full-term and preterm infants. Child Development, 54, Rose, S. A., Feldman, J. F., & Jankowski, J. J. (2001). Attention and recognition memory in the 1st year of life: A longitudinal study of preterm and full-term infants. Developmental Psychology, 37, Rose, S. A., Feldman, J. F., McCarton, C. M., Wolfson, J. (1988). Information processing in seven-month-old infants as a function of risk status. Child Development, 59, Salapatek, P. (1975). Pattern perception in early infancy. In L. B. Cohen & P. Salapatek (Eds.), Infant perception: From sensation to cognition (Vol. 1, pp ). New York: Academic Press. Salapatek, P., & Kessen, W. (1966). Visual scanning of triangles by the human newborn. Journal of Experimental Child Psychology, 3, Schiller, P. (1985). A model for the generation of visually guided saccadic eye movements. In D. Rose & V. G. Dobson (Eds.), Models of the visual cortex (pp ). Chichester, UK: Wiley. Schiller, P. (1998). The neural control of visually guided eye movements. In J. E. Richards (Ed.), Cognitive neuroscience of attention (pp. 5-50). Mahwah, NJ: Erlbaum Associates, Inc. Schmuckler, M. A. (2001). What is ecological validity? A dimensional analysis. Infancy, 2, Schneider, G. E. (1969). Two visual systems. Science, 163, Schölmerich, A., Leyendecker, B., & Keller, H. (1995). The study of early interaction in a contextual perspective: Culture, communication, and eye contact. In J. Valsiner (Ed.), Child development within culturally structured environments. Vol. III: Comparative-cultural and constructivist perspectives (pp ). Norwood, NJ: Ablex. Shapley, R., & Perry, V. H. (1986). Cat and monkey retinal ganglion cells and their visual functional roles. Trends in Neurosciences, 9, Slater, A. (1995). Individual differences in infancy and later IQ. Journal of Child Psychology and Psychiatry and Allied Disciplines, 36, Slater, A., Morison, V., & Rose, D. (1983). Locus of habituation in the human newborn. Perception, 12, Slater, A., Morison, V, & Somers, M. (1988). Orientation discrimination and cortical functioning in the human newborn. Perception, 17, Sokol, S., & Jones, K. (1979). Implicit time of pattern evoked potentials in infants: An index of maturation of spatial vision. Vision Research, 19, Sprague, J. M., & Meikle, T. H. (1965). The role of the superior colliculus in visually guided behavior. Experimental Neurology, 11, Spungen, L. B., Kurtzberg, D., & Vaughan, H. G. (1985). Patterns of looking behavior in full-term and low birth weight infants at 40 weeks post-conceptional age. Developmental and Behavioral Pediatrics, 6,

30 Introduction Stechler, G., & Latz, E. (1966). Some observations on attention and arousal in the human infant. Journal of the American Academy of Child Psychology, 5, Stelmach, L. B., Campsall, J. M., & Herdman, C. M. (1997). Attentional and ocular movements. Journal of Experimental Psychology: Human Perception and Performance, 23, Stifter, C. A., & Moyer, D. (1991). The regulation of positive affect: Gaze aversion activity during mother-infant interaction. Infant Behavior and Development, 14, Stone, L. J., Smith, H. T., & Murphy, L. B. (Eds.) (1973). The competent infant: Research and commentary. New York: Basic Books. Tauber, E. S., & Koffler, S. (1966). Optomotor responses in human infants to apparent motion: Evidence of innateness. Science, 152, Taylor, H. G., Hack, M., & Klein, N. K. (1998). Attention deficits in children with <750 gm birth weight. Child Neuropsychology, 4, Teller, D. Y., & Bornstein, M. (1987). Infant color vision and color perception. In P. Salapatek & L. B. Cohen (Eds.), Handbook of infant perception (Vol. 1, pp ). Orlando, FL: Academic Press. Trevarthen, C. B. (1968). Two mechanisms of vision in primates. Psychologische Forschung, 31, Tronick, E. (1972). Stimulus control and the growth of the infant s effective visual field. Perception and Psychophysics, 11, Turkewitz, G., & Kenny, P. A. (1985). The role of developmental limitations of sensory/ perceptual organization. Journal of Developmental and Behavioral Pediatrics, 6, Ungerleider, L. G., & Mishkin, M. (1982). Two cortical visual systems. In D. J. Ingle, M. A. Goodale, & R. J. W. Mansfield (Eds.), Analysis of visual behavior (pp ). Cambridge, MA: MIT Press. Van Essen, D. C. (1985). Functional organization of primate visual cortex. In A. Peters & E. G. Jones (Eds.), Cerebral cortex (Vol. 3, pp ). New York: Plenum. Van Essen, D. C., & Maunsell, J. H. R. (1983). Hierarchical organization and functional streams in the visual cortex. Trends in Neurosciences, 6, von Hofsten, C., & Rosander, K. (1996). Development of smooth pursuit tracking in young infants. Vision Research, 37, Wilcox, B. M., & Clayton, F. L. (1968). Infant visual fixation on motion pictures of the human face. Journal of Experimental Child Psychology, 6, Wolke, D. (1987). Environmental and developmental neonatology. Journal of Reproductive and Infant Psychology, 5, Wolke, D., & Meyer, R. (1999). Cognitive status, language attainment, and prereading skills of 6-year-old very preterm children and their peers: The Bavarian Longitudinal Study. Developmental Medicine and Child Neurology, 41, Wright, R. D., & Ward, L. M. (1998). The control of visual attention. In R. D. Wright (Ed.), Visual attention (pp ). New York: Oxford University Press. 29

31

32 Chapter 2 Developmental Changes in Visual Scanning of Dynamic Faces and Abstract Stimuli in Infants Abstract The characteristics of scanning patterns between the ages of 6 and 26 weeks were investigated through repeated assessments of 10 infants. Eye movements were recorded using a corneal reflection system while the infants looked at two dynamic stimuli: the naturally moving face of their mother and an abstract stimulus. Results indicated that the way infants scanned these stimuli stabilized only after 18 weeks, which is slightly later than the ages reported in the literature on infants scanning of static stimuli. This effect was especially prominent for the abstract stimulus. From the 14-week session on, infants adapted their scanning behavior to the stimulus characteristics. When scanning the video of their mother s face, infants directed their gaze at the mouth and eye region most often. Even at the youngest age, there was no indication of an edge effect. This chapter is published as: Hunnius, S., & Geuze, R. H. (2004). Developmental changes in visual scanning of dynamic faces and abstract stimuli in infants: A longitudinal study. Infancy, 6,

33 Chapter 2 INTRODUCTION When infants are born, their motor skills are very limited, and the fact that they have very little control over their limbs restricts the way in which they are able to explore the world around them. The oculomotor system unlike other motor systems approximates its mature state several months after birth. The infant exercises this system every day from birth on. This makes vision one of the most important channels through which babies learn about the world surrounding them. However, during the first months of life, eye movements and visual acuity are also subject to certain constraints. For example, during the first month, eye movements are often inaccurate and unreliable (Atkinson, 1992), and infants of 1 to 3 months of age who are fixating a salient stimulus often have difficulty shifting their gaze away toward another interesting stimulus (Stechler & Latz, 1966; Butcher, Kalverboer, & Geuze, 2000; see Chapter 3). Development of Scanning Scanning is the pattern of eye movements to fixate different parts of an object of examination. As the visual system matures, the characteristics of infants scanning patterns change. At birth, infants already show complex active scanning behaviors, but they often fail to direct their gaze toward a stimulus present in their field of vision (Haith, 1980; Bronson, 1990a). Around the age of 1 month, babies tend to look especially at edges or outer contours of a stimulus pattern, but they mostly ignore the inner parts of a stimulus (Salapatek, 1975; Milewski, 1976). They also have the tendency to fixate single locations of stimuli for long periods, a behavior that has been referred to as staring (Bronson, 1996). Infants of 2 months of age scan more locations and different features of stimuli (Bronson, 1982, 1990a, 1996; Salapatek & Kessen, 1966; Leahy, 1976). The changes in infants scanning behavior can be described as a gradual transition to adult-like scanning (Bronson, 1994). As infants grow older, they engage more frequently in an advanced scanning mode, showing more brief fixations and more extensive scanning, whereas the infant-like way of scanning characterized by periods with extremely long fixations directed to single salient features becomes less prominent. Around 3 months, infants can make accurate, efficient eye movements. Salient features in the environment still attract their gaze, but they have gained volitional, strategic control over their scanning behavior (Bronson, 1994). It is known from studies with adults that characteristics of the stimuli influence the way they are scanned. Changes in the way of scanning have been shown to be related to physical components of the stimuli such as luminance or texture (Mackworth & Morandi, 1967; Antes, 1974; Buswell, 1935) but also to semantic aspects (Loftus & Mackworth, 1978) or familiarity (Althoff & Cohen, 1999) and experience (Chapman & Underwood, 1998). There are indications that infants adapt their scanning patterns to stimulus characteristics increasingly more during the first few months of life (Johnson 32

34 Developmental Changes in Visual Scanning & Johnson, 2000). Further, it seems that stimulus characteristics can elicit more or less mature scanning behavior in young infants. For example, if the internal elements of a stimulus are moving, they are more likely to attract the gaze of even 1-month-old infants (Bushnell, 1979), whereas in infants of 13 weeks of age a flickering stimulus evokes a scanning pattern similar to that observed with a static stimulus at a much younger age (Bronson, 1990a). Face Scanning Human faces are among the most important stimuli in the visual environment of a baby. Infants are confronted with faces very frequently from birth on and show attraction for faces or facelike stimuli already at an early age (Fantz, 1961; Morton & Johnson, 1991). Moreover, faces are of large psychosocial significance for a baby because mostly they appear in sight and draw the infant s attention in a situation of interaction and communication. Infants reactions to faces have been studied frequently, and there has also been considerable interest in how exactly infants examine faces. Infants younger than 2 months of age show limited scanning mostly of the perimeter of faces, whereas infants older than 2 months become more likely to fixate the internal elements of faces. When looking at the internal features of a face, they pay most attention to the eyes (Haith, Bergman, & Moore, 1977; Maurer & Salapatek, 1976). The borders of the face have been shown to be attractive regions of the face even for infants of 2 or 3 months of age (Haith et al., 1977; Hainline, 1978), whereas the mouth is looked at quite seldom (Haith et al., 1977). Previous studies thus suggest a developmental shift from scanning directed at the edges of a stimulus to scanning of the internal elements similar to that described earlier for geometric stimuli. At least for schematic faces, there are indications that the effect of edge attraction might be less prominent if the internal features of the face are moving (Girton, 1979). Young infants are usually exposed to faces in social situations. Adults (especially parents ) interactions with babies have been shown to be highly similar across individuals even from different cultures and to be characterized by elements like attracting attention and eye contact, displaying greeting responses, and making exaggerated facial movements and signals of pleasure (Papoušek & Papoušek, 1987). It is therefore remarkable that only very few studies have made use of moving or talking faces as stimuli (e.g., Haith et al., 1977). Most studies focusing on young infants scanning patterns of faces have used photographs or drawings of faces (e.g., Hainline, 1978), and when real faces have been used, they often were still faces (e.g., Maurer & Salapatek, 1976; Bronson, 1982). The use of stimuli that are of limited ecological validity has been addressed and criticized frequently (Neisser, 1976; Schmuckler, 2001), as the generalizability of the obtained results is questionable. This study examines how infants scan naturally moving faces like those they are confronted with in daily life. 33

35 Chapter 2 Measuring Eye Movements in Infants The technique of infrared corneal photography was first applied to human infants in the 1960s (Salapatek & Kessen, 1966; Haith, 1969) and always had to cope with a number of problems inherent to using a complex and highly sensitive technique with delicate and unpredictable subjects. The technical progress and the experience of the last 40 years have enabled several improvements of the original method, which enhance the quality and accuracy of the measurements. Examples are the increase of the sampling rate from 2-6 Hz to Hz (Aslin, 1981; Hainline, 1981; Bronson, 1982) to improve the accuracy in determining fixation durations and the implementation of a calibration (Bronson, 1983; Harris, Hainline, & Abramov, 1981) to enhance the spatial accuracy of the measurement. Incorporating a head-tracker into the eyetracking system now has reduced the need to restrain the infant s head, which was quite unnatural and often distressing for the infant. This study combines the recent improvements in infant eye-tracking techniques with an intense longitudinal design. It also demonstrates the limitations of measuring eye movements in young infants and reports ways of handling the emerging problems of optimizing the data acquisition and dealing with missing data in a longitudinal design. Aims of the Study Several studies have compared infants attending to moving and static stimuli, and the attention-enhancing effect of stimulus movement has been described frequently (e.g., Wilcox & Clayton, 1968; Tronick, 1972; Carpenter, 1974). There are also indications that infants might scan moving stimuli in a different way than static displays (Bronson, 1990a; Johnson & Johnson, 2000). However, whereas the development of scanning of static stimuli has been studied widely, it is still unknown how infants scan non-static stimuli and how scanning of dynamic stimuli develops throughout infancy. Thus, our first goal was to describe infants scanning of dynamic stimuli and its development during the first few months of infancy. We chose to investigate infants scanning of two different dynamic stimuli: the infant s mother s face as it moved in a natural way and an abstract stimulus. The scanning of these two stimuli was examined between 6 and 26 weeks of age because this age period covers an interval in which the infant s oculomotor and visual system is changing rapidly (see e.g., Atkinson, 1984). Infants were examined with intervals of 4 weeks between test sessions to provide enough measurements to precisely describe the development on the one hand but prevent possible training and habituation effects on the other. As sketched above, stimulus characteristics might influence the way stimuli are scanned already in infancy. Thus, our second goal was to examine whether the scanning patterns elicited by a moving face differed from the scanning patterns elicited by an abstract dynamic stimulus and from which age on infants tailored their scanning behavior to the characteristics of the stimuli. To obtain an abstract stimulus that was 34

36 Developmental Changes in Visual Scanning comparable to the face video with regard to movement, colors, and luminance, but which had no facelike characteristics, the video of the mother s face was scrambled. As stimulus characteristics might not only influence the way infants scan them but also how these scanning patterns evolve, we also investigated whether there were differences in how the scanning patterns of the mother and the abstract stimulus developed during the first few months of life. The third goal of the study was to determine which regions of the face were frequently looked at and how this developed as infants were growing older. The question was whether the patterns of scanning of faces found in earlier studies that used less ecologically relevant representations of faces or even static faces are also found for naturally moving faces. We expected that the movement of highly salient internal features would reduce the clear effect of edge preference in very young infants and the frequency of edge fixations in the older infants. We also expected that it would lead to a more equal distribution of fixations over the regions of the face around the different internal features in contrast to results of earlier studies that have emphasized the infants preference to fixate the eye region. METHOD Participants Ten infants (5 girls; 5 boys) participated in the longitudinal study. The mothers of the infants were contacted through childbirth education classes, midwives, or gym classes. Parents were told about the course and goals of the study and gave their written informed consent. The research was approved by the local Medical Ethics Committee. All infants had been born after a gestation period between 37 and 42 weeks, had a birth weight above 2800 g, and no history of pre- or perinatal complications. All infants scored within their age range on the Bayley Scales of Infant Development (BSID-II; Bayley, 1993) at 12 and 24 weeks of age. Measurement sessions were conducted at 6, 10, 14, 18, 22, and 26 weeks, calculated from the due date. Mean ages were 46.7 days (SD = 3.8), 73.0 days (SD = 2.4), days (SD = 3.4), days (SD = 2.3), days (SD = 3.4), and days (SD = 4.4). If a measurement session was unsuccessful because the infant was not in the required state of alert wakefulness or other problems occurred, a new appointment was made. Despite attempts to retest, data for an additional 10 infants were not included because fewer than 5 of their 6 test sessions could be analyzed due to fussiness, crying, or technical problems. Nine of the 10 infants included in the analysis completed all 6 test sessions; one infant completed 5 sessions. Procedure Appointments were made at a time of the day when the mothers expected their baby to be awake and able to stay alert for about half an hour. After arriving at the 35

37 Chapter 2 laboratory, the infant was given some time to get used to the new environment. When the infant was in state 3 or 4 of Prechtl s scale of alertness (awake, eyes open, some spontaneous movements, no crying; Prechtl & Beintema, 1964), the experiment was started. Apparatus To carry out the assessments at the different ages under the same circumstances, a setup suitable for infants from 1 to 6 months of age was developed. The infant was seated in an infant chair in front of a 21 inch monitor at a distance of 35 cm. The seat was tilted backwards (about 45 degrees) to provide enough support for the younger infants and to keep the older ones from leaning forward. The infant s head was slightly stabilized, especially when the infant was young, but head-movements were not severely restricted to avoid distressing the babies and to allow them a natural way of moving while looking at the stimuli. Only the screen of the monitor was visible. The frame and the other equipment necessary to run the experiment and to record eye movements were concealed behind a gray curtain, which filled approximately 180 degrees of the infant s visual field. During the task, one experimenter stood behind the baby to support the infant s head when necessary, while the second experimenter controlled the presentation of the stimuli and the measuring equipment. The baby s face and the display visible to the baby were shown on a video monitor, which made it possible for the experimenter to watch the infant s behavior while running the task. Stimuli Two dynamic stimuli were presented to the baby: a short video of the baby s mother s face, in which the mother was looking, smiling, and nodding to the baby as she would in a normal interaction, and an abstract moving figure. Both stimuli were in color. The video recording of the mother s face was made during a preliminary visit of mother and baby to the lab, shortly before the sessions in which the data were collected began. The mothers were asked to start with an attention getting movement (like nodding or greeting) and then to continue in a way that felt natural to them. As infants are sensitive to other persons gaze direction during interactions (Hains & Muir, 1996), mothers were asked not to avert their gaze during the recording. The videos were digitalized for use in a computerized experimental design. The abstract stimulus was derived from the digital video of the mother by carrying out a number of transformations in a graphic computer program (Corel PHOTO- PAINT 9), such that it no longer resembled a face. During the transformation the image of the mother was rotated, scrambled, and distorted. This frame-by-frame procedure ensured that the two stimuli used were comparable in terms of total dynamics, color, and luminance. One frame from each type of video is presented as a stimulus example in Figure

38 Developmental Changes in Visual Scanning Figure 2.1. Example of (A) the mother stimulus and (B) the abstract stimulus. Both stimuli had the same size and appeared against a gray screen. At the distance of 35 cm, the stimuli were 30 by 40 degrees in size. The luminance of both figures was approximately 50 lux in the center of the stimulus. Each stimulus was presented to the baby for 30 s. The order in which they were shown was pseudorandomized. Measurement of Eye Movements The eye movements of the infants were measured using a corneal reflection eye-tracking system (ASL, model 504). The infrared remote camera was positioned on a pan/tilt base underneath the monitor at a distance of about 50 cm from the eyes of the infant. Eye position data were sampled at 50 Hz. As the infants could move their head in a natural, relatively free way, their head position was tracked by a sensor, which was attached to a soft fabric hat and coupled to a magnetic head-tracker (Polhemus Fastrak). The head position data were used 37

39 Chapter 2 constantly during the experiment to direct the remote camera, which recorded an image of the eye. The camera followed slow horizontal and vertical head-movements automatically and was also able to recover an eye-image quickly after a rapid movement or even a turn away from and back to the monitor. A video recording of the location of the current fixation superimposed on the video display allowed the experimenter to see where the infant was looking at all times. As eye geometries have been shown to differ considerably between infants (Bronson, 1990b), every infant s visual field was calibrated prior to each measurement session. The calibration stimulus was a flashing black-and-white concentric square that appeared on the monitor first in the top left and then in the bottom right corner of the area in which the abstract and the face stimulus were presented. The calibration stimulus was accompanied by a short quacking noise. The experimenters judged whether the baby was fixating the stimulus before carrying out the calibration. Each calibration was tested to evaluate the quality of measurement during the experimental session. A spiral of about 12 degrees in diameter was presented to the infant on the monitor. The spiral moved across the gray screen and stopped at five different locations (the four corners and the center of the subsequent stimulus) where it shrank to a size of about 3 degrees with a gaudy dot in the center. As the baby followed the moving stimulus and fixated the points at which it was still, the experimenter could watch the infant s focus of gaze on the monitors and judge the quality of measurement. If necessary, the calibration procedure was repeated. Analysis Eye position data. To determine the number and length of fixations during stimulus exposure, saccades, which naturally mark the beginning and end of a fixation, had to be identified from the eye position data. The displacement was calculated from the horizontal and vertical eye position data using Pythagoras s theorem and smoothed with a 5-point window (100 ms). The onset of a saccade was defined as (a) pupil diameter above 0 as an indication of a valid signal and (b) change in fixation position larger than a threshold value. The threshold value adopted for saccade onset in this study was a mean displacement of.6 degrees per sample point over 3 sample points (60 ms). This is equivalent to an eye movement velocity of 30 degrees/s. Experience in using eye-tracking methods with infants has taught us that the data yielded by an eye-tracker are often incomplete or contain artifacts. Examples of frequently occurring problems are missing data due to pupil loss after a posture change or a rapid head-movement. Artifacts can be caused by fussing, crying, or screwing up the eyes. Behavioral coding. To correct errors and complete the data set, the video recordings of each infant s face were coded off-line. They were played back half-frame by half-frame (20 ms intervals) and compared with the available information from 38

40 Developmental Changes in Visual Scanning the eye-tracker data files. Looking with narrowed eyes was included if a part of the pupil was visible. The coding was carried out by two observers. The interobserver reliability (across subjects and across infants ages) for the classification of an eye movement indicated by the eye-tracker as a real eye movement versus an artifact was 94.7%. For detecting an eye movement in the absence of an eye-tracker signal, it was 96.9%. The reliability between observers for the onset and the length of an eye movement was 92.5% and 94.6%, respectively. Finally, for judging whether the infant was looking at the stimulus display or not, the agreement was 92.0%. With these corrections and completions on the basis of the videotaped eye movements it was possible to obtain complete data on whether the infant was looking to or away from the stimulus display and on the relative displacement of the eyes. The latter were used to determine the number and duration of fixations. Coding of location of fixations. As the test of calibration carried out before every stimulus presentation revealed, the quality of eye-tracker measurement differed considerably across sessions. When the calibration was successful and allowed a reliable measurement of absolute eye position, data on the actual location of the fixations on the respective stimulus could be collected. Figure 2.2. Scheme of zones for one frame of a mother stimulus. For this frame, the regions of the face consist of the following zones: mouth - 4D, 4E; eyes - 3D, 3E; edge - 1D, 1E, 2C, 2D, 2E, 2F, 3C, 3F, 4C, 4F; body - 5B, 5C, 5D, 5E, 5F, 6A, 6B, 6C, 6D, 6E, 6F, 6G; background - 1A, 1B, 1C, 1F, 1G, 1H, 2A, 2B, 2G, 2H, 3A, 3B, 3G, 3H, 4A, 4B, 4G, 4H, 5A, 5G, 5H, 6H. The location of the fixations was scored off-line by a single observer, using the video recordings of the stimulus display with the sequence of the infant s fixations superimposed on it. Every stimulus was divided into a number of zones, and each fixation was if possible assigned to one of these areas. For the face stimulus, the categories were eyes, mouth, edge (including the hairline and the ears as well as the lower edge of the face), background, and body (neck, hair lying on the shoulders). The zones were determined separately for each stimulus using grid lines (for an example, see Figure 2.2). As the stimulus was moving, the areas could 39