ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2017

Digital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Social Sciences 145

Joint Attention in Development

Insights from Children with Autism and InfantSiblings

EMILIA THORUP

ISSN 1652-9030ISBN 978-91-513-0020-7urn:nbn:se:uu:diva-327117

Dissertation presented at Uppsala University to be publicly examined in Sal IV,Universitetshuset, Biskopsgatan 3, Uppsala, Friday, 22 September 2017 at 13:15 for thedegree of Doctor of Philosophy. The examination will be conducted in English. Facultyexaminer: Peter Mundy (University of California, UC Davis MIND Institute).

AbstractThorup, E. 2017. Joint Attention in Development. Insights from Children with Autismand Infant Siblings. Digital Comprehensive Summaries of Uppsala Dissertations fromthe Faculty of Social Sciences 145. 91 pp. Uppsala: Acta Universitatis Upsaliensis.ISBN 978-91-513-0020-7.

Compared to other children, children with Autism Spectrum Disorder (ASD) are known toengage less in joint attention - the sharing of attention between two individuals toward a commonobject or event. Joint attention behaviors - for example gaze following, alternating gaze, andpointing - play an important role in early development, as they provide a foundation for learningand social interaction. Study I and Study II focused on infant siblings of children with ASD.These infants, often termed high risk (HR) infants, have an increased probability of receivinga later ASD diagnosis. Studying them therefore allows for the detection of early signs of ASD.Live eye tracking was used to investigate different joint attention behaviors at 10 months ofage. Study I showed that omitting the head movement that usually accompany experimenters’eye gaze shifts in similar designs reduced gaze following performance in the HR group, but notin a group of infants at low risk (LR) for ASD. HR infants may thus be less sensitive to eyeinformation, or may need more salient cues in order to follow gaze optimally. Study II focusedon the infants’ tendency to initiate joint attention by alternating their gaze between a personand an event. LR infants engaged more in alternating gaze than HR infants, and less alternatinggaze in infancy was associated with more ASD symptoms at 18 months. This relation remainedwhen controlling for visual disengagement and general social interest in infancy. Study IIIexplored the role of joint attention later in development, by investigating the microstructureof the looking behaviors of autistic and typically developing children (~6 years old). Theresults indicated that seeing somebody look at an object influenced the processing of that objectless in autistic children than in the typically developing controls. Both groups followed gazeeffectively, suggesting that differences in joint attention at this age may be subtle, but detectablewith eye tracking technology. Together, the studies contribute to our understanding of the rolethat joint attention atypicalities play both in the early development of infants at risk for ASD,and later in the development of children with a confirmed diagnosis.

Keywords: Autism Spectrum Disorder, Joint Attention, Gaze following, Alternating gaze,Social cognition, Eye tracking, Infant siblings

Emilia Thorup, Department of Psychology, Box 1225, Uppsala University, SE-75142Uppsala, Sweden.

© Emilia Thorup 2017

ISSN 1652-9030ISBN 978-91-513-0020-7urn:nbn:se:uu:diva-327117 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-327117)

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Thorup, E., Nyström, P., Gredebäck, G., Bölte, S., Falck-Ytter,

T., the EASE Team (2016). Altered gaze following during live interaction in infants at risk for autism: An eye tracking study. Molecular Autism, 7(1):12.

II Thorup, E., Nyström, P., Gredebäck, G., Bölte, S., Falck-Ytter, T., the EASE Team. Reduced alternating gaze during social in-teraction in infancy is associated with elevated symptoms of au-tism in toddlerhood. Manuscript submitted for publication.

III Thorup, E., Kleberg, J. L., Falck-Ytter (2017). Gaze following in children with autism: Do high interest objects boost perfor-mance? Journal of Autism and Developmental Disorders, 47(3):626-635.

Reprints were made with permission from the respective publishers.

Contents

Introduction ..................................................................................................... 9 Autism Spectrum Disorder ....................................................................... 10 

A Brief Historical Note ........................................................................ 10 Prevalence and Comorbidities ............................................................. 11 Defining and Assessing ASD .............................................................. 12 Outcomes and Intervention .................................................................. 13 Causes .................................................................................................. 14 Neurobiology ....................................................................................... 15 Cognitive Models ................................................................................ 15 An Increased Focus on Dimensionality and Heterogeneity ................. 16 The Infant Sibling Paradigm and Early Development ......................... 17 

Joint Attention .......................................................................................... 18 Joint Attention and ASD ...................................................................... 20 Gaze Following and Response to Joint Attention in Typical Development ........................................................................................ 22 Gaze Following and Response to Joint Attention in ASD ................... 23 Studies of Gaze Following and RJA in Infant Siblings ....................... 25 Key Findings and Knowledge Gaps (I) ............................................... 26 Initiation of Joint Attention in Typical Development .......................... 27 Initiation of Joint Attention in ASD .................................................... 28 Studies of IJA in Infant Siblings .......................................................... 29 Key Findings and Knowledge Gaps (II) .............................................. 30 

Aims of the Thesis .................................................................................... 30 

Methods ........................................................................................................ 32 Participants ............................................................................................... 32 

Study I and Study II ............................................................................. 32 Study III ............................................................................................... 33 

Apparatus and Procedures ........................................................................ 34 Study I and Study II ............................................................................. 34 Study III ............................................................................................... 37 

Measures and Analyses ............................................................................ 38 Study I .................................................................................................. 38 Study II ................................................................................................ 39 Study III ............................................................................................... 40 

Study I ........................................................................................................... 42 Background .............................................................................................. 42 Primary Results ........................................................................................ 42 Conclusions .............................................................................................. 43 

Study II ......................................................................................................... 45 Background .............................................................................................. 45 Primary Results ........................................................................................ 46 

Alternating Gaze, Disengagement and Social Preference at 10 Months ................................................................................................. 46 Longitudinal Relations Between 10-month Alternating Gaze and 18-month Symptom Level ................................................................... 47 Longitudinal Relations Between 10-month Alternating Gaze and 18-month Gestural IJA ........................................................................ 48 

Conclusions .............................................................................................. 49 

Study III ........................................................................................................ 50 Background .............................................................................................. 50 Primary Results ........................................................................................ 51 

Gaze Following Accuracy ................................................................... 51 First Fixation Durations ....................................................................... 51 Total Looking Time at Objects ............................................................ 52 Parental Ratings of Circumscribed Interests ........................................ 52 

Conclusions .............................................................................................. 54 

General Discussion ....................................................................................... 55 Early Development ................................................................................... 55 

Gaze Following .................................................................................... 55 Alternating Gaze .................................................................................. 56 General Conclusions from Study I and Study II .................................. 57 

Later Development ................................................................................... 58 Gaze Following Accuracy ................................................................... 58 Looking Time Measures, Object Processing, and Possible Consequences for Learning ................................................................. 59 Do the Findings of Study III Indicate Specifically Social Atypicalities? ....................................................................................... 60 A Final Remark on the Observed Looking Patterns of Children with ASD ..................................................................................................... 61 

Findings with Relevance for Typical Development ................................. 61 Methodological Implications .................................................................... 63 Clinical Implications ................................................................................ 65 Limitations ............................................................................................... 65 Future Directions ...................................................................................... 67 Final Conclusions ..................................................................................... 69 

Summary in Swedish .................................................................................... 70 

Acknowledgements ....................................................................................... 73 

References ..................................................................................................... 75 

Abbreviations

ADI-R Autism Diagnostic Interview – Revised ADOS Autism Diagnostic Observation Schedule AOSI Autism Observation Scale for Infants AOI Area of Interest ASD Autism Spectrum Disorder BAP Broader Autism Phenotype DS Difference Score DSM Diagnostic and Statistical Manual of Mental Disorders ESCS Early Social Communication Scales EASE Early Autism Sweden EEG Electroencephalogram HR High Risk IJA Initiation of Joint Attention LR Low Risk MSEL Mullen Scales of Early Learning PDD – NOS Pervasive and Developmental Disorder – Not Otherwise Spec-

ified RBS – R Repetitive Behavior Scale – Revised RJA Response to Joint Attention STAT Screening Tool for Autism in Two-year-olds TD Typically Developing/Typical Development ToM Theory of Mind VABS Vineland Adaptive Behavior Scale WISC Wechsler Intelligence Scale for Children

9

Introduction

In infancy and early childhood, interaction with parents and other caregivers plays an important role in a child’s development. In dyadic interaction, the infant learns and develops from direct communication and interplay with a parent. In triadic attention sharing, the infant and parent mutually direct their attention toward a common object or event. Such triadic attention sharing is typically referred to as joint attention and infants begin to engage in joint attention related behaviors as early as their first year of life (e.g. Beuker, Rommelse, Donders, & Buitelaar, 2013; Gredebäck, Fikke, & Melinder, 2010). Joint attention is considered an important foundation for learning, in-teraction, and development in several areas (e.g. Baldwin, 1991; Charman et al., 2000; Mundy, Sullivan, & Mastergeorge, 2009; Tomasello, Carpenter, Call, Behne, & Moll, 2005). For example, by following other people’s gaze, infants get guidance as to what elements in the environment are currently ben-eficial to attend to. They may also eventually learn that what others are looking at may be a cue to what they are thinking about. In addition, hearing others name the objects they are looking at provides opportunities to learn words (Baldwin, 1991).

Importantly, not only do infants take advantage of the learning opportunities that are initiated by others, they also actively create such opportunities by di-recting others’ attention. An infant may look back and forth between a parent and an interesting object, or, slightly later in development, point toward the object. The parent may respond by naming the object, or even by handing it to the infant who then gets to explore it and learn about its functions. The infant and parent may also get an opportunity to share each other’s affective states regarding the object. The fact that joint attention plays an important role in development also entails that if a child does not engage in it to the same extent as others, development may be negatively affected.

As a group, children with Autism Spectrum Disorder (ASD) differ from other children in terms of social interaction with others, and this includes engage-ment in joint attention. Although it is well-known that autistic children both initiate and respond to joint attention less than other children (e.g. Chiang, Soong, Lin, & Rogers, 2008; Dawson et al., 2004), many aspects of the rela-tion between joint attention and ASD warrant further investigation. Gaze fol-

10

lowing is an example of an area where some studies report impairment in chil-dren with ASD (e.g. Falck-Ytter, Fernell, Hedvall, von Hofsten, & Gillberg, 2012) whereas others find typical performance (e.g. Leekam, Hunnisett, & Moore, 1998). The circumstances under which children with ASD do and do not follow gaze therefore need to be further explored. There is also a need to learn more about the role that joint attention may play in the early develop-ment of children who are later diagnosed with ASD. Since diagnosis before the ages of 2-3 years is highly uncommon, focusing on infant siblings of chil-dren with ASD (who themselves have an increased probability of a later diag-nosis) has proven to be a successful means of studying early development (Zwaigenbaum et al., 2007). Over the past 10-15 years, the infant sibling par-adigm has increased our knowledge about the early development of children with ASD, but significant knowledge gaps still need to be filled.

The overarching aim of this thesis is to increase our knowledge about the re-lation between various joint attention behaviors and ASD, both before and after diagnosis. By including studies of infant siblings of children with ASD as well as a study of older children with a confirmed diagnosis, it is my hope that this thesis will contribute to our understanding of the role that joint atten-tion may play in the life and development of children with or at risk for ASD.

This thesis will start by a general overview of ASD. Among other things, some of what we know about causes, neurobiology and cognitive functioning will be accounted for. Thereafter, the concept of joint attention will be discussed in more detail, both in the context of typical development and ASD. The in-dividual studies will then be presented. Finally, I will end with providing an integrated discussion of the results and my thoughts on how the findings may contribute to the field.

Autism Spectrum Disorder A Brief Historical Note In 1943, a paper titled “Autistic disturbances of affective contact” (Kanner, 1943) was published. The author, child psychiatrist Leo Kanner, described 11 children who shared certain characteristics that markedly set them apart from other children. The most outstanding feature, according to Kanner, was the children’s “inability to relate themselves in the ordinary way to people and situations”. Kanner emphasized that these children had shown atypical devel-opment from the very beginning of their lives. As infants, they had not made anticipatory motor adjustments when being picked up. Some of the children had displayed delayed language development, and once they started speaking,

11

they seemed to use language in a different manner than most children; lan-guage was not primarily used for communication, but for repeating previously heard words and phrases, or for reciting impressive amounts of facts learnt by heart. Some of the children enjoyed jumping up and down or watching spin-ning objects. They were easily frightened by loud noises and upset by changes in routines. Only a year after Kanner’s seminal paper, pediatrician Hans Asperger (1944) published a report describing four highly intelligent boys who displayed difficulties with social interaction and engaged in specific ob-sessive interests.

Many years have passed since Kanner (1943) and Asperger (1944) first de-scribed what is now known as Autism Spectrum Disorder (ASD). Although many of their observations are still accurate, the concept of autism has gone through many changes over the years, and has suffered from multiple miscon-ceptions. Previously categorized as ‘childhood schizophrenia’, it was not until 1980 that ‘infantile autism’, as it was then termed, was listed as a separate disorder in the Diagnostic and Statistical Manual of Mental Disorders (DSM). Although Kanner himself postulated a biological origin for the phenomenon he described, a view of ASD as being the consequence of cold and unrespon-sive parenting, so-called “refrigerator mothers” (Bettelheim, 1967) gained much popularity in the sixties and seventies. In the late nineties, ASD became the target of a different controversy, as a study (Wakefield et al., 1998) claim-ing it to be caused by the measles-mumps-rubella vaccine received widespread attention. The results were later shown to have been fabricated, and the paper was withdrawn from the journal. As will be accounted for in the following sections, we today know that ASD is neither caused by parental styles nor vaccines. Rather, ASD is highly heritable, with certain environmental factors and gene by environment interactions playing a role as well (Lai, Lombardo, & Baron-Cohen, 2014).

Prevalence and Comorbidities The first study to investigate the prevalence of ASD reported that the condition affected less than .5% of the population (Lotter 1966). Since then, prevalence estimates have increased, and ASD is today typically reported to occur in about 1–1.5% of the population (e.g. Baird et al., 2006; Baron-Cohen et al., 2009b; Idring et al., 2012; Saemundsen, Magnusson, Georgsdottir, Egilsson, & Rafnsson, 2013; Wingate et al., 2014). However, prevalence estimates of up to 2.5% (Kim et al., 2011) have been reported. The pattern of increasing estimates is likely explained by changes in how ASD has been defined over time, as well as by an increased awareness and enhanced diagnostic methods (Elsabbagh et al., 2012a). Whether the higher reports also reflect an increase in environmental risk factors is unknown (Lai et al., 2014). ASD is 2-3 times more common in males than in females (Baird et al., 2006; Idring et al., 2012;

12

Kim et al., 2011; Saemundsen et al., 2013), which may in part be due to un-derdiagnosing of females, but also to biological female-specific protective ef-fects and/or male-specific risk factors (Baron-Cohen et al., 2011; Lai et al., 2014). About 45% of all individuals with ASD have intellectual disability (Idring et al., 2012; Saemundsen et al., 2013). Other psychiatric, medical and developmental comorbidities affect more than 70% of all individuals with ASD, with depression, anxiety disorders and ADHD being the most common co-occurring conditions (Lugnegard, Hallerback, & Gillberg, 2011; Mattila et al., 2010; Simonoff et al., 2008).

Defining and Assessing ASD According to the Diagnostic and Statistical Manual of Mental Disorders–Fifth Edition (DSM-5), ASD is characterized by persistent deficits in social com-munication and social interaction, as well as by the presence of restricted and repetitive patterns of behaviors, interests or activities. The symptoms must have been present in the early developmental period, they must cause clini-cally significant impairment in functioning, and they must not be better ex-plained by intellectual disability or global developmental delay. The recent transition from the Diagnostic and Statistical Manual of Mental Disorders–Fourth Edition–Text Revision (DSM-IV-TR) entailed substantial changes in terms of ASD criteria, with the most prominent being the merge of Autistic disorder, Asperger’s disorder, and Pervasive Developmental Disorder-Not Otherwise Specified (PDD-NOS) into one diagnostic category - Autism Spec-trum Disorder. The merge was grounded on the empirical finding that the clas-sification of subtypes varied substantially across clinical sites, thus rendering the categorical division non-meaningful (Lord et al., 2012b). The transition from the DSM-IV-TR also meant merging deficits in social interaction and social communication (previously two separate criteria) into one criterion; re-moving atypical language from the criteria; and introducing the impairment criterion. The transition from DSM-IV to DSM-5 may be viewed as represent-ing a move from a more categorical conception of ASD to a more dimensional view (Lord & Jones, 2012).

Research suggests that it is possible to reliably diagnose ASD already at the age of 2 years (Chawarska, Klin, Paul, & Volkmar, 2007; Lord, 1995; Stone et al., 1999), but diagnosis before the age of 3 is still relatively uncommon (Boyd, Odom, Humphreys, & Sam, 2010). A review of studies published be-tween 1990 and 2012 report that the mean age of diagnosis varies between 3 and 10 years, and reveals a trend toward earlier diagnosis (Daniels & Mandell, 2014). Factors that are associated with earlier diagnosis are higher symptom severity and parental concern, and lower language ability (Daniels & Mandell, 2014; Salomone, Charman, McConachie, & Warreyn, 2016).

13

The core components of an ASD assessment are an in-depth interview with the parent, focusing on early development and specific symptoms, as well as a behavioral observation of the child (Ozonoff, Goodlin-Jones, & Solomon, 2005). Both should preferably be conducted using standardized methods, and the instruments that are generally considered “gold standard” and recom-mended as first choice by several international clinical guidelines (e.g., NICE.org.uk) are the Autism Diagnostic Observation Schedule – second edi-tion (ADOS-2; Lord et al., 2012c) and the Autism Diagnostic Interview-re-vised (ADI-R; Lord, Rutter, & LeCouteur, 1994). Assessments of intellectual functioning as well as adaptive functioning should also be included, and pro-vide an important foundation for intervention (Ozonoff et al., 2005). The Wechsler Intelligence Scales (e.g. WISC-V; Wechsler, 2014) are the most widely used instruments of cognitive functioning in verbal children, and adap-tive functioning is typically assessed by the use of parental report measures such as the Vineland Adaptive Behavior Scales (VABS; Sparrow, Balla, & Cicchetti, 1984). Tests of related domains such as attention and executive functioning, as well as assessments of any suspected comorbidities, should be included if necessary. Genetic testing, particularly chromosomal microarray analysis, can be used to identify certain rare forms of ASD (Geschwind & State, 2015) but is not always performed. ASD assessments should be inter-disciplinary, and are ideally conducted by a team including psychologists, physicians, speech and language pathologists, and other professionals if needed (Ozonoff et al., 2005).

Outcomes and Intervention Having ASD is associated with several negative outcomes, such as increased risk of being bullied in childhood (Schroeder, Cappadocia, Bebko, Pepler, & Weiss, 2014); loneliness and lower levels of social support in adolescence (Lasgaard, Nielsen, Eriksen, & Goossens, 2010); and lower perceived quality of life (Barneveld, Swaab, Fagel, van Engeland, & de Sonneville, 2014) and higher levels of stress (Hirvikoski & Blomqvist, 2015) in adulthood. ASD, in both high- and low-functioning individuals, is associated with an increased risk of premature mortality (Hirvikoski et al., 2016). Causes of premature death are highly variable and include somatic diseases as well as suicide. Apart from the medical conditions associated with low-functioning ASD, lack of social support, insufficient coping skills and a lower tendency to seek med-ical attention may be factors underlying the increased mortality (Hirvikoski et al., 2016). Being a parent of a child with ASD is associated with higher levels of stress (Hayes & Watson, 2013), higher work absence, and lower income (McEvilly, Wicks, & Dalman, 2015).

ASD is generally considered a lifelong condition, but it has been suggested that a minority of autistic individuals may reach a stage where they no longer

14

fulfill diagnostic criteria (Fein et al., 2013). As of today, no pharmaceutical cure exists for the core ASD symptoms, although psychopharmacological agents are often used to treat comorbid symptoms (Mohiuddin & Ghaziuddin, 2013). Available behavioral interventions include early intensive behavioral intervention, social skills training, parent-mediated intervention, educational approaches toward parents and teachers, as well as targeted approaches aimed at specific areas of low functioning - for example, language, joint attention, emotion recognition, imitation, etc. Many behavioral intervention techniques share the overarching goals of promoting communication, independence, and quality of life (Lai et al., 2014). Most behavioral interventions have low to moderate effects (Lai et al., 2014). It is believed however, that earlier inter-vention may be associated with more positive outcome (Boyd et al., 2010), and specifically that offering intervention before the full autistic phenotype is in place may alter development and result in lower symptoms expression (Dawson, 2008). Indeed, a few attempts to engage infants at risk for ASD in early intervention have been made lately, with promising results (Green et al., 2017; Rogers et al., 2014). When discussing intervention, it is important to acknowledge the neurodiversity perspective, and to note that many high-func-tioning autistic adults consider their ASD an integral part of their personality, rather than a deficit in need of treatment (Jaarsma & Welin, 2012; Kapp, Gillespie-Lynch, Sherman, & Hutman, 2013; Ortega, 2009).

Causes A multitude of studies have shown that genes play an important role in the development of ASD. Twin studies have reported heritability estimates rang-ing from 38% to over 90% (e.g. Bailey et al., 1995; Colvert et al., 2015; Hallmayer et al., 2011; Lichtenstein, Carlstrom, Rastam, Gillberg, & Anckarsater, 2010). Up to 25% of all cases of ASD are estimated to be caused by rare monogenic conditions, such as Fragile X syndrome and tuberous scle-rosis, and rare copy number variants affecting genes involved in brain devel-opment (Geschwind & State, 2015). The majority of the remaining cases is most likely explained by a complex genetic model where a combination of specific common genetic variants, with small individual effect sizes, may un-derlie the autistic phenotype (Geschwind & State, 2015). De novo mutations, which are not inherited from the parents, but occur in the gamete or fertilized egg, are more common in simplex families (with only one affected individual) than in multiplex families (Iossifov et al., 2014; Levy et al., 2011; Sebat et al., 2007). No study has reported 100% concordance rates in monozygotic twins, entailing that environmental factors play a role as well. It has been consistently shown that non-shared environment, i.e., factors that differ between siblings, play a bigger part than shared environment (Ronald & Hoekstra, 2011). Envi-ronmental factors most likely act as triggers, with the autistic phenotype being the result of a gene by environment interaction. Known environmental risk

15

factors include advanced maternal (Sandin et al., 2012) and paternal age (Hultman, Sandin, Levine, Lichtenstein, & Reichenberg, 2011); maternal in-fection or medication use during pregnancy (Gardener, Spiegelman, & Buka, 2009); prenatal exposure to environmental toxins such as traffic-related air pollution (Volk, Lurmann, Penfold, Hertz-Picciotto, & McConnell, 2013) and pesticides (Roberts et al., 2007); as well as perinatal complications (Kolevzon, Gross, & Reichenberg, 2007; Willfors et al., 2017).

Neurobiology On the neurological level, a view of ASD as a disorder characterized by early overgrowth and altered connectivity has emerged over the years (Courchesne et al., 2007; Rane et al., 2015). Multiple studies have presented evidence of increased neuronal growth in the first years of life (Hazlett et al., 2011; Redcay & Courchesne, 2005). A recent study of infants at risk for ASD demonstrated that a pattern of increased cortical surface growth between 6 and 12 months, followed by an increased growth rate of the total brain volume between 12 and 24 months, was associated with a subsequent diagnosis of ASD (Hazlett et al., 2017). Evidence suggests that the increased growth rate slows down later in childhood (Redcay & Courchesne, 2005) and that it may even be followed by a period of decline in some individuals (Courchesne, Campbell, & Solso, 2011). These later stages may be the consequence of extensive pruning, par-ticularly in pre-frontal areas, which in turn may lead to the consistently re-ported pattern of decreased connectivity in the pre-frontal-cortex, as well as decreased long-range connectivity involving several brain areas (Rane et al., 2015). Increased short-distance connections, possibly compensatory, have been suggested to occur in some brain areas (Belmonte et al., 2004), and in-creased cortical-subcortical connectivity has also been reported (Di Martino et al., 2014). Some evidence of correlations between hypoconnectivity and autistic traits on a behavioral level have been presented (Gotts et al., 2012; von dem Hagen, Stoyanova, Baron-Cohen, & Calder, 2013). Whether the pat-terns of aberrant connectivity are widespread across the brain or whether they predominate in the so-called ‘social brain’ network – a set of neuroanatomical structures argued to underpin social perception and functioning (Brothers, 1990) – is currently discussed (Elsabbagh & Johnson, 2016; Gotts et al., 2012).

Cognitive Models Multiple models have tried to describe autistic behavior from a cognitive per-spective. Whereas some accounts place an emphasis on the social domain, others argue for the need to investigate more general differences across do-mains. The earliest social theories focused on differences in terms of social cognition. The Theory of Mind (ToM) account (Baron-Cohen, 1995; Baron-

16

Cohen, Leslie, & Frith, 1985), proposing a difficulty with metalizing (repre-senting the mental states of self and others) as a core deficiency in ASD, is perhaps the most well-known cognitive theory of ASD. From a social motiva-tion perspective on the other hand, it has been argued that differences in terms of social cognition may be second to an early diminished social interest. The general idea behind this line of reasoning is that infants who later receive an ASD diagnosis are less motivated by social stimuli than other infants, which in turn leads to fewer social learning opportunities, and thus impacts develop-ment in multiple areas (Chevallier, Kohls, Troiani, Brodkin, & Schultz, 2012; Dawson, Bernier, & Ring, 2012). Proponents of the non-social accounts argue that a focus on the social domain will fail to explain important non-social as-pects of ASD. Individuals with ASD have been shown to display a preference for local over global features, and to display superior performance on tasks where such a bias is advantageous (Frith, 1970; Shah & Frith, 1983). Several accounts have emphasized that enhanced low-level perceptual processing, and a tendency toward bottom-up rather than top-down control, can explain fea-tures such as hyper-sensitivity, savant skills, generalization difficulties, and attention to detail (Happe & Frith, 2006; Mottron et al., 2013; Mottron, Dawson, Soulieres, Hubert, & Burack, 2006; Van Boxtel & Lu, 2013).

An Increased Focus on Dimensionality and Heterogeneity Advances in the fields of both genetics and behavioral sciences have led to changes in how we conceptualize ASD. As mentioned earlier, a shift toward a dimensional perspective has occurred over the years. That is, rather than considering ASD a discrete entity, many agree that it should be viewed as representing the extreme end of a normal variation (e.g. Constantino & Todd, 2003). This shift is partially driven by the realization that autistic-like traits are present to various degrees in many more than those with an ASD diagno-sis. The term Broader Autism Phenotype (BAP) is commonly used to describe this type of sub-clinical traits and personality features in individuals who do not meet full diagnostic criteria. Such traits are more common in relatives of autistic individuals than in the population at large, and evidence suggest them to be etiologically linked to ASD (e.g. Constantino et al., 2006; Lundström et al., 2012; Sucksmith, Roth, & Hoekstra, 2011). Over the years, there has also come to be a greater emphasis on heterogeneity. ASD is characterized by a great degree of diversity in terms of interactional styles, language level, cog-nitive functioning, etc. This, in combination with the notion of divergent ge-netic pathways, has led many to prefer a view of ASD as representing not one but many disorders (e.g. Geschwind & Levitt, 2007; Happé, Ronald, & Plomin, 2006).

17

The Infant Sibling Paradigm and Early Development The fact that ASD is rarely diagnosed before the age of 2-3 years (Boyd et al., 2010) entails that knowledge about early signs and development has been scarce. Previous research had to rely on retrospective methods such as parental report or analyses of home videos - methods that suffer from limitations and are prone to bias. The last decades’ emerging view of ASD as a predominantly genetic disorder has however given rise to a novel means of studying early development by focusing on younger siblings of diagnosed children. In a fam-ily where at least one child has an ASD, the probability of a younger sibling receiving a diagnosis is 7-20% (Gronborg, Schendel, & Parner, 2013; Messinger et al., 2015; Ozonoff et al., 2011), and thus much higher than the general prevalence of ~1.0-1.5% (e.g. Baird et al., 2006; Wingate et al., 2014). Over the past 10-15 years, a large number of infant sibling studies have emerged. These studies typically include a so-called high risk (HR) group, made up of infants with at least one older sibling with an ASD, as well as a low risk (LR) group of control infants with at least one typically developing (TD) older sibling and no family history of ASD (Zwaigenbaum et al., 2007). The infants are followed regularly from early on and until they are old enough to go through diagnostic assessment, typically at 24 or 36 months. The study design allows for different types of analyses to be made at different stages of the research project. Cross-sectional comparisons between HR and LR infants can be made at any age, and may provide useful information concerning how infants with a family history of ASD differ from those without it. When the infants are old enough to go through diagnostic assessment, longitudinal com-parisons can be made between different outcome groups, including those who receive an ASD diagnosis. By including multiple measurement points, it is also possible to map early trajectories.

Studies of younger siblings of children with ASD have contributed substan-tially to our knowledge about early development. In a review of the field, Jones, Gliga, Bedford, Charman, and Johnson (2014) conclude that most symptoms of ASD become apparent between 12 and 18 months. These involve multiple domains, and include atypical eye contact (Zwaigenbaum et al., 2005); fewer gestures (Landa, Holman, & Garrett-Mayer, 2007; Zwaigenbaum et al., 2005); fewer directed vocalizations (Ozonoff et al., 2010); less social smiling (Ozonoff et al., 2010; Zwaigenbaum et al., 2005); lower response to name (Nadig et al., 2007; Zwaigenbaum et al., 2005); less imitation (Macari et al., 2012; Young et al., 2011); less engagement in joint attention (Landa, Gross, Stuart, & Faherty, 2013; Rozga et al., 2011); more non-functional repetitive play (Christensen et al., 2010); and more atypical exploratory behaviors (Ozonoff et al., 2008). At this age, infants with later ASD also score lower on tests of both motor development (Christensen et al.,

18

2010; Landa & Garrett-Mayer, 2006) and impressive as well as expressive language (Landa & Garrett-Mayer, 2006; Zwaigenbaum et al., 2005).

Apart from a few studies that have reported deviances during the first year - lower attention to faces at 6 months (Chawarska, Macari, & Shic, 2013) and a pattern of declining attention to eyes between 2 and 6 months (Jones & Klin, 2013) – the majority of the conducted studies indicate that 6-month-olds with later ASD do not differ from other infants in terms of social-communicative behaviors (Ozonoff et al., 2010; Rozga et al., 2011; Zwaigenbaum et al., 2005). Outside the social domain however, a few more early deviances have been detected during the first year, including slower attention disengagement (Elison et al., 2013); poor postural control (Flanagan, Landa, Bhat, & Bauman, 2012); increased perceptual sensitivity (Clifford, Hudry, Elsabbagh, Charman, & Johnson, 2013); lower activity level (Zwaigenbaum et al., 2005); and dif-ferences in terms of early consonant production (Paul, Fuerst, Ramsay, Chawarska, & Klin, 2011). Neuroimaging has also revealed early differences not detectable in behavior, such as aberrant neurological responses to direct gaze at 6-10 months (Elsabbagh et al., 2012b) and, as noted previously, in-creased cortical surface growth rates between 6 and 12 months (Hazlett et al., 2017). Finally, studies investigating developmental trajectories have sug-gested that different subgroups of children with later ASD may display differ-ent patterns of early development (Chawarska et al., 2014; Landa, Gross, Stuart, & Bauman, 2012; Macari et al., 2012), lending support to the view that ASD represents a heterogeneous condition.

Joint Attention The term joint attention refers to the sharing of attention between two individ-uals toward a common object or event (Bruner, 1975; Scaife & Bruner, 1975). Imagine two friends admiring a painting at an arts exhibition. While viewing the painting, they may occasionally avert their gaze from it and look at each other. They may also talk about their common focus of attention – one may ask the other about his opinion of the painting, or perhaps express her own thoughts. She might point toward a part of the painting that she finds particu-larly interesting, and her friend will immediately look in the direction of her pointing. They may then both laugh at the sight of the same clever detail. The ability to engage in joint attention thus provides a foundation that allows the two friends to both share their visual attention, and to share each other’s thoughts and feelings regarding the common focus of interest.

When joint attention is studied, researchers usually focus on overt behaviors such as gaze (and point) following, or the use of gaze and gestures to initiate a common focus of attention. Definitions of the concept vary however, and

19

assume different levels of mental awareness. As an example of a lean defini-tion, Butterworth (1998) suggests that joint attention may be defined “simply as looking where someone else is looking”. In contrast, Tomasello et al. (2005) argue that joint attention is not achieved until the child has reached a developmental stage where s/he is able to represent others’ attention as goal directed and affected by choices. Regardless of what level of mentalistic abil-ities one assumes, behaviors such as gaze following, alternating gaze (i.e., looking back and forth between another person and an event of interest), and pointing, are pivotal in that they serve the function of aligning the attention of individuals in time and space, which provides an important foundation for de-velopment and interaction. In this thesis, I adhere to the view that joint atten-tion, when it is fully developed, extends beyond the directly observable be-havior that individuals engaging in it perform (e.g., gaze following), and that it also involves a mental awareness of other’s attention. However, overt be-haviors such as alternating gaze and gaze following start to develop before the infant can be expected to have such an awareness. In this thesis, I will be re-ferring to early occurring behaviors such as gaze following and alternating gaze as joint attention behaviors, as they represent important components of the construct joint attention. The fact that, for example, gaze following may or may not involve an understanding of the other’s looking behavior as refer-ential will however be discussed in more detail later on.

Infants start to engage in joint attention behaviors as early as the first half of their first year of life (D'Entremont, 2000; D'Entremont, Hains, & Muir, 1997; Gredebäck et al., 2010; Perra & Gattis, 2010), and the ability to do so plays an important role for their development in several areas. Infants learn that fol-lowing another’s gaze is likely to result in an interesting sight, and that direct-ing another’s gaze may result in the other providing useful information about the common focus of attention. Both gaze following and initiation of joint attention play important roles for language development. Following the adult’s gaze enables the child to map a novel word, spoken by the adult, to the correct referent (Baldwin, 1991). Furthermore, adults respond to children’s joint attention bids by providing labels for objects attended to by the children (Tomasello & Farrar, 1986, but see Akhtar & Gernsbacher, 2007, for a critical comment on the importance of joint attention for word learning). Not surpris-ingly, joint attention in infancy/toddlerhood has been shown to correlate both concurrently and longitudinally with language ability (Beuker et al., 2013; Cochet & Byrne, 2016; Morales et al., 2000; Morales, Mundy, & Rojas, 1998; Mundy et al., 2007; Mundy & Gomes, 1998). But the role of joint attention in learning and development extends the language domain. Early joint attention behaviors have been shown to correlate with a later understanding of other’s mental states and intentions (Brooks & Meltzoff, 2015; Camaioni, Perucchini, Bellagamba, & Colonnesi, 2004; Charman et al., 2000), and has thus been argued to be a foundation for ToM (Sodian & Kristen-Antonow, 2015). Joint

20

attention skills have also been shown to contribute to cooperation success in early childhood (Wu, Pan, Su, & Gros-Louis, 2013), and to be associated with aspects of executive function, such as self-regulation (Morales, Mundy, Crowson, Neal, & Delgado, 2005; Van Hecke et al., 2012) and response inhi-bition in reward-based learning (Nichols, Fox, & Mundy, 2005). It has been shown that joint attention facilitates learning and is associated with a deeper level of processing in infancy (Kopp & Lindenberger, 2011) and toddlerhood (Hirotani, Stets, Striano, & Friederici, 2009). Joint attention has also been shown to affect information processing in adults (Bockler, Knoblich, & Sebanz, 2011; Frischen & Tipper, 2004), thus demonstrating that it does not only play an important role in early development, but continues to do so throughout life (e.g. Mundy et al., 2009). The ability to engage in joint atten-tion has even been suggested as an evolutionary foundation for human com-munication and culture (Tomasello et al., 2005).

When studying joint attention, researchers often differentiate between re-sponse to joint attention (RJA) and initiation of joint attention (IJA). Gaze following is typically measured as an indicator of RJA, and studies of IJA may include behaviors such as pointing, showing, and alternating gaze. Evidence suggests that RJA and IJA reflect partially different processes. They differ in terms of developmental trajectories, with RJA having an earlier onset than IJA (Beuker et al., 2013; Gredebäck et al., 2010); and longitudinal correlations within each type of joint attention are stronger than correlations between the two types at a given time point (Mundy et al., 2007). RJA and IJA have also been shown to be partially dissociated in terms of underlying neurological processes (Mundy et al., 2009; Redcay, Kleiner, & Saxe, 2012; Schilbach et al., 2010); and they have been shown to contribute differently to language de-velopment in childhood (Mundy & Jarrold, 2010); as well as to cognition and information processing in adulthood (Kim & Mundy, 2012).

Joint Attention and ASD Over the years, a multitude of studies with diverse methodologies have demonstrated that children with ASD show deficits in joint attention (e.g. Charman, 2003; Chiang et al., 2008; Dawson et al., 2004; Falck-Ytter et al., 2012; Leekam, López, & Moore, 2000; Maljaars, Noens, Jansen, Scholte, & van Berckelaer-Onnes, 2011; Mundy, Sigman, & Kasari, 1994; Mundy, Sigman, Ungerer, & Sherman, 1986; Naber et al., 2008; Sigman & Ruskin, 1999; Warreyn, Roeyers, & De Groote, 2005). Assessing joint attention is an important part of diagnostic assessment, and measures of both RJA and IJA are included in the ADOS-2 and the ADI-R, the gold standard instruments for diagnosing ASD. Studies of infant siblings have identified lower engagement in joint attention as one of the earliest signs of ASD, emerging as early as around the first birthday (Landa et al., 2007; Rozga et al., 2011). Considering

21

the important role that joint attention has in early development, less engage-ment in it can be expected to lead to missed learning opportunities and to have cascading effects on several developmental areas (e.g. Charman, 2003; Mundy et al., 2009).

The previously accounted for relation between joint attention and language is well established in ASD as well as in typical development, and in a recent meta-analysis, Bottema-Beutel (2016) concluded that this relation is in fact stronger in ASD than in TD. The author suggested that there may be a thresh-old of joint attention abilities above which language develops sufficiently, and that the stronger association in ASD may be due to more children in this group falling below that threshold. The meta-analysis also confirmed that although both RJA and IJA are related to language, the association is stronger for RJA. The association between joint attention and aspects of executive functioning has also been confirmed in ASD (Dawson et al., 2002), and it has been shown that joint attention impairments contribute to cooperation-related difficulties in children with ASD (Colombi et al., 2009). Finally, and as will be more thoroughly discussed later, some intriguing evidence has suggested that en-gaging in joint attention may not have the same beneficial effects on infor-mation processing in ASD as in typical development (Falck-Ytter, Thorup, & Bölte, 2015b; Mundy, Kim, McIntyre, Lerro, & Jarrold, 2016).

Although there is evidence of impairment both in terms of RJA and IJA in children with ASD, deficits in IJA have been shown to persist longer in devel-opment than RJA-related deficits (Gotham, Risi, Pickles, & Lord, 2007; Lord et al., 2000; Mundy et al., 1994; Sigman & Ruskin, 1999). However, most RJA studies focus on gaze following accuracy – that is, they measure whether or not the child responds to another person’s gaze cue by making a gaze shift from the person’s face to a target object. As will be discussed later, one of the papers in this thesis will explore the hypothesis that autistic children who do follow gaze may still differ from other children in terms of how they process the gazed at objects. It may thus be premature to conclude that older children with ASD are unimpaired in terms of RJA.

Considering the partial dissociation between the two aspects of joint attention (e.g. Mundy et al., 2009), the following sections will discuss our current knowledge of gaze following and initiation of joint attention separately. Find-ings regarding typical development will be presented first, before findings re-garding children with ASD and findings from studies of infant siblings will be accounted for. Each section will end with a paragraph in which I highlight what I think are the key findings and knowledge gaps that have led up to the studies in this thesis.

22

Gaze Following and Response to Joint Attention in Typical Development From birth on, TD infants show a visual preference for human faces (Johnson, Dziurawiec, Ellis, & Morton, 1991). They prefer faces with open eyes over faces with closed eyes (Batki, Baron-Cohen, Wheelwright, Connellan, & Ahluwalia, 2000), and faces with direct gaze over those with averted gaze (Farroni, Csibra, Simion, & Johnson, 2002). These early preferences are be-lieved to facilitate social-cognitive development by allowing the infant to fo-cus on social stimuli (Johnson, Senju, & Tomalski, 2015). Newborns have also been shown to display a rudimentary sensitivity to gaze direction; they are faster to allocate a visual target if the target location is cued by the gaze direc-tion of a schematic face (Farroni, Massaccesi, Pividori, & Johnson, 2004). This tendency is often referred to as gaze cueing, or reactive gaze following, and indicates a very early reliance on other people’s gaze cues as a means of processing the world. Fast and involuntary, gaze cueing differs from the more volitional, spontaneous gaze following that infants start to engage in later in development, and that will be the primary focus of this section.

Studies of spontaneous gaze following typically entail a person, live or on video, who initiates eye contact and then looks in the direction of one of sev-eral objects. The looking toward the target is almost always accompanied by a head turn, and sometimes with pointing gestures and/or vocalizations such as calling the child’s name. TD infants start to spontaneously follow the gaze of others around 3-4 months of age, and the ability becomes increasingly sta-ble during the subsequent months (D'Entremont, 2000; D'Entremont et al., 1997; Gredebäck et al., 2010; Perra & Gattis, 2010). Early gaze following is relatively rudimentary with young infants following the general direction of gaze and only being able to detect objects within their immediate visual field (Butterworth & Jarrett, 1991). The ability however gradually becomes more precise, and by 9 months infants are able to ignore a closer object in order to follow gaze to a more peripheral target (Flom, Deak, Phill, & Pick, 2004). During their second year of life, children even start to follow gaze to objects located behind them (Beuker et al., 2013; Deak, Flom, & Pick, 2000). The speed by which infants respond to others’ gaze shifts also becomes increas-ingly faster during the first and second year of life (Gredebäck et al., 2010; Van Hecke et al., 2012).

As noted above, early gaze following does not require any mentalistic abili-ties, but could be a consequence of basic learning processes. That is, the infant may learn through experience to expect an interesting sight in the location of another’s gaze direction, and thus becomes more likely to respond to gaze cues by looking in the same direction (Deak, Krasno, Triesch, Lewis, & Sepeta, 2014; Moore & Corkum, 1994; Triesch, Teuscher, Deak, & Carlson, 2006;

23

although, the above discussed cueing effect in newborns could be argued to indicate some kind of innate bias). Toward the end of the first year, gaze fol-lowing becomes more fine tuned to specific aspects of the stimulus, and oth-ers’ eyes in particular. For example, it has been shown that 9-month-olds are equally likely to follow the head movement of a person with closed as with open eyes. At 10 months however, infants are more likely to follow the gaze of a person with open rather than closed eyes, suggesting, perhaps, that they now appreciate the link between eyes and seeing (Brooks & Meltzoff, 2005, but see Moore & Corkum, 1998, suggesting that infants do not follow eye direction without an accompanying head movement until the age of 18 months). Studies using looking times (Senju, Csibra, & Johnson, 2008) and event-related potentials (Senju, Johnson, & Csibra, 2006) have provided some evidence suggesting that 9-month-olds can encode the relation between gaze direction and targets. It has also been shown that 8-month-olds look longer at an empty location when someone else has previously looked in that direction, indicating that they expect the other’s gaze to be directed at a target (Csibra & Volein, 2008). Further, 9-month-olds look longer at faces gazing in the direc-tion of a target compared to faces gazing to the opposite side of the target, even if the face is moving in the opposite direction. This suggests that infants at this age prioritize gaze direction over facial movement (Senju et al., 2008), whereas 4-5-month-olds have been shown to rely on movement over gaze direction (Farroni, Johnson, Brockbank, & Simion, 2000).

I would like to end this section by acknowledging the fact that although gaze following is by far the most common measure of RJA, it is not its only mani-festation. The majority of the research has focused on visual joint attention, but joint attention is possible, although probably understudied, in other mo-dalities as well (see Botero, 2016, for a suggestion of how joint attention through touch can be studied, and Bigelow, 2003, for a study of joint attention development in blind infants). Even in the visual mode, it has been demon-strated that parents and infants also obtain joint attention by following each other’s hand movements (Yu & Smith, 2013).

Gaze Following and Response to Joint Attention in ASD A majority of studies suggest the area of reflexive gaze cueing to be intact in children with ASD (e.g. Chawarska, Klin, & Volkmar, 2003; Kylliäinen & Hietanen, 2004; Okada, Sato, Murai, Kubota, & Toichi, 2004; Swettenham, Condie, Campbell, Milne, & Coleman, 2003), whereas a substantial minority of studies suggest otherwise (e.g. Goldberg et al., 2008; Johnson et al., 2005; Ristic et al., 2005). For a review of the area, see Nation & Penny, 2008. When it comes to spontaneous gaze following, a large number of studies have shown that children with ASD engage less in it than both TD and developmentally delayed children (e.g. Chawarska et al., 2003; Chiang et al., 2008; Dawson et

24

al., 2004; Falck-Ytter et al., 2012; Leekam, Baron-Cohen, Perrett, Milders, & Brown, 1997; Leekam et al., 2000; Mundy et al., 1994; Mundy et al., 1986; Sigman & Ruskin, 1999). In this area as well however, there are quite a few studies that have reported similar performance between children with ASD and other children (Akechi et al., 2011; Falck-Ytter et al., 2015b; Freeth, Chapman, Ropar, & Mitchell, 2010; Leekam et al., 1998). Some of the studies (Leekam et al., 2000; Mundy et al., 1994) showed that deficits were most ev-ident in autistic children with lower mental age, thus suggesting that the ability to follow gaze may be predominantly impaired early in development.

It is not clear what the mechanisms behind the often reported early lower gaze following accuracy of autistic children are. Some early studies (Baron-Cohen, 1989; Leekam et al., 1997) showed that children with ASD were equally good as control children at determining at which object a model was looking when specifically asked to do so. The authors argued that this ruled out difficulties with detecting eye gaze direction as a possible cause behind the lower sponta-neous gaze following of the autistic children. However, using different meth-ods, some more recent studies have reported evidence of impaired eye gaze direction detection in children and adults with ASD (d'Arc et al., 2017; Howard et al., 2000; Webster & Potter, 2008). It is important to note that none of these studies investigated children younger than 10 years, and that it is therefore not possible to determine whether decreased gaze direction detection is a cause or a consequence of less gaze following engagement (or other as-pects of having an ASD diagnosis over several years).

Apart from measuring gaze following accuracy, a few studies have included measures of looking times. Falck-Ytter et al. (2015b) found that 3-year-olds with ASD were equally likely as TD and developmentally delayed children to follow gaze, but that the control children displayed a longer first fixation to-ward the attended object as compared to the unattended object than did the autistic children. This was suggested to reflect that the control children show a processing bias for objects being the focus of another person’s attention, but that this may not be the case for children with ASD. That is, seeing someone else look at an object would not affect the status of the object for autistic chil-dren as it may for other children. Measuring total looking durations rather than first fixation durations, and using static pictures rather than videos, Freeth et al. (2010) made a similar finding when showing that following gaze was as-sociated with an increase in fixation time to the target object in a group of TD adolescents, but not in adolescents with high-functioning ASD. Relatedly, two studies (Akechi et al., 2011; Gliga et al., 2012) have demonstrated that chil-dren with or at risk for ASD spend less time looking at attended objects in a word learning context, despite following gaze to the same extent as other chil-dren. Interestingly, the ASD/risk groups in these studies also learned fewer words than the control children, suggesting that the reduced processing time

25

may be a disadvantage for learning. This is a hypothesis that warrants further investigation. Nevertheless, if the looking patterns of autistic children really entails reduced learning opportunities, it would be of utmost importance to investigate whether these patterns can be altered (Mundy, 2016).

Studies of Gaze Following and RJA in Infant Siblings Whether newborns with later ASD display the same preference for eyes and direct gaze as TD infants (Batki et al., 2000; Farroni et al., 2002) is currently not known, since this has not been investigated in at-risk populations. Few studies of infant siblings have included measurement points prior to 6 months. Jones and Klin (2013) however, measured infants’ looking times to eyes re-peatedly between 2 and 24 months. They discovered that infants who were later diagnosed with ASD displayed a decline in attention to eyes between 2 and 6 months that was not present in TD infants. Importantly though, the in-fants with later ASD did not look less at eyes than TD infants at the initial measurement. The results should be interpreted with some caution, consider-ing that the ASD sample was very small and included only boys. Nevertheless, if valid, the findings do not support the presence of an innate insensitivity to eye gaze. Rather, they seem to indicate that infants with later ASD become gradually less interested in eyes during their first half year of life. Using elec-troencephalogram (EEG), another study has demonstrated that 6-10 month-olds with later ASD do not differentiate in terms of their neurological re-sponses between faces with gaze moving toward vs. away from them, as TD infants do (Elsabbagh et al., 2012a). This finding may indicate that infants with later ASD are less sensitive to gaze direction.

A number of infant sibling studies have investigated spontaneous gaze follow-ing, thus providing more insight on very early development. Two studies have reported lower levels of gaze following accuracy in HR infants than in LR controls (Cassel et al., 2007; Ibanez, Grantz, & Messinger, 2013), but two others did not detect any group differences (Goldberg et al., 2005; Yirmiya et al., 2006). However, none of these studies reported on the children’s later di-agnostic outcome. Of the studies that have followed infants until they were old enough to go through diagnostic assessment, the majority report some as-sociation between early gaze following and later ASD symptomatology (Bedford et al., 2012; Brian et al., 2008; Landa et al., 2007; Rozga et al., 2011; Sullivan et al., 2007; Yoder, Stone, Walden, & Malesa, 2009) but a number of studies do not find such an association (Chawarska et al., 2014; Ibanez et al., 2013; Macari et al., 2012). Not all reported group differences are related to gaze following accuracy. Bedford et al. (2012) did not detect any group dif-ferences in terms of this metric, but made a finding similar to the previously discussed looking time findings - namely that 13-month-olds with later socio-communicative problems, including ASD, spent less time looking at the object

26

attended to by the model. The finding was suggested to reflect a lower sensi-tivity to the referential meaning of the gaze cue.

The conducted infant sibling studies differ greatly in terms of methodology. The study by Bedford et al. (2012), for example, used eye tracking in response to pre-recorded stimuli, whereas most studies used standardized instruments administered during interaction. The studies also differ in terms of what types of gaze cues that were used. Whereas some used a silent head/gaze turn, others included pointing and/or vocalization. In a study of 15-month-old infant sib-lings, Presmanes, Walden, Stone, and Yoder (2007) specifically addressed the issue of cue redundancy. They discovered that it was only at an intermediate level (in that case vocalization and head/gaze shift) that a group difference - lower accuracy in the HR group - arose. When more cues were used in com-bination, both groups were equally successful, and when fewer cues were used, none of the groups managed to follow gaze. A later follow-up study revealed that this early pattern of gaze following accuracy was predictive of later social impairment (Yoder et al., 2009). The results of another study of 14-month-olds (Sullivan et al., 2007) also suggested that infants at risk for ASD may benefit more from increased cue saliency. It is interesting to note that in all of the different cue variations that were used in the two studies, the model turned her entire head toward the target object. Considering that infants with later ASD have previously been shown to be less sensitive to eye infor-mation than other infants (Elsabbagh et al., 2012b), an interesting extension would be to remove the head turn and present the infants with a model who used only her eyes to gaze in the direction of the target.

Key Findings and Knowledge Gaps (I) As has been recently accounted for, several studies have indicated that typi-cally developing 9-10-month-olds appreciate the link between eyes and gazed-at objects (Brooks & Meltzoff, 2005; Senju et al., 2008; Senju et al., 2006) and that they thus understand that another person is looking at something when his/her eyes are focused in the direction of that object. Younger infants, on the other hand, have been shown to rely on motion over eye gaze (Farroni et al., 2000) and to even follow the head movement of a person with closed eyes (Brooks & Meltzoff, 2005). Infants at risk for ASD have been shown to be less sensitive to eye information compared to other infants (Elsabbagh et al., 2012b), which could potentially imply that these infants would be less respon-sive to eye gaze cues in a gaze following context. However, the extent to which infants at high and low risk for ASD rely on information from the eyes vs. the head when following gaze has not been studied previously. This issue is addressed in Study I of this thesis.

27

Another question that is currently without a definite answer is the extent to which older children with ASD are impaired in terms of gaze following. Some studies have reported lower gaze following accuracy in this group (e.g. Falck-Ytter et al., 2012), whereas others have not found any group differences (e.g. Leekam et al., 1998). A number of recent studies have indicated that applying other metrics than gaze following accuracy may be useful, and focusing on looking time measures has proved to be an alternative means of investigating how children with ASD may differ from others. In particular, in our previous study described above (Falck-Ytter et al., 2015b), we found that children with ASD, despite following gaze, did not differentiate between attended and un-attended objects in terms of first fixation durations to the same extent as other children. This was interpreted as reflecting that object processing may be less influenced by other people’s looking behavior in autistic children compared to TD children, which could affect learning opportunities negatively. How-ever, as this was a small sample study, replication of the initial finding is needed. Moreover, from a clinical point of view it would be valuable to know if the pattern of less differentiating between attended and unattended objects in ASD could be altered through the use of more interesting objects. To ad-dress these issues was the aim of Study II.

Initiation of Joint Attention in Typical Development Around 8-9 months of age, infants begin alternating their gaze between an interaction partner and an object or event (Beuker et al., 2013). Between 8 and 10 months, the alternating of gaze is increasingly paired with anticipatory smiling; suggesting an attempt at communicating positive affect about the ob-ject gazed upon (Parlade et al., 2009; Venezia, Messinger, Thorp, & Mundy, 2004). Around 10-14 months, infants also start engaging in gestural IJA be-haviors such as pointing and showing (Beuker et al., 2013; Carpenter, Nagell, & Tomasello, 1998). In contrast to gaze following, research suggests that the early development of IJA may not be linear, but rather characterized by an early increase followed by a u-shaped pattern of decline and rebound between 12 and 18 months (Mundy et al., 2007).

IJA behaviors vary in terms of both function and form. With regard to func-tion, they can be either declarative – aimed at attention sharing - or imperative – aimed at obtaining a desired object or activity (Bates, Camaioni, & Volterra, 1975). A child pointing at an airplane in the sky in order to show it to a parent is an example of declarative IJA, whereas a child pointing to a cookie jar be-cause she wants a cookie is an imperative behavior. The two behaviors may share the same topography, and they may both comprise joint attention, but their function is significantly different, with attention sharing being a goal of the declarative behavior, but rather a means of the imperative behavior. As accounted for in the previous section, the form of IJA behaviors spans from

28

earlier occurring, more subtle behaviors such as alternating gaze and child-initiated eye contact, to later occurring gestural behaviors, such as pointing and showing. Some (e.g. Stone, Coonrod, & Ousley, 2000; Wetherby & Prizant, 2002) include vocal behavior as well.

As should be evident from the text above, studies of IJA measure a variety of different behaviors. One of the papers in this thesis will focus on one particular IJA behavior – namely alternating gaze. This behavior may serve the function of declarative joint attention; by looking back and forth between an object and a person, the child aims to draw the person’s attention to the object with the goal of sharing an experience. Alternatively, the child may look from an object to an adult in order to obtain social information about the object; the child will then use the adult’s reaction to regulate his or her own behavior toward the object (i.e., approach or withdrawal). This is referred to as social referencing and is something that infants encountering new or ambiguous objects start to engage in toward the end of the first year (Sorce, Emde, Campos, & Klinnert, 1985; Stenberg & Hagekull, 1997; Striano & Rochat, 2000). As a measure of IJA, alternating gaze has the advantage of occurring early in development, and it has been shown to display high individual stability between 9 and 18 months (Mundy et al., 2007). Nevertheless, it evidently represents a less explicit form of IJA than, for example, pointing, which has led some to question its suita-bility as an IJA measure (Pickard & Ingersoll, 2015; Tomasello, Moore, & Dunham, 1995).

IJA in typical development is not as well studied as gaze following, and many studies on the area come from the ASD field. Focusing on a typical population though, Van Hecke et al. (2007) showed that higher levels of IJA at 12 months is associated with more optimal social functioning at 30 months, thus demon-strating that the relation between IJA and social behavior is not restricted to ASD. Considering that IJA reflects a more active attempt at sharing attention than gaze following, it has been suggested to be more related to social moti-vation and reward. Indeed, imaging of neurotypical adults have confirmed that IJA is associated with increased activity in reward-related brain areas, such as the ventral striatum (Gordon, Eilbott, Feldman, Pelphrey, & Vander Wyk, 2013; Schilbach et al., 2010).

Initiation of Joint Attention in ASD A number of studies have shown that children with ASD do not initiate joint attention to the same extent as TD children and children with developmental delay (e.g. Chiang et al., 2008; Dawson et al., 2004; Goldberg et al., 2005; Mundy et al., 1994; Mundy et al., 1986; Sigman & Ruskin, 1999; Warreyn et al., 2005). In comparison to gaze following, problems with IJA seem to persist longer in development (Mundy et al., 1994; Sigman & Ruskin, 1999). It has

29

been shown that autistic children are relatively more impaired at declarative joint attention initiations than at imperative joint attention bids (Curcio, 1978; Mundy et al., 1986), suggesting that these children may direct another’s atten-tion as a means of obtaining a desired object, but that they are less likely to initiate joint attention for the sake of attention sharing. Children with ASD are also less likely than other children to combine IJA with the expression of pos-itive affect, suggesting that they are less motivated to share affective states with others (Kasari, Sigman, Mundy, & Yirmiya, 1990). In terms of the dif-ferent types of IJA behaviors, it has been suggested that the performance of autistic children differs more from that of typically developing children in terms of gestures than in terms of eye contact behaviors such as alternating gaze (Chiang et al., 2008; Mundy et al., 1994). Also, and somewhat surpris-ingly, a study of 3-year-olds with ASD failed to find a correlation between eye contact-based and gestural IJA (Pickard & Ingersoll, 2015).

Studies of IJA in Infant Siblings Infant sibling studies have investigated IJA behaviors early in development, and found them to be less frequent in HR than LR infants (Gangi, Messinger, Martin, & Cuccaro, 2016; Goldberg et al., 2005). Other interesting results in-clude the finding that the presence of certain genotypes associated with less optimal dopaminergic function is related to lower levels of IJA in an HR, but not an LR sample (Gangi et al., 2016); and the finding that HR infants engage in fewer IJA bids accompanied with social smiling than LR infants (Gangi, Ibanez, & Messinger, 2014). Whereas the studies mentioned so forth used combined measures of IJA, a few other studies have investigated eye contact-based IJA separately. Two studies with outcome data reported lower perfor-mance at 12–14 months in those infants who later received an ASD diagnosis (Landa et al., 2007; Macari et al., 2012), whereas a third study did not detect such a group difference (Rozga et al., 2011). In addition, one study without outcome data (Cassel et al., 2007) reported lower levels of eye contact IJA in HR than LR infants at 15 months (but not at any other time point between 8 and 18 months), whereas another study (Yirmiya et al., 2006) did not find such a difference at 14 months. Targeting alternating gaze in the function of social referencing, Cornew, Dobkins, Akshoomoff, McCleery, and Carver (2012) showed that 18-month-olds with later ASD were both less likely and slower to engage in this behavior.

To the best of my knowledge, only one study has prospectively investigated IJA before the first birthday, and its relation to later diagnostic outcome. Ibanez et al. (2013) showed that the frequency by which infants engaged in IJA at 8 months could predict diagnostic outcome at 30 months. This study also demonstrated that IJA development in infants with a later ASD diagnosis

30

show the same u-shaped pattern as detected in a typical sample (Mundy et al., 2007).

Key Findings and Knowledge Gaps (II) Although alternating gaze can be viewed as a rather subtle behavior, the fre-quency by which an infant engages in it may play an important role for devel-opment. By looking back and forth between a parent and a ball sitting on a shelf, the infant is able to communicate his interest in the ball to the parent. The response this evokes may involve both opportunities for social interaction (the parent acknowledging the infant’s behavior by attending to him), word learning (the parent commenting on the ball and naming it) and motoric ex-ploration (the parent handing the ball to the infant). Despite this powerful role that alternating gaze may play in terms of infant-initiated learning opportuni-ties, little is known regarding how infants at risk for ASD may differ from other infants in terms of this behavior. As recently mentioned, only one study (Ibanez et al., 2013) has prospectively investigated the relation between IJA before the first birthday and later ASD symptoms. This study did not target alternating gaze specifically, but measured IJA in a broader sense. Also, pre-vious studies have not carefully ruled out potential confounding factors that could explain the observed relation between IJA in infancy and later out-comes. In particular, it is important to study general attentional factors such as visual disengagement as well as general reductions in engagement with the social environment. Infants later diagnosed with ASD have previously been shown to be slower to disengage their attention from one object to another, compared to TD infants (Elison et al., 2013; Elsabbagh et al., 2013; Zwaigenbaum et al., 2005). They have also been shown to spend less time looking at social stimuli (Chawarska et al., 2013), thus reflecting a potentially lower social interest. Both slower disengagement and a lower social interest could be expected to affect the frequency by which infants engage in alternat-ing gaze. Indeed, a study of 3-year-olds with ASD (Schietecatte, Roeyers, & Warreyn, 2012) reported a concurrent relation between disengagement laten-cies and alternating gaze, and a trend toward a relation between social interest and alternating gaze. As this study did not include a control group, it is un-known whether these relations are specific to ASD or not. No previous study has investigated how differences in terms of attentional measures may influ-ence the longitudinal relation between alternating gaze and ASD symptoms, and doing so was one of the aims of Study II.

Aims of the Thesis The overarching aim of the studies in this thesis was to study various joint attention behaviors and their role in the development of children with or at

31

risk for ASD. Study I and Study II focused on early development, thus inves-tigating behaviors of interest as early as at 10 months in a sample of infants at high vs. low risk for ASD. Previous research has shown that typically devel-oping 10-month-olds rely on information from the eyes when following gaze (Brooks & Meltzoff, 2005), whereas another study has indicated that infants who later receive an ASD diagnosis may be less sensitive to eye information compared to TD infants (Elsabbagh et al., 2012b). The main goal of Study I was therefore to investigate whether infant siblings of children with ASD would differ from other infants in terms on their reliance on eye vs. head in-formation in a gaze following context. In Study II, the focus was on alternating gaze as a measure of initiation of joint attention. First, we wanted to examine whether the frequency by which 10-month-olds alternated gaze between an interaction partner and an interesting event would be related to autistic symp-toms at 18 months. Considering that it has been questioned to what degree alternating gaze really captures IJA (Pickard & Ingersoll, 2015), we also wanted to explore the validity of the measurement by investigating its relation with later occurring gestural IJA behaviors. Finally, Study II also aimed to investigate whether other attentional behaviors – disengagement latencies and social interest – may influence the potential relation between alternating gaze and ASD symptoms. Joint attention behaviors are typically measured either using naturalistic measures of interaction, or pre-recorded eye tracking stim-uli. A common aim of Study I and Study II was to combine the benefits of both approaches by using eye tracking during live interaction.

In Study III, the focus was shifted from early development and infants at risk to older children with a confirmed ASD diagnosis. The main goal of this study was to investigate gaze following from a different angle, and to further explore the hypothesis that children with ASD may differ from TD children in terms of how they process objects they see others look at. The first aim was to rep-licate the finding that autistic individuals differentiate less between attended and unattended objects in terms of first fixation durations (Falck-Ytter et al., 2015b) in a new sample. Since a lower processing bias for attended objects may have negative consequences - for example reduced learning (Akechi et al., 2011; Gliga et al., 2012) - the second aim was to test whether it would be possible to alter the pattern of less differentiation between attended and unat-tended objects by varying the interest level of the target objects.

32

Methods

Participants Study I and Study II All participants in Study I and Study II were part of a longitudinal project (Early Autism Sweden; EASE), following infant siblings of children with ASD and control infants from the age of 5 months, with multiple assessments up to age 6 years. For a general overview of the main assessments and measures in the study, see www.eurosibs.eu/research, (but note that the table only includes measures that are shared within the EUROSIBS network and not site-specific measures). Infants in the HR group were recruited through advertisements, the project’s website (www.smasyskon.se) and clinical units. Infants in this group were required to have at least one older sibling with di-agnosed ASD. LR infants were recruited from a database of families who had expressed interest in participating in research. These infants were required to have at least one older sibling with typical development and no first-hand rel-atives with known or suspected ASD. Exclusion criteria for both groups were pre-term birth (< 36 weeks) and confirmed or suspected medical problems. Both the HR and the LR group comprised primarily of middle-class families from the greater Stockholm, Sweden, area, and the socio-economic status did not differ between groups. The samples of Study I and II were largely over-lapping. The HR and LR samples did not differ in terms of either age or de-velopmental level, assessed by the Mullen Scales of Early Learning (MSEL; Mullen, 1995) at any visit. For participant characteristics, see Table 1 and Ta-ble 2.

Table 1. Participant characteristics by group, Study I, final sample. M/SD.

Measure HR Group N = 47 (21 boys)

LR Group N = 17 (11 boys)

Age (months) 10.25/0.45 10.27/0.58

MSEL Total Score 98.45/13.96 96.35/11.70

SES1 -0.06/0.85 0.14/0.84 1Socioeconomic status calculated on the basis of parental education and income (equal weigh-ing), expressed as a z-score (for this measure N in the HR group = 45, as two families did not disclose this information).

33

Table 2. Participant characteristics by group, Study II, final sample. M/SD.

Measure HR group N = 51 (22 boys)

LR group N = 16 (10 boys)

Assessment at 10 months of age 10.45/0.43 10.37/0.55

Assessment at 18 months of age 18.51/0.76 18.51/0.94

MSEL Total Score 10 months

99.92/13.50 101.44/11.64

MSEL Total Score 18 months 97.29/15.67 97.75/13.54

SES1 -0.02/0.84 0.08/0.84 1Socioeconomic status calculated on the basis of parental education and income (equal weigh-ing), expressed as a z-score (for this measure, N in the HR group = 50, as two families did not disclose this information).

Study III The participants in the ASD group in Study III were recruited from the Autism Center for Young Children in Stockholm, Sweden. All children in this group had a confirmed diagnosis of ASD, including Autistic disorder, Asperger’s disorder, and PDD-NOS. Diagnosis was corroborated using the ADOS-2. The children of the control-group were age- and gender-matched to the ASD group and recruited from the same database as the LR children in Study I and Study II. These children were typically developing according to parental report and did not have any medical or developmental diagnoses. As in Study I and II, most families were middle-class families from the greater Stockholm, Swe-den, area. The two groups did not differ statistically in terms of age or non-verbal IQ (NVIQ), but there was a trend toward higher NVIQ in the TD group. For participant characteristics, see Table 3.

Table 3. Participant characteristics, Study III, final sample. M/SD.

Measure ASD N = 16 (12 boys)

TD N = 17 (12 boys)

Age (months) 81.56/15.42 74.12/15.51

NVIQ1 105.23/19.20 115.76/11.01

ADOS-2 Total Score 14.31/5.77 N/A 1NVIQ was assessed using the standard non-verbal subtests of WPPSI-III or WISC-IV (based on 13 children in the ASD group, as it was not possible to determine NVIQ in three children).

34

Apparatus and Procedures Study I and Study II Study I and Study II were conducted using a live eye tracking procedure (e.g. Nyström, Bölte, & Falck-Ytter, in press), where the gaze behavior of the in-fants was recorded during interaction using a Tobii TX300 eye tracker. The experiments in Study I and II were part of the same eye tracking session, which comprised a number of tasks and lasted approximately 10-15 minutes in total. The infant was seated on the lap of the parent, 200 cm from a scene area, where the experimenter was seated at a low table. At each side of the table was an oblong transparent lamp (that was only activated during the Study II experiment). The eye tracker was placed at a table between the scene area and the infant, and recorded the gaze of the infant with a sample rate of 120 Hz. Two video cameras recorded the child’s behavior and the scene area. The ses-sion started with a calibration procedure in which the experimenter moved a squeaky toy across pre-defined calibration points to attract the attention of the infant. Calibration was validated via online inspection of gaze replay on a monitor in the background of the room, and repeated if necessary.

For the experiment of Study I, which measured gaze following, two wooden screens were placed on top of the table, one on each side of the experimenter. Each screen had a hole at approximately the same height as the experimenter’s face. The experiment included four trials, during which the experimenter made pairs of puppets appear through the holes. Each trial started by the experi-menter calling the child’s name in order to elicit eye contact. In the Eyes and Head condition, the experimenter then turned his/her head toward one of the puppets while saying “Oj!” (a Swedish interjection expressing surprise or ex-citement). The experimenter kept looking at the puppet for 4 seconds before turning back to look at the infant. The Eyes Only condition mimicked the Eyes and Head condition, except for the fact that the experimenter kept his/her head still in a forward position, moving only the eyes in the direction of one of the puppets. Each block contained four trials, two in each condition, and the order of the conditions was counterbalanced between blocks. For a sketch of the experimental setting, see Figure 1.

35

Figure 1. Experimental setting, Study I. Infant and parent were seated 200 cm from the experimenter. A Tobii TX300 eye tracker, placed on the table in front of the in-fant, recorded the infant’s gaze. Two cameras, not visible in the sketch, recorded the infant’s behavior and the stimulus area.

The Study II experiment, that measured alternating gaze, took place at the end of the live eye tracking session, and the wooden screens were then removed from the table. The experiment included four trials. The experimenter started by attracting the infant’s attention. S/he then turned on the lamp on his/her right side, by the use of a remote control hidden under the table. Once the lamp was turned on, lights began to flash, changing color approximately every sec-ond. The lights were flashing for 10 seconds, during which the experimenter remained still facing the infant, intermittently speaking softly, and thus providing the infant with an opportunity to initiate joint attention. If the infant made an explicit attempt at directing the experimenter’s attention toward the lights, for example, by pointing or saying “lamp” or “there”, the experimenter responded by turning toward the lights and commenting on them. The ra-tionale behind this was to avoid extinguishing joint attention behaviors by be-ing unresponsive. The four trials were identical, except for the fact that the lights on the experimenter’s left side were activated on the first and third trials, and the lights on the right side were activated on the second and fourth trials. See Figure 2.

36

Figure 2. Sketch of experimental setting, IJA-task, Study II.

Study II also included a disengagement task that took place in the middle of the eye tracking session. During this task, that was a modified version of the disengagement task from the Autism Observation Scale for Infants (AOSI; Bryson, Zwaigenbaum, McDermott, Rombough, & Brian, 2008), the experi-menter presented the infant with two rattles. The experimenter started by shak-ing one rattle at approximately 20 cm to the right of his/her face, while keeping the second rattle hidden behind the table. Once the infant fixated on the active rattle, the experimenter began shaking the second rattle on the left side of his/her face. Six disengagement trials were presented, alternating between the left and right sides.

The live eye tracking experiment was performed by six individual experiment-ers (2 males, 4 females), who had all been trained according to a written pro-tocol and a video template.

Study II included data from a follow-up visit when the children were 18 months old, during which they were assessed with the ADOS. The Toddler Module (ADOS-T; Lord, Luyster, Gotham, & Guthrie, 2012a), developed for children below 30 months of age, was used. The ADOS is a semi-structured play-observation, designed to elicit certain behaviors of interest. The admin-istration results in a total measure, reflecting the degree of autistic symptoms, as well as individual scorings of the frequency and quality of various behav-iors. The ADOS-T was administered by three experienced clinical psycholo-gists (2 males, 1 female).

37

Study III Study III was conducted using traditional eye tracking with pre-recorded video stimuli. The gaze behavior of the children was recorded with a Tobii T120 eye tracker and the stimuli were displayed on a 17-inch monitor. After a five-point calibration was conducted, 16 video clips, lasting 10 seconds each, were pre-sented intermixed with other stimuli and attention grabbers, and divided into three blocks with breaks in between. Each video showed a female model seated at a table, on which four objects were placed. See Figure 3. In the be-ginning of the video, the face of the model was covered by an attention grabber and a voice greeted the child by saying “hello”. The attention grabber was then removed, and the model turned her head (and eyes) toward one of the objects. The model kept looking at the object for 5 seconds, before turning her head back to a forward position. The videos comprised two conditions with eight videos in each. The high interest condition included two sets of objects - one with toy vehicles and one with model trains. Vehicles and trains were chosen because they are documented common interest areas in ASD, and have previ-ously been used in studies measuring attention to objects related to circum-scribed interests (Sasson, Elison, Turner-Brown, Dichter, & Bodfish, 2011; Sasson & Touchstone, 2013; Sasson, Turner-Brown, Holtzclaw, Lam, & Bodfish, 2008). The baseline condition also comprised two sets of objects - one with house plants with green leafs and one with flowering house plants. The objects in the two conditions were roughly equivalent in terms of size and perceptual detail. Each of the four objects comprising a stimulus set was at-tended to once by the model and the order of the videos was pseudorandom-ized so that no more than two videos from the same condition occurred in a row.

The parents of the children in Study III filled out the Repetitive Behavior Scale-Revised (RBS-R), and the five-factor solution (Lam & Aman, 2007) was used to assess the presence of repetitive behaviors and restricted interests.

38

Figure 3. Screenshots of the stimulus video, Study III, showing the attention grabber covering the model’s face, the model engaging in direct gaze, the model attending to one of the objects, and then engaging in direct gaze again. Areas of interest (AOIs) are highlighted. The upper row depicts one of two stimuli sets in the high interest condition; the lower row depicts one of two stimuli sets in the baseline condition.

Measures and Analyses Study I The recordings from the eye tracking were manually coded by a lab member who was blind to infant risk status as well as to the study hypotheses. Coding was conducted with the Tobii Studio 3.2.3. software, and based on gaze re-plays (video of the stimulus area with infants’ gaze superimposed). A time window was defined that started with the experimenter engaging in direct gaze just before s/he initiated the gaze shift toward the puppet, and ended with the experimenter shifting gaze from the puppet back to the infant. Only trials where the infant first fixated on the experimenter’s face and then one of the puppets within this time window were included. Trials where the infant looked at the attended object after fixating on the experimenter were coded as con-gruent. Trials where the infant looked at the unattended object after fixating on the experimenter were coded as incongruent. Trials where the infant fixated on the experimenter, but did not look at any of the puppets were coded as other, and were not included in the primary analysis, but compared across groups. Fixations had to be at least 200 milliseconds long to be included, and each infant had to contribute at least two valid trials per condition. As the primary measure, we computed a difference score (DS); the number of incon-gruent gaze shifts was subtracted from the number of congruent gaze shifts made by each infant. A positive DS would thus indicate that the infant made more congruent than incongruent gaze shifts. Due to the DS’s being negatively skewed, non-parametric tests were used. In order to test the hypothesis that the two groups were differently affected by the experimental manipulation, we subtracted the performance (DS) in the Eyes Only condition from that in the Eyes and Head condition. The obtained measure reflects performance re-duction in the Eyes Only condition relative to the Eyes and Head condition,

39

and comparing this measure between groups is analogous to testing for a group by condition interaction effect in a 2 x 2 ANOVA. Statistical analyses were performed in SPSS (SPSS Inc., Chicago, IL) and two-tailed probabilities were used.

Study II Data text files were exported from the Tobii Studio Software and analyzed in MATLAB (MathWorks Inc., Natick, MA), using the TimeStudio project framework (version 3.15; timestudioproject.com; Nyström, Falck-Ytter, & Gredebäck, 2016), and SPSS. For each recording of the IJA task, gaze data was extracted from four individually defined AOIs, covering the face of the experimenter, the two lamps, and the entire scene area. As manual inspections revealed that the lights sometimes were flashing for slightly shorter than 10 seconds, only the first 9 seconds were analyzed for all trials.

Trials with less than 50% gaze data during the time the lights were flashing were excluded from the analysis. We further required at least two valid trials from each infant. As the primary measure of alternating gaze, we used the number of gaze alternations between the face of the experimenter and the flashing lights. In order to assess for general differences in terms of looking patterns and durations, we also compared the total number of fixations on the scene area and the total looking times in the different AOIs across groups.

For the disengagement task, data was extracted from two AOIs covering the rattles. As the dependent measure, we used the time it took the infant’s gaze to enter the AOI covering the second rattle once it became visible to the infant. Each infant had to contribute at least two valid disengagement trials in order to be included in the analysis, and all trials where the infants made predictive gaze shifts (disengaged from the first rattle before the second was presented) were excluded.

In order to assess for differences in terms of attention to the experimenter, a social preference quotient was calculated during all parts of the live eye track-ing session except the IJA task and the disengagement task. The session in-cludes a variety of components during which the experimenter interacts with the infant. For example, the experimenter presents and plays with different toys and makes small talk. The present objects are never near the experi-menter’s face, which allowed us to calculate a ratio of relative looking time toward the face. The total amount of time that the infants spent looking at the experimenter’s face was divided by the total amount of time they spent look-ing anywhere in the scene area, including the experimenter and various ob-

40

jects. Higher values would thus indicate a higher preference for the experi-menter over the non-social elements. In order to be included in this analysis, the infants had to contribute at least 25% valid data.

As a general measure of ASD symptomatology, the Total Social Affect and Restricted Repetitive Behavior Score from the ADOS-T was used. As recom-mended by Esler et al. (2015), the calibrated severity score of this measure was used. Calibrated severity scores range from 1–10, with higher values in-dicating more symptoms and a higher degree of concern. To investigate the relation between early alternating gaze and later gestural IJA, we used the ADOS-T items showing and pointing. Both are scored on a four-point scale where lower values indicate higher incidence and quality of the behaviors. Since all three ADOS-T variables were positively skewed, non-parametric tests were used for the correlations and group comparisons involving these variables.

Study III Data was analyzed using Tobii Studio 3.2.3. software. A fixation filter (Tobii Fixation Filter) with a velocity threshold of 35 pixels/window and a distance threshold of 35 pixels was applied, and fixations shorter than 60 milliseconds were discarded. Seven rectangular AOIs were defined. One covered the atten-tion grabber in the beginning of the video, and the other six covered the model’s face, the four objects and the entire screen. First fixation durations within each of the six later AOIs were extracted from when the model’s face was revealed until the end of the clip. Gaze shifts were coded manually, using gaze replays (recordings of the stimuli with the participant’s gaze and AOIs superimposed). For a trial to be included in the analysis, the child had to first fixate on the model’s face, and then on one of the objects. Trials where the gaze moved from the face to the attended object were coded as congruent, and trials where the gaze moved from the face to one of the two objects at the unattended side were coded as incongruent. Trials where the gaze moved from the face to the unattended object at the same side of the table as the attended object were removed from the analysis, since visual inspection had revealed that the children relatively often looked toward the objects immediately after seeing the model start her gaze shift, relying on the general direction of her gaze rather than paying attention to which specific object she would look at. It was therefore not possible to assure whether such responses reflected “ac-curate” gaze following based on the general direction of gaze or an “inaccu-rate” response. The groups did not differ in terms of the number of such trials.

Three dependent measures were defined. 1) Accuracy was calculated as a pro-portion where the number of congruent gaze shifts was divided by the total number of (congruent and incongruent) gaze shifts. An accuracy score of .33

41

would indicate chance performance (due to the above described removal of trials where the child looked at the object next to the attended), and we ex-pected the performance of each group to exceed this value in both conditions. 2) For the analysis of first fixation durations, we first calculated the mean first fixation durations at attended and unattended objects in each condition. The averages were then used to calculate a proportional measure where the first fixation duration at attended objects was the numerator and the sum of the first fixation durations at attended and unattended objects was the denominator. We termed this measure the Attended–Unattended Fixation Index (AUF-in-dex). Values above .5 would indicate longer first fixations at attended than unattended objects, whereas values below .5 would indicate longer first fixa-tions at unattended objects. Only data from trials with congruent gaze shifts was included in this analysis. 3) Total looking time at objects was defined as the total duration of fixations on objects divided by the total duration of fixa-tions on the screen, thus resulting in a measure indicating the proportion of time that was spent looking at objects. All trials, regardless of whether the children followed gaze or not, were included in this analysis, as were looks at attended as well as unattended objects. Looking time is a measure known to indicate interest (Holmqvist et al., 2011), and the purpose of this measure was thus to assess object type preferences. We expected the autistic children to spend more time looking at the objects in the high interest condition as com-pared to the baseline condition, but did not have any specific hypotheses re-garding this measure in the TD group. Each child had to contribute at least two valid trials per condition in the accuracy analysis, and at least one valid trial per condition in the first fixation duration analysis in order to be included. Statistical analyses were performed in SPSS. All variables were normally dis-tributed, and parametric tests were therefore used. Two-tailed probabilities were used throughout.

42

Study I

Background Typically developing infants start to appreciate the critical role of the other’s eyes in a gaze following context around 9-10 months of age (Brooks & Meltzoff, 2005; Senju et al., 2008). Some evidence suggests that infants with later ASD may be less sensitive to eye information than other infants (Elsabbagh et al., 2012b). Gaze following studies typically measure responses to a model who turns her head toward the object she is looking at, and by the use of such a paradigm, infants with later ASD have been shown to follow gaze to the same extent as TD infants (Bedford et al., 2012). However, this design entails that we cannot know whether the infants rely predominantly on information from the eyes or from the head movement. The aim of the study was therefore to investigate whether this would differ between infants at high vs. low familial risk for ASD. We hypothesized that excluding the head move-ment that typically accompanies the eye gaze shift would affect gaze follow-ing accuracy more in the HR group than in the LR group.

Primary Results Preliminary analyses confirmed that the groups did not differ in terms of the amount of data produced in either condition; the number of trials coded as other; the duration of trials; or the time that had to be allocated to engaging the infants in eye contact (all p-values > .25). There was no indication that individual differences between experimenters affected the results, and no ef-fects of infant gender (all p-values > .25).

In line with the hypothesis, the performance reduction when omitting head information was significantly larger in the HR group than in the LR group (HR: M = 2.02, SD = 2.68; LR: M = 0.47, SD = 1.66), U = 240.00, p = .01, r = −0.31 (Mann–Whitney U test; Figure 4). Bonferroni-corrected follow-up tests revealed that the HR infants were less likely to follow gaze in the Eyes Only as compared to the Eyes and Head condition, p < .001, r = −0.65, whereas the performance of the LR group did not differ between conditions, p > .25, r = −0.25 (related samples Wilcoxon signed rank tests). The accuracy difference scores did not differ between groups in either the Eyes and Head

43

condition, U = 316.50, p > .25, r = −0.16, or the Eyes Only condition, U = 307.00, p >.25, r = −0.18.

The gaze following performance of both groups was significantly above chance level (i.e., an accuracy difference score of zero) in both the Eyes and Head condition (HR p < .001; LR p = 0.003) and the Eyes Only condition (HR p = 0.008; LR p = .005; one-sample Wilcoxon signed rank tests).

Figure 4. The gaze following accuracy of the HR group was significantly higher in the Eyes and Head condition than in the Eyes Only condition, whereas the perfor-mance of the LR group did not differ between conditions. Error bars represent stand-ard errors.

Conclusions The experiment showed that both HR and LR infants were able to follow gaze above chance level both in response to a model who used only the eyes as a gaze cue, and to a model who used eyes and head cues in combination. How-ever, omitting the head movement led to a significant decline in gaze follow-ing in the HR group, but not in the LR group. This suggests that the former group relies more on the head movement when following gaze. At least two possible explanations may account for this finding. First, HR infants may be less sensitive to eye information than LR infants. This would be in line with a previous ERP finding indicating a lower sensitivity to gaze direction in infants

44

with later ASD (Elsabbagh et al., 2012b). Alternatively, the pattern of results may not be specific to eye information, but may reflect that the HR infants generally need more salient cues in order to follow gaze optimally. That is, their attention may be better captured by the large head turns compared to the subtler eye gaze shifts. This interpretation is in line with a previous finding showing that HR infants perform worse than LR infants at an intermediate level of cue redundancy, but not when excessive cues are used in combination (Presmanes et al., 2007). Regardless of which explanation is correct, the find-ings of the current study highlight the importance of separating information from the eyes from information from the head when studying gaze following and ASD.

45

Study II

Background A number of studies have demonstrated that there is a relation between less initiation of joint attention early in life and later ASD (Ibanez et al., 2013; Landa et al., 2007; Rozga et al., 2011; Yoder et al., 2009). The studies included a variety of behaviors, ranging from eye contact to explicit gestures. Only two (Landa et al., 2007; Macari et al., 2012) focused on alternating gaze specifi-cally and the only study that was conducted before the first birthday (Ibanez et al., 2013) measured IJA in a broader sense. None of the prospective studies controlled for the potential influence of other attentional measures on which infants with later ASD have been shown to differ from other infants, such as attention disengagement (Elison et al., 2013; Elsabbagh et al., 2013; Zwaigenbaum et al., 2005)  and general social interest (Chawarska et al., 2013).

The fact that alternating gaze emerges earlier in development than gestural IJA suggests that the former is a better target for early detection. However, some uncertainty exists regarding the validity and usefulness of alternating gaze as a measure of IJA in relation to ASD. On the one hand, alternating gaze has been shown to display a high level of individual consistency between 9 and 18 months, thus suggesting that it can be reliably measured early in life (Mundy et al., 2007). On the other hand, a study of autistic preschoolers found eye-contact based and gestural IJA to be unrelated to each other, and that only the gestural behaviors were associated with other joint attention-relevant met-rics such as gaze following and language (Pickard & Ingersoll, 2015). Further, some studies have indicated that it may be predominantly in terms of the ges-tural IJA behaviors that the performance of autistic children differs from that of TD children (Chiang et al., 2008; Mundy et al., 1994).

Against this background, the main aims of Study II were to 1) investigate whether a measure of alternating gaze at 10 months could be used to discrim-inate between infants based on risk status as well as to predict later symptom level; and 2) to investigate the validity of the alternating gaze measure by fo-cusing on its relation with later gestural IJA behaviors. In addition, we wanted to explore the effects of some aspects of looking behavior that have previously

46

been suggested to account for some of the variation in IJA – namely attention disengagement and social preference (Schietecatte et al., 2012).

Primary Results Alternating Gaze, Disengagement and Social Preference at 10 Months At 10 months, the LR group made more gaze alternations between the exper-imenter and the lights than the HR group, t(65) = 2.66, p = .012, r = .31 (in-dependent samples t-test). See Figure 5. The performance on the disengage-ment task did not differ between groups, t(55) = 0.62, p > .25, r = .08. For descriptive statistics for this and other metrics, see Table 4. A significant neg-ative correlation between disengagement and alternating gaze was discovered in the LR group, r(10) = - .77, p = .004, suggesting that longer latencies to disengage were associated with less alternating gaze. In the HR group, a mar-ginally significant trend in the same direction was detected, r(43) = - .28, p = .062. The groups did not differ in terms of social preference, t(63) = 0.69, p > .25, r =.09; and social preference and alternating gaze did not correlate in ei-ther the HR group, r(48) = .23, p = .109, or the LR group, r(13) = -.23, p > .25.

Table 4. Descriptive statistics by group for 10-month attentional measures and 18-month ADOS-T scores. M/SD.

Measure HR group LR group

Disengagement latency (s) 10 months1 0.74/0.25 0.79/0.20

Social Preference 10 months2,3 0.44/0.12 0.47/0.10

ADOS-T Total score4 18 months

3.18/1.74 2.06/1.53

ADOS-T Showing 18 months

0.92/0.91 0.44/0.81

ADOS-T Pointing 18 months

0.71/0.81 0.38/0.50

1 Based on 45 HR and 12 LR infants; 2 Based on 50 HR and 15 LR infants; 3 Looking time at model divided by total looking time at stage area; 4Calibrated severity score

47

Figure 5. Number of gaze alternations between the face of the experimenter and the flashing lights. Error bars represent standard errors. * p < .05.

Longitudinal Relations Between 10-month Alternating Gaze and 18-month Symptom Level Significant negative correlations between alternating gaze at 10 months and total ADOS-T score at 18 months were found in both the HR group, rs(49) = -.28, p = .048, and the LR group, rs(14) = -.56, p =.024. See figure 6. Making fewer gaze alternations at 10 months was thus associated with more ASD symptoms at 18 months. The correlational analyses suggest that this relation was similar in both groups, which indicates that the group factor may not con-tribute explained variance in the outcome variables when alternating gaze is included in the model. To address this statistically, we analyzed the data using a multiple regression analysis with ADOS-T score as the dependent variable. First, in order to test whether the relation between alternating gaze and the total ADOS-T score was different in the two groups, an unrestricted model (separate beta weights allowed for group, alternating gaze and group*alternat-ing gaze interaction) was compared with a restricted model where the beta weight for the interaction term was set to zero (i.e., no interaction term al-lowed). There was no indication that the unrestricted model fit the data signif-icantly better than the restricted model, F(1,64) = 0.05, p > .25; and the re-stricted model was thus retained in subsequent analyses. This model was sig-nificant, R2 = 0.16, F(2, 64) = 6.29, p = .003, showing that only alternating gaze was a significant predictor of symptom level, β = -.31, t = -2.61, p = .011 (group status: β = .18, t = 1.46, p = .15). With no other factors included in the model, alternating gaze explained 14% of the variance in ADOS-T scores, p

48

= .002. Due to the skewness of the ADOS-T variable, the procedure was re-peated using a log-transformed version of the same variable. All patterns of significant results remained the same.

No correlations between disengagement at 10 months and total ADOS-T score (HR group: rs(43) = -.10, p > .25 LR group: rs(10) = .32, p > .25) or social preference at 10 months and total ADOS-T score (HR group: rs(48) = .18, p = .202; LR group: rs(13) = .14, p > .25) were found.

Figure 6. Alternating gaze plotted against total ADOS-T scores. Figure shows indi-vidual data points and separate regression lines for each group.

Longitudinal Relations Between 10-month Alternating Gaze and 18-month Gestural IJA Making fewer gaze alternations at 10 months was associated with less show-ing at 18 months, as measured by ADOS-T in both the HR group, rs(49) = -.33, p = .018, and the LR group, rs(14) = -.50, p = .047. A trend toward a negative correlation between alternating gaze and the ADOS-T item pointing was found in the HR-group, rs(49) = -.25, p = .082, indicating that fewer gaze alternations at 10 months was associated with less pointing at 18 months. In the LR group, the relation was similar in magnitude and direction, but did not reach statistical significance, rs(14) = - 0.32, p = .223.

49

Conclusions The study showed that LR infants made more gaze alternations between an interaction partner and an interesting event (flashing lights) than HR infants at 10 months. In addition, making fewer gaze alternations at 10 months was associated with more ASD symptoms at 18 months, and this relation was sim-ilar in both groups. Importantly, we were also able to show that infants who used more alternating gaze early in life tended to engage more in pointing and showing in toddlerhood. This strengthens the validity of the alternating gaze measure by demonstrating that it is related to later occurring gestural IJA be-haviors. By showing that the relation between alternating gaze in infancy and later ASD symptomatology was unrelated to differences in terms of disen-gagement latencies and general social preference, the study provides an im-portant addition to existing work in the area (Ibanez et al., 2013; Landa et al., 2007; Rozga et al., 2011; Yoder et al., 2009).

50

Study III

Background Although some studies have demonstrated lower degrees of gaze following in children with ASD than in TD and developmentally delayed children (e.g. Chawarska et al., 2003; Falck-Ytter et al., 2012; Leekam et al., 1997), other studies have reported comparable performance between groups (e.g. Akechi et al., 2011; Freeth et al., 2010; Leekam et al., 1998). This shows that the relation between RJA and ASD is complex, and not necessarily characterized by lower gaze following accuracy. A number of recent studies have suggested that children with or at risk for ASD who do follow gaze may still display subtle but important deviances in comparison to other children. One study (Bedford et al., 2012) revealed that infants with later ASD spent less total time looking at the object attended to by the model, and two other studies (Falck-Ytter et al., 2015b; Swanson & Siller, 2013) reported differences in terms of the micro-structure of the looking behavior of autistic children. Specifically, Falck-Ytter et al. (2015b) discovered that TD and developmentally delayed 3-year-olds differentiated more between attended and unattended objects in terms of the length of the first fixation than autistic children of the same age. This was interpreted as reflecting a weaker processing bias for attended ob-jects in the ASD group. The first aim of Study III was to replicate the pattern discovered by Falck-Ytter et al. (2015b) in an older and more high functioning sample. Our hypothesis was that even high-functioning 6-year-olds, who are expected to follow gaze accurately, would display a lower processing bias for attended objects as compared to TD children.

The second aim of the study was to investigate whether it would be possible to alter the pattern of less differentiating between attended and unattended ob-jects in ASD by manipulating the interest level of the target objects. Circum-scribed interests are present in 75-95% of all children with high-functioning ASD (Klin, Danovitch, Merz, & Volkmar, 2007; Turner-Brown, Lam, Holtzclaw, Dichter, & Bodfish, 2011). They are intense interest areas that usu-ally represent a great source of joy for the individual, but may also be very time-consuming and interfere with other areas of life (Mercier, Mottron, & Belleville, 2000). It has previously been shown that autistic children show more visual attention toward objects associated with circumscribed interests (Sasson et al., 2011; Sasson et al., 2008), and that such objects can be used to

51

increase social interaction (Baker, 2000; Boyd, Conroy, Mancil, Nakao, & Alter, 2007) and joint attention (Kryzak, Bauer, Jones, & Sturmey, 2013; Kryzak & Jones, 2014). We therefore wanted to investigate whether using ob-jects that are commonly related to circumscribed interests as gaze targets would lead to an increased processing bias for attended objects in ASD.

Primary Results No group differences were found in terms of the number of valid trials in either condition or analysis (all p-values ≥ .092). No general differences in first fix-ation durations between groups or object types were found, and no effect of NVIQ was detected (all p-values > .25).

Gaze Following Accuracy As expected, both groups followed gaze above chance level in both the high interest condition (ASD: t(15) = 8.24, p < .001; TD: t(16) = 6.05, p < .001) and the baseline condition (ASD: t(15) = 6.08, p < .001; TD: t(16) = 5.11, p < .001; one sample t-tests comparing the means to .33). For descriptive statistics for this and other measures, see Table 5. A repeated measures ANOVA on the accuracy scores revealed no main effect of condition, F(1, 31) = 0.10, p > .25, partial η2 = .003, or group, F(1, 31) = 2.49, p = .125, partial η2 = .07, and no interaction effect between group and condition, F(1, 31) = 0.02, p > .25, partial η2 = .001.

First Fixation Durations A repeated measures ANOVA with condition as within subjects variable and group as between subjects variable revealed a group by condition interaction effect, F(1, 31) = 11.31, p = .002, partial η2 = .27, but no main effect of con-dition, F(1, 31) = 0.13, p > .25, partial η2 = .004, or group, F(1, 31) = 0.86, p > .25, partial η2 = .03. See Figure 7. In order to understand the interaction effect, a series of Bonferroni-corrected follow-up tests were performed. A higher AUF-index in the TD as compared to ASD group was found in the baseline condition, t(31) = 2.95, p = .024, d = 1.06, whereas no group differ-ence was found in the high interest condition, t(31) = 1.47, p > .25, d = 0.53 (independent samples t-tests). Furthermore, no difference between conditions was found in the ASD group, t(15) = 1.65, p > .25, d = 0.85, but in the TD group a significantly higher AUF-index was found in the baseline as compared to the high interest condition, t(16) = 4.06, p = .004, d = 2.03 (paired-samples t-tests).

52

Table 5. Descriptive statistics by group for accuracy measures, first fixation dura-tion raw scores and RBS-R scores. M/SD.

Measure type Specific measure ASD TD

Accuracy1 high inter-est condition

0.75/0.20 0.64/0.21

Accuracy1 baseline condition

0.73/0.26 0.63/0.24

First fixation dura-tion raw scores (s)

Attended object high interest condition

0.38/0.27 0.36/0.31

Unattended object high interest condition

0.32/0.13 0.41/0.15

Attended object baseline condition

0.30/0.13 0.54/0.39

Unattended object baseline condition

0.44/0.30 0.39/0.23

RBS-R Restricted Interest score

2.85/2.152 0.35/0.61

1N congruent gaze shifts divided by total N (congruent and incongruent) gaze shifts; 2Based on 13 children

Total Looking Time at Objects A repeated measures ANOVA with total looking time as the dependent meas-ure, condition as within subjects factor and group as between subjects factor revealed a main effect of condition, F(1, 31) = 11.77 p = .002, partial η2 = .28, with more looking time at high interest compared to baseline objects. No main effect of group, F(1, 31) = 0.64, p > .25, partial η2 = .02, and no interaction effect between group and condition, F(1, 31) = 0.38, p > .25, partial η2 = .01, was found. See Figure 8. The analysis thus showed that the children, regard-less of group status, looked more at the objects in the high interest condition than in the baseline condition.

Parental Ratings of Circumscribed Interests The ASD group received a higher score than the TD group on the Restricted Interests subscale of the RBS-R, t(28) = 4.56, p = .001, confirming the ex-pected higher rate of circumscribed interests in the former group.

53

Figure 7. AUF-index (first fixation duration at attended object divided by sum of first fixation durations at attended and unattended objects, averaged across trials) for the two groups across conditions. Error bars represent standard errors. **p < .01.

Figure 8. Total looking time at objects divided by total looking time at screen for the two groups across conditions. Error bars represent standard errors.

54

Conclusions In line with our hypothesis, the groups did not differ in terms of gaze following accuracy, and they both followed gaze above chance level. Children of both groups spent more total time looking at trains and vehicles than at plants, sug-gesting that our manipulation of the interest level was successful. As expected, the pattern of results from the previous study (Falck-Ytter et. al., 2015b) was replicated in the baseline, but not in the high interest condition. That is, when ordinary objects were used as gaze targets, children with ASD differentiated less between attended and unattended objects than control children. This may reflect that another’s gaze changes the status of an object for TD children, but not for children with ASD. The second aim of the study was to explore whether including highly interesting objects as gaze targets would result in increased differentiation between attended and unattended objects in the au-tistic children. This was not the case; no difference between conditions was found in the ASD group. Surprisingly, a difference between conditions was instead found in the TD group, with more differentiating between attended and unattended objects in the baseline than in the high interest condition. This finding may suggest that in typical development, another’s gaze may influence object processing more when the objects themselves are less interesting.

55

General Discussion

The overarching aim of this thesis was to investigate the relation between joint attention and ASD, both before and after diagnosis. Study I and Study II fo-cused on early development, and at an age (10 months) where the full autistic phenotype is not yet observable. By investigating gaze following - an im-portant component of visual RJA - and alternating gaze - an early means by which infants direct others’ attention - the aim was to assess both initiation of and response to joint attention in infants at high familial risk for ASD. In Study III, the focus was shifted from early development to a time point later in child-hood (~ 6 years). Whereas impairment in terms of IJA are known to persist at this age, several studies have indicated that RJA problems tend to become less pronounced with increasing age and verbal ability (Gotham et al., 2007; Leekam et al., 1998; Mundy et al., 1994). Study III aimed to clarify what role RJA may play in the life and development of older children with ASD by moving beyond gaze following accuracy, and focusing on the subsequent pro-cessing of the target objects.

In the following sections, I will first discuss some theoretical implications of the findings regarding joint attention and ASD or ASD risk. This discussion is organized in two parts: early and later development. After that, some find-ings with relevance for our understanding of joint attention in typical devel-opment will be accounted for. The following sections will then highlight some methodological as well as clinical implications, before limitations and future directions are discussed.

Early Development Gaze Following In Study I, we showed that omitting the head movement that typically accom-panies the eye gaze shift when a person looks at an object had a greater impact on the gaze following accuracy of HR than LR infants. Since the distinction between eye- and head movement is typically not made in gaze following studies, this is an important contribution to the field. When previous studies have concluded that children with ASD or infants at risk for ASD follow gaze (e.g. Bedford et al., 2012; Falck-Ytter et al., 2015b), it is possible that these

56

children have followed the model’s head movement, rather than relying on her eye direction. We have offered two potential explanations for the pattern of results in Study I. First, it is possible that the lower performance in the Eyes Only condition was due to an ASD-specific lower sensitivity to eye gaze. Sec-ond, the lower performance in the Eyes Only condition may reflect that the HR infants generally need more salient cues than LR infants in order to follow gaze optimally. That is, whereas the subtle eye gaze cue was enough to evoke optimal gaze following in the LR group, this was not the case for all infants in the HR group, where the addition of the more salient head cue thus resulted in improved performance. Although it is beyond the scope of Study I to answer the question of which explanation applies, both have some empirical support in previous research. In favor of the hypothesis of a lower sensitivity to eye information, Elsabbagh et al. (2012b) have shown that infants with later ASD differentiate less between faces with gaze shifting toward vs. away from them than TD infants. Diagnosed individuals are known to display difficulties in-terpreting information from the eyes (Baron-Cohen, Jolliffe, Mortimore, & Robertson, 1997; Baron-Cohen, Wheelwright, & Jolliffe, 1997) and to look less at eyes in general (Dalton et al., 2005; Klin, Jones, Schultz, Volkmar, & Cohen, 2002; Spezio, Adolphs, Hurley, & Piven, 2007). In favor of the second hypothesis (a generally higher need for more salient directional cues), Presmanes et al. (2007) have demonstrated that HR infants display lower gaze following accuracy than LR infants in response to a model who vocalizes and turns her head, but that the group difference disappears when more cues (e.g., pointing) are added. The results of the study by Presmanes et al. (2007) and Study I are not directly comparable, since the studies differ in terms of the type of cue combinations that were used. However, the common finding is that both studies suggest that increasing cue saliency has a greater effect on HR than LR infants, and this has also been indicated in a study by Sullivan et al. (2007). Finally, a third possible explanation for the difference in perfor-mance between the two conditions in the HR group should perhaps be consid-ered. Younger infants have been shown to rely on movement over gaze direc-tion (Farroni et al., 2000). It is thus possible that the infants in the HR group would experience a developmental delay that entails that movement is still a more powerful cue for this group. However, considering that the two groups did not differ from each other in terms of general developmental level, such a higher sensitivity to motion would have to represent a rather specific delay. Also, this explanation does not contradict any of the other two, since a high reliance on motion early in life is likely to be replaced by an increased sensi-tivity to other types of cues.

Alternating Gaze The first aim of Study II was to test whether 10-month-olds’ tendency to al-ternate their gaze between an interaction partner and an unusual event could

57

be used to differentiate between infants based on risk status, as well as to pre-dict later ASD symptomatology. The answer to both of these questions proved to be in the affirmative. The infants in the HR group made fewer gaze alterna-tions between the experimenter and the flashing lights than the infants in the LR group. Moreover, less gaze alternation at 10 months was associated with more ASD symptoms at 18 months. This supports the hypothesis that the measure may have potential for predicting later emerging ASD symptoms (Ibanez et al., 2013). The second aim was to investigate the relation between early alternating gaze and later gestural IJA. We were able to show that alter-nating gaze at 10 months was significantly associated with showing at 18 months, and marginally associated with pointing at 18 months. This demon-strates that alternating gaze and gestural IJA are indeed related, and strength-ens the validity of the alternating gaze measure. The fact that only three infants pointed during our experiment also confirms that alternating gaze has the ad-vantage of being measurable earlier than gestural IJA.

Finally, taking advantage of eye tracking technology, Study II aimed to inves-tigate the role of disengagement ability and general social preference on alter-nating gaze, both concurrently and in relation to later ASD symptoms. Disen-gagement latencies have previously been shown to contribute variance to IJA in autistic children (Schietecatte et al., 2012) and, not surprisingly, we also found that longer disengagement latencies were associated with less alternat-ing gaze in both groups (albeit just marginally in the HR group). Latency to disengage was however not associated with either group status or ASD symp-tomatology in this study, suggesting that the relation between IJA and disen-gagement is unrelated to ASD. It is notable that previous studies have reported atypically long disengagement latencies in children with later ASD (Elison et al., 2013; Elsabbagh et al., 2013; Zwaigenbaum et al., 2005). None of these studies included measurements at 10 months, and Elsabbagh et al. (2013) found a group difference at 14 months, but not 7, suggesting that disengage-ment difficulties may emerge later. Considering that Elison et al. (2013) how-ever did report an effect at 7 months, the subject of disengagement latencies in early development clearly needs further investigation. In none of the infant groups in Study II was relative looking time at the experimenter related to alternating gaze, suggesting that the results could not be explained by a differ-ence in general preference for social stimuli. Combined with the main results, this finding suggests that infants with later ASD do not differ from TD infants in terms of more passive social orienting, but that they do not actively seek to share experiences with others to the same extent as other infants.

General Conclusions from Study I and Study II One of the common goals of infant sibling studies is to shed light on the order of emergence of various behavioral atypicalities, which will contribute to our

58

understanding of which features of ASD are primary, and which may repre-sent secondary effects. As mentioned in the introduction, the majority of the studies that have been conducted with infant siblings suggest that the social development of those infants who later receive an ASD diagnosis is relatively typical during the first year of life, but that deviances become apparent during the second year (Jones et al., 2014). The fact that we were able to detect group differences as early as 10 months, both in terms of gaze following and alter-nating gaze, suggests that joint attention behaviors may represent one of the earliest areas of social deviance. Study II assessed alternating gaze only 1-2 months after the behavior emerges at a group level in TD children (Beuker et al., 2013), and the results therefore could indicate that infants with later autis-tic symptoms may be less likely to initiate joint attention from the beginning. IJA has previously been related to motivation and reward (Gordon et al., 2013; Schilbach, 2010), and our findings may reflect that HR infants find sharing experiences with others less rewarding. Alternatively, the findings may indi-cate that HR infants are less likely to use social referencing, that is to look at others in order to obtain information about an object or event. Considering that infants and children learn many skills and behaviors in the context of in-teraction, a lower motivation to share experiences and a lower tendency to use social referencing could both be expected to have extensive consequences on development and learning. In conclusion, the two studies show that infants at risk for ASD differ from other infants both in terms of their responses to other people’s attentional cues, and in terms of their own tendency to direct others’ attention. These findings have in common that they are both likely to result in limited social learning opportunities.

Later Development Gaze Following Accuracy Study III aimed to measure gaze following and RJA in older children with and without an ASD diagnosis. The children in the ASD group were of average intelligence and characterized as high-functioning. As expected, this study did not reveal any group differences in terms of gaze following accuracy, and both groups of children were able to follow gaze above chance level in both condi-tions. The results thus add to a growing body of evidence suggesting that older children with ASD are able to follow gaze to the same extent as other children (Akechi et al., 2011; Falck-Ytter et al., 2015b; e.g. Leekam et al., 1998; Mundy et al., 1994). This is however not the same as to say that autistic chil-dren do not differ from TD children in terms of RJA. In Study III we wanted to move beyond gaze following accuracy, and to further explore the possibility that individuals with ASD display differences related to RJA that are better

59

captured by other measures (e.g. Falck-Ytter et al., 2015b; Swanson & Siller, 2013).

Looking Time Measures, Object Processing, and Possible Consequences for Learning In a previous study (Falck-Ytter et al., 2015b), we discovered that children with ASD did not differentiate as much as other children between attended and unattended objects in terms of their first fixation durations at the objects. In Study III, we replicated this finding in a condition where ordinary objects (plants) were used as gaze targets. We argue that the combined results of Study III and Falck-Ytter et al. (2015b) suggest that whereas seeing another person look at an object gives that object a special status for TD children, this may not be the case for children with ASD. The fact that the autistic groups dis-played intact gaze following accuracy, but nevertheless differed from TD chil-dren in the subsequent processing of the target objects, opens up the possibility that gaze following in ASD may be more reflexive rather than entailing a full appreciation of the communicative nature of the gaze cue. This hypothesis fits rather nicely with what has been concluded by some studies of gaze following and word learning. Akechi et al. (2011) demonstrated that both autistic and TD children followed gaze above chance level in a word learning context, but that only the TD group spent more time looking at the object attended to by the model than at the unattended object. The TD group also learned more words than the group with ASD, thus suggesting a link between looking time and word learning. Similarly, Gliga et al. (2012) found intact gaze following but reduced word learning and looking time to attended objects in a group of 3-year-olds at familial risk for ASD. These studies suggest that accurate gaze following may not be enough for word learning to take place in children with or at risk for ASD. The reduced looking time to the attended objects may in-dicate a failure to understand the referential or communicative meaning of the gaze cue, which in turn may result in lower word learning. Also in line with this notion, intact gaze following but reduced looking time at attended objects has been demonstrated both in HR infants (Bedford et al., 2012) and adoles-cents with ASD (Freeth et al., 2010). Although these studies used total looking durations, whereas Study III and Falck-Ytter et al. (2015b) focused on first fixation durations, the results of all studies can be interpreted as being in line with the hypothesis that seeing someone look at something does not affect children with or at risk for ASD in the same manner as it affects TD children. It is possible that although autistic children follow gaze, they may not mentally share the others’ attention to the same extent as other children. A lower degree of attention sharing during gaze following may have consequences beyond ‘just’ word learning. Previous research has shown that engagement in joint attention is associated with deeper information processing and enhanced

60

memory in TD individuals (Bockler et al., 2011; Frischen & Tipper, 2004; Hirotani et al., 2009; Kopp & Lindenberger, 2011). Interestingly, engagement in joint attention has recently been shown to affect recognition memory less in children with ASD than in other children (Mundy et al., 2016; however, the effect in this study pertained to IJA and not RJA), suggesting that joint atten-tion may not be associated with the same learning benefits in ASD as in TD.

Apart from replicating a previous finding, Study III also aimed to investigate whether the pattern of less differentiation between attended and unattended objects in ASD could be altered by varying the interest level of the target ob-jects. The short answer to this question turned out to be ‘no’. Although the total looking time metric indicated that we succeeded in our manipulation of the interest level, including highly interesting objects did not result in more differentiating between attended and unattended objects in the ASD group. The fact that we actually detected an opposite pattern in the TD group (which will be discussed later on) suggests that we were wrong in assuming that using objects that are more interesting would result in more differentiating in terms of first fixation durations. The fact that we were unsuccessful in our approach to alter the pattern of first fixation lengths in autistic children therefore does not rule out the possibility that this can be achieved using other methods.

Do the Findings of Study III Indicate Specifically Social Atypicalities? The finding that fixation durations of children with ASD are less affected by contextual cues is not necessarily limited to the social domain. It has previ-ously been shown that adults with ASD differentiate less between contextually congruent and incongruent pictures in terms of first fixation durations than neurotypical individuals (Benson, Castelhano, Au-Yeung, & Rayner, 2012). It has also been demonstrated that whereas typically developing individuals display a pattern of increasing fixation lengths during scene viewing (e.g. Irwin & Zelinsky, 2002; Pannasch, Helmert, Herbold, Roth, & Walter, 2008), fixation lengths of HR infants are less variable (Wass et al., 2015). This im-plies that the results of Study III and Falck-Ytter et al. (2015b) do not neces-sarily represent a specific RJA alteration. Rather, the results may reflect a broader tendency of fixations being less susceptible to contextual factors in ASD than in TD. This would be in line with the reasoning of some accounts that focus on more general perceptual and cognitive differences between au-tistic and TD individuals. It has, for example, been argued that a favoring of local over global processing may be a core characteristic of the autistic phe-notype (e.g. Mottron et al., 2006; Van Boxtel & Lu, 2013). Similarly, individ-uals with ASD have been argued to be less influenced by previous experiences when processing sensory information, which could be attributed to reduced

61

top-down processing (Pellicano & Burr, 2012) or enhanced bottom-up pro-cessing (Brock, 2012). It should be noted however, that even if fixation lengths are generally less affected by contextual factors in ASD than in TD, this could still be expected to have specific consequences in the social domain.

A Final Remark on the Observed Looking Patterns of Children with ASD In the previous paragraphs, I have argued that the observed tendency of autis-tic children to be less influenced by other’s looking behavior may impact learning negatively. I would like to end this section by acknowledging that such a tendency may however not only be associated with disadvantages. The sensory processing style of autistic individuals has been associated with ben-eficial features such as enhanced pattern detection, superior performance on tasks requiring attention to detail, and with the presence of talent and savant skills (Baron-Cohen, Ashwin, Ashwin, Tavassoli, & Chakrabarti, 2009a; Happe & Frith, 2006; Mottron et al., 2013). Similarly, a lower influence of others’ looking behavior could potentially represent an advantage in some cir-cumstances. Although it is usually beneficial to focus on the same things as others early in life, one can easily imagine situations later in development when the opposite may be true. To be less affected by others’ looking patterns may entail greater possibilities to see things ‘as they really are’ and to provide new perspectives and insights.

Findings with Relevance for Typical Development Although the focus of this thesis is on ASD, each of the three studies have contributed to our knowledge concerning joint attention behaviors in typical development as well. Study I showed that the infants of both groups were able to follow gaze even when the model kept her head still and only used eye gaze as a directional cue. In a previous study, Moore and Corkum (1998) demon-strated that typically developing 18-19-month-olds, but not 15-16-month-olds, followed the gaze of a model who used only her eyes, suggesting that the abil-ity to follow gaze based on eye direction only does not develop until the sec-ond year of life. Their task can however be argued to have been substantially more demanding than ours. The objects in that task were placed outside the child’s visual field, which has previously been demonstrated to be too difficult for infants under 12 months (Butterworth & Jarrett, 1991), and the experi-menter shifted gaze silently. In our study, the objects were within the infant’s visual field, and the experimenter started each trial by calling the infant’s name. Our study thus demonstrates that in response to an experimenter who indicates her intent to communicate, even 10-month-olds are able to follow

62

gaze to near objects in the absence of head cues. This is in line with previous research suggesting that sensitivity for the role of others’ eyes develops be-tween 8 and 10 months (Brooks & Meltzoff, 2005; Csibra & Volein, 2008; Senju et al., 2008; Senju et al., 2006).

As mentioned in the introduction, relatively little research has focused on the relation between IJA behaviors and social functioning in typical development. Much of what we know about this relation comes from the ASD field. A pre-vious infant sibling study (Ibanez et al., 2013) reported a stronger relation be-tween IJA and ASD symptoms in an HR than in an LR sample, and found that only the interaction between IJA and risk status was a significant predictor of later symptom level, which could indicate that the relation between IJA and social functioning is specific to ASD. The results of Study II do not support this, as we found similar significant relations between IJA and ASD symptoms in both groups. This is in line with a previous finding of an association be-tween IJA frequency at 12 months and social functioning at 30 months in a TD sample (Van Hecke et al., 2007). The combined findings from that study and ours suggest that a higher tendency to engage in IJA early in life is related to more optimal later social functioning in general. As such, the results of Study II are also congruent with the view that ASD represents the extreme end of a continuum, rather than a discrete entity (e.g. Lundström et al., 2012).

In Study III, we made an unexpected finding regarding the TD sample. As previously discussed, one of the aims of the study was to assess whether in-cluding a highly interesting target object would increase the first fixation du-ration differentiation between attended and unattended objects in the ASD group. This was not the case, as the AUF-index of this group did not differ between conditions. Instead, we discovered that the TD group displayed a higher AUF-index in response to the baseline objects than to the high interest objects. We suggested that this might indicate that another’s gaze has a greater influence on object processing when the target objects are less interesting than when the objects themselves are highly interesting. In other words, the gaze cue may have a stronger effect when it is directed toward an object that is not perceived as particularly interesting, as compared to when the target object in itself is highly interesting or salient. Although this post-hoc explanation war-rants further investigation, Study III has contributed to our understanding of how characteristics of the target objects influence RJA and processing in typ-ical development.

63

Methodological Implications Studies of joint attention are often conducted during real time interaction, for example, by the use of standardized instruments such as the Early Social Com-munication Scales (ESCS; Mundy, Hogan, & Doehring, 1996) or the Screening Tool for Autism in Two-year-olds (STAT; Stone et al., 2000). Conducting measurements during interaction has the advantage of providing a high degree of ecological validity. Lately, an increasing number of studies have started to instead use pre-recorded video stimuli in combination with eye tracking (e.g. Akechi, Kikuchi, Tojo, Osanai, & Hasegawa, 2013; Akechi et al., 2011; Bedford et al., 2012; Falck-Ytter et al., 2012; Falck-Ytter et al., 2015b; Gliga et al., 2012). Eye tracking has the advantage of generating very exact measures in terms of various looking time metrics such as latencies, total looking time, and first fixation durations. Although these kind of metrics are useful for con-ducting fine-grained analyses of gaze behavior, eye tracking, in its traditional application, is associated with one major drawback – namely the use of pre-recorded stimuli. Although it is very common to use video stimuli in studies of social cognition, evidence suggests that some caution should be taken when doing so. It has been shown that humans look at other humans differently when observing them in real life, as opposed to when watching them on video (Foulsham, Walker, & Kingstone, 2011; Laidlaw, Foulsham, Kuhn, & Kingstone, 2011). In addition, imaging studies have revealed that watching stimuli live vs. on video results in differences in terms of neurological activa-tion, both in adults (Redcay et al., 2010) and infants (Shimada & Hiraki, 2006). A common goal of Study I and Study II was to combine the advantages of naturalistic and eye tracking-based methods. To do so, we developed a live eye tracking paradigm, in which the gaze of the infant is recorded during in-teraction. Live eye tracking has previously been successfully used to measure gaze following in TD infants (Gredebäck et al., 2010) and older children with ASD (Falck-Ytter, 2015; Falck-Ytter, Carlström, & Johansson, 2015a). Alt-hough it is my hope that our use of live stimuli enhances the validity and gen-eralizability of the results of both studies, the live application may be espe-cially important for Study II. It could be argued that following the gaze of a person on video will allow the viewer to see what the other person sees, which may be useful even if the other person is not real. In contrast, trying to direct the attention of a pre-recorded person on screen is pointless. As soon as infants display some level of appreciation for the difference between a real person and a person on screen, attempts to measure IJA in response to pre-recorded stimuli will therefore be fruitless.

In addition to increasing the ecological validity, using eye tracking in a live setting also has the advantage of allowing for greater flexibility than when video stimuli are used. Eye tracking with young infants is typically associated with substantial data loss, since infants tend to look away from the screen, get

64

fussy, etc. While conducting experiments live, the experimenter has the pos-sibility to attempt to re-engage the infants’ attention as needed, which likely results in an increased number of valid trials. Study I and Study II are both successful examples of how eye tracking technology can be used in a live set-ting in order to increase both data quality and validity. However, the point of this paragraph is not to argue that live eye tracking is under all circumstances superior to more naturalistic methods or to video-based eye tracking. It is my conviction that all three approaches have their advantages as well as disad-vantages, and that they complement each other. Whereas screen-based eye tracking can contribute with a high degree of standardization, clinical instru-ments and more naturalistic paradigms have a higher potential for capturing behaviors as they naturally occur. Thus, I believe that a variety of methods are needed to better understand complex phenomena such as joint attention.

In contrast to Study I and II, Study III was conducted with video stimuli. The reason for this was primarily practical, as the data preparation and analysis associated with live eye tracking is considerably more time- and resource con-suming. As mentioned in the previous section though, there is reason to be-lieve that gaze following could be less sensitive to the live vs. video distinction than IJA. In addition, a number of studies using pre-recorded stimuli to assess gaze following in ASD have generated highly interesting and replicable results (e.g. Akechi et al., 2011; Bedford et al., 2012; Falck-Ytter et al., 2015b; Gliga et al., 2012). Nevertheless, replication of the findings of Study III in a live setting would be preferable. The main methodological contribution of Study III is, in my view, its use of the first fixation duration measure. Although not as commonly used as total looking times, first fixation durations are believed to indicate information acquisition and cognitive processing of visual ele-ments (Holmqvist et al., 2011). It has been shown that first fixations are longer in response to unexpected objects than to expected objects (De Graef, Christiaens, & d'Ydewalle, 1990; Loftus & Mackworth, 1978), and longer first fixation durations are typically considered an indication of deeper processing. We have argued that an advantage of the first fixation duration measure is that it has the potential to capture processing biases that occur on very short time-scales, thus without confounding influence of factors such as sustained atten-tion. To my knowledge, only two previous studies have used first fixation du-rations in a gaze following context (Falck-Ytter et al., 2015b; Swanson & Siller, 2013). These two studies and Study III are all congruent in that they report a pattern of less differentiation between attended and unattended objects in ASD than in TD. Together, the studies therefore suggest that the first fixa-tion duration metric can be used to generate replicable findings that contribute to our understanding of information processing in an RJA context.

65

Clinical Implications All three studies in this this thesis are experimental studies aiming predomi-nately at contributing to our understanding of the development of joint atten-tion behaviors in infants and children with or at risk for ASD. The ambition behind conducting the studies was not to contribute directly to either ASD detection or intervention. Nevertheless, the results of the studies entail some implications that may have clinical value. In Study II, we showed that the fre-quency of gaze alternations between an interaction partner and an event at 10 months could predict later ASD symptomatology. This early measure alone explained 14% of the variance of later ADOS-T scores, thus contributing more explained variance than the group factor. This demonstrates that early alter-nating gaze, which is quick and easy to measure, may show potential to be used in early screening.

The fact that Study III demonstrated that children with ASD may differ from other children in terms of the processing of the objects that are the common focus of attention is also interesting from a clinical point of view. As already accounted for, this may indicate that autistic children benefit less than TD children from the learning opportunities associated with joint attention. If this is the case, teaching children with ASD to follow gaze may not be enough. In addition, interventions should aim to target the associated object processing, for example by attempting to reinforce longer fixations at attended objects (which can be achieved by the use of gaze-contingent eye tracking), or by explicitly teaching that objects looked at by others may be particularly inter-esting.

Limitations The infants of Study I were at the time of analysis not old enough for diagnos-tic instruments to be used, and the comparisons were therefore only based on group status. Around 20% of the infants in the HR group are expected to later meet diagnostic criteria for ASD (Messinger et al., 2015; Ozonoff et al., 2011). Of the rest, some are expected to display typical development and others to display milder developmental concerns including language delays (Zwaigenbaum et al., 2007). The fact that only a minority of the HR infants are expected to receive an ASD diagnosis represents a common issue for all studies conducting analyses based on risk status. If our sole interest were the infants with a later diagnosis, the design could be argued to be underpowered. However, adopting a dimensional view of ASD, and knowing that undiag-nosed family members of autistic individuals share many characteristics with those who are diagnosed (e.g. Constantino et al., 2006; Lundstrom et al., 2012; Sucksmith et al., 2011), I would argue that comparing individuals based on

66

group status has merit on its own. The fact that we discovered an effect in-volving group status in Study I suggests that the HR infants as a group differed from the LR infants. Whether the observed effect was driven by those infants who will receive a later diagnosis, or whether it reflects a more general differ-ence between infants with and without a familial history of ASD – and thus perhaps be more related to the BAP than to diagnosis specifically – remains to be seen once the infants are old enough to go through diagnostic assessment.

Unlike Study I, Study II included follow-up data when the infants were 18 months old and assessed with the ADOS-T. Eighteen months is an interesting age for follow-up, since signs of ASD are then detectable in several areas of development (Jones et al., 2014). Nevertheless, ASD is typically not diag-nosed until slightly later, and we therefore refrained from conducting categor-ical analyses based on the outcome data. Although Study II can inform us re-garding the relation between alternating gaze and autistic symptoms, no con-clusions regarding actual diagnostic status can be drawn.

In my view, the greatest limitations of Study III are related to the first fixation duration metric. Although I have previously argued that the use of this meas-ure is one of the greatest contributions of the study, the measure is also asso-ciated with a level of uncertainty. First fixation durations have not been ex-tensively used in studies of information processing during scene viewing. Alt-hough evidence suggests that the measure likely indicates degree of infor-mation processing (Holmqvist et al., 2011; Loftus & Mackworth, 1978), further studies of how first fixations are affected by various factors are needed.

As mentioned, the AUF-index is intended to reflect the degree to which at-tended objects are associated with a processing bias compared to unattended objects. However, it is difficult to determine a limit for when such a processing bias exists. Previous studies have shown that first fixations tend to become increasingly longer over time during scene viewing (e.g. Antes, 1974; Galpin & Underwood, 2005; Irwin & Zelinsky, 2002; Pannasch et al., 2008). This entails that first fixations at unattended objects should be expected to be longer due to the objects always being fixated later than the attended objects (in the current analysis), and testing the values against .5 is therefore not applicable. This is not a problem for the group comparisons, but it nevertheless renders the interpretation of the results more complicated. It is also problematic that fixation lengths may differ between groups in other aspects than the one being the focus of Study III. One study (Wass et al., 2015) has indicated that HR infants display generally shorter fixations, and that they do not display the typical pattern of increasing fixation lengths during scene viewing. If the same is true for older children with ASD, it would entail that the first fixations of the ASD group should not be expected to be longer at unattended objects be-cause of them being fixated later than attended objects. However, this would

67

bias the AUF-index upward in the ASD group relative to the TD group, and could therefore not explain the finding of a higher AUF-index in TD compared to ASD. Finally, by focusing on the relation between fixations at attended and unattended objects, more general group differences are controlled for.

Another weakness of Study III is that we did not directly assess for object type preference. Although the use of total looking time as an indication of interest seems relatively straightforward, and the parents reported a higher incidence of circumscribed interests in the ASD group than the TD group, it would have been preferable to also ask the participants to disclose information about the specific interest areas.

In addition to their individual limitations, all three studies suffer from some problems related to sample size and amount of data. Both Study I and Study II included substantially more HR than LR infants. This is relatively common in infant sibling studies, partially because the HR children are expected to form several subgroups after diagnostic assessment (Zwaigenbaum et al., 2007). Nevertheless, the uneven group sizes imply different power to detect within-group effects. Importantly though, the unequal sample sizes do not af-fect between-group or interaction effects. Although not sharing the problem of unequal sample sizes, Study III suffered from substantial data loss. This was partly due to our decision to include four target objects in the experiment. During analysis, it was discovered that the children relatively often looked toward one of the two objects that were placed in the general direction of the model’s gaze, before she had finished her gaze shift. We were therefore not able to determine whether a look to the object next to (and on the same side as) the attended object reflected a “correct but too fast” response, or an “in-correct” response. Because of this, all such trials were excluded (note that their number did not differ between groups). When analyzing first fixation dura-tions, we were only interested in those trials where the children did follow gaze, which inevitably resulted in the exclusion of those trials where the chil-dren did not follow gaze, thus reducing the number of trials included in this analysis even more. In hindsight, it would have been preferable to only include one object on each side of the table. Although the loss of trials with incorrect responses is inevitable, this loss could have been compensated for to some extent by including a higher total number of trials.

Future Directions In the discussion of Study I, two possible interpretations of the pattern of re-sults were presented. We suggested that the fact that the performance of the HR infants was reduced when head information was absent could be due to either a lower sensitivity to eye information or to a more general need for

68

increased cue saliency. Further research should aim to tease apart these ex-planatory models. Brooks and Meltzoff (2005) have shown that whereas 10-month-old infants follow gaze significantly more often when the model’s eyes are open than when they are closed, the behavior of 9-month-olds do not differ between conditions. To my knowledge, nobody has tested whether 10-month-olds at risk for ASD follow the head movement of a model with closed eyes. If less differentiation between conditions in such an experiment were to be found in an HR sample, this could be interpreted as supporting the hypothesis of a lower sensitivity specifically to eye information. If on the other hand HR infants, like typically developing 10-month-olds, would display higher gaze following accuracy in response to open eyes than to closed eyes, this hypoth-esis would not be supported. Thus, applying Brooks and Meltzoffs (2005) par-adigm on an HR sample may cast some light on the mechanisms behind the findings of Study I.

The fact that Study I and Study II focused on 10-month-olds is a strength since it allows for the detection of early differences in behavior. However, it is not possible to determine whether the differences represent impairment or delay. We thus do not know whether omitting head movement will still affect the gaze following performance of the HR infants negatively a few months later, once they have gained more experience with gaze following in different con-texts. Similarly, we cannot tell whether the HR infants of Study II will con-tinue to display less alternating gaze than the LR infants. In order to determine whether the obtained differences will remain during development or not, fu-ture studies should aim to include multiple measurement points between the first and second birthday (data collected at 14 and 18 months will later be analyzed in the EASE project). It would also be interesting to combine the results of Study I and Study II and to investigate the relations between early gaze following and early alternating gaze, as well as the relations between these measurements and more general developmental measures.

The fact that we were unsuccessful in our attempt to change the fixation pat-terns of autistic children by increasing the interest level of the objects in Study III does not necessarily entail that this cannot be achieved by other methods. In the future, it may be interesting to investigate whether it is possible to obtain the typical pattern of longer fixations at attended objects by the use of positive reinforcement techniques. One could, for example, use stimuli in which the target object changes contingent on the length of the child’s first fixation, so that fixating on attended objects longer would result in a new interesting sight (i.e., gaze contingent eye tracking methods). The finding that less interesting target objects are associated with relatively longer first fixation durations after gaze following in typical development also warrants further investigation. In addition to replicating the current finding – where the focus was on interesting

69

and neutral objects - it would be interesting to investigate the effect associated with objects with other emotional valences as well.

All three studies included in this thesis focused on children with or at risk for ASD, as well as on TD children. We are therefore not able to draw any con-clusions regarding whether the discovered effects are associated specifically with ASD, or whether they may also be associated with any of the commonly co-occurring conditions. Future studies should therefore aim to include groups of children with or at risk for conditions such as ADHD, specific language impairment and intellectual disability.

Final Conclusions The studies in this thesis have used different eye tracking approaches to in-vestigate aspects of joint attention both early and later in development. We have seen that infants at familial risk for ASD differ from other infants in terms of both gaze following and alternating gaze, and even before their first birthday. These joint attention behaviors thus represent some of the earliest detectable alterations in the social domain. In the third study, we learned that although older autistic children follow gaze, they may be less influenced by others’ looking behavior during object processing. This shows that joint atten-tion continues to be an area where children with ASD differ from other chil-dren in ways that may affect learning and social interaction.

Together, the studies contribute to our understanding of the role that joint at-tention may play in the early development of infants at risk for ASD, as well as in the later development of diagnosed children. Eventually, such an in-creased understanding may contribute to the development of better diagnostics and early intervention programs.

70

Summary in Swedish

Denna avhandling undersöker hur små barn som antingen har Autism-spektrumtillstånd (AST) eller har en ökad sannolikhet för att senare få AST skiljer sig från andra barn avseende delad uppmärksamhet (eng. joint attent-ion). Delad uppmärksamhet är ett begrepp som används för att beskriva hur två personer tillsammans uppmärksammar ett objekt. Ett barn som pekar ut ett flygplan på himlen för sin förälder kan exemplifiera delad uppmärksamhet. Konceptet kan delas upp i två komponenter: initiering av delad uppmärksam-het – barnet som pekar ut flygplanet – och respons på delad uppmärksamhet – föräldern som följer barnets pekning och blickriktning. Förmågan att dela upp-märksamhet börjar utvecklas under det första levnadsåret, och har betydelse för barnets utveckling och inlärning.

AST är en neuropsykiatrisk funktionsnedsättning som karakteriseras av avvi-kelser i förmågan till socialt samspel och kommunikation, samt av förekoms-ten av repetitiva beteendemönster. AST diagnostiseras sällan före 2-3-årsål-dern, vilket medför begränsningar i hur mycket vi kan lära oss om den tidig-aste utvecklingen genom att studera barn med diagnos. Att AST i hög grad har genetiska orsaker innebär dock att vi vet att i en familj där ett eller flera barn har en AST-diagnos är sannolikheten att ett yngre syskon också ska få en så-dan diagnos 7-20%, d.v.s. betydligt högre än den allmänna prevalensen på 1 - 1.5%. Dessutom brukar ytterligare en andel av de yngre syskonen uppvisa autismlika drag. För att undersöka tidig utveckling väljer därför många fors-kare att studera yngre syskon till barn med AST. Två av studierna i denna avhandling fokuserar på yngre syskon till barn med AST, medan den tredje studien fokuserar på diagnostiserade barn. I samtliga studier använde vi oss av ögonrörelsemätning, som är en metod för att med hjälp av infrarött ljus få exakta mått på hur barn tittar. I två av studierna mätte vi barnens blick medan de interagerade med en annan person (experimentledaren). I den tredje studien användes istället förinspelat videomaterial som barnen fick titta på.

Man vet att barn med AST är mindre benägna att dela uppmärksamhet än andra barn, och att detta sannolikt kan påverka deras utveckling negativt, men många aspekter av hur delad uppmärksamhet och AST hänger ihop är fortfa-rande okända. Den första studien i denna avhandling undersökte hur 10 må-nader gamla barn med ökad sannolikhet för AST skiljer sig från andra barn avseende förmåga att följa blick, vilket ofta används som ett mått på respons

71

på delad uppmärksamhet. Barnen fick se en person titta på ett av två objekt. Personen vred ibland hela huvudet mot objektet, och höll ibland huvudet stilla och riktade bara ögonen mot objektet. Resultaten visade att medan de barn som inte hade något äldre syskon med AST var lika benägna att följa blick i båda betingelserna, så följde barnen med ett äldre syskon med AST blick i större utsträckning då personen använde både ögon och huvud än när hen en-bart använde sina ögon. Detta skulle kunna bero på att barn med förhöjd san-nolikhet för AST är mindre känsliga för ögoninformation, eller på att de mer generellt behöver tydligare signaler för att följa blick optimalt.

I den andra studien var fokus på barnets initiering av delad uppmärksamhet. Ett av de tidigaste sätten på vilka barn initierar delad uppmärksamhet är ge-nom att titta fram och tillbaka mellan en interaktionspartner och ett intressant objekt eller en händelse. I studien visade vi att barn med förhöjd sannolikhet för AST vid 10 månader gjorde färre sådana blickväxlingar mellan en person och en ovanlig händelse (lampor som började blinka) än vad barn utan förhöjd sannolikhet för AST gjorde. När barnen blev 18 månader mätte vi förekoms-ten av AST-symptom. Vi kunde då påvisa en relation mellan blickväxlingar vid 10 månader och AST-symptom vid 18 månader, d.v.s. de barn som gjorde färre blickväxlingar tidigt hade en högre förekomst av AST-symptom senare. Denna relation kunde inte förklaras av skillnader i barnens förmåga att flytta sin uppmärksamhet från ett objekt till ett annat, eller av deras generella prefe-rens för människor över icke-sociala objekt. Slutligen kunde vi visa att de barn som använde fler blickväxlingar tidigt var mer benägna att använda pekningar och att visa saker för andra vid 18 månader. Att initiera delad uppmärksamhet genom blickväxlingar eller genom att peka eller visa är beteenden som får andra att kommunicera och interagera med barnet. Om ett barn använder färre sådana beteenden tidigt i livet kan det därför ha negativa konsekvenser för den senare utvecklingen.

I den tredje studien undersökte vi förmåga att följa blick hos lite äldre barn med AST och en kontrollgrupp. Barnens ögonrörelser spelades denna gång in medan de såg en video av en person som tittade på ett av fyra objekt. Resulta-ten av denna studie visade att barnen med AST inte var sämre på att följa blick än barnen i kontrollgruppen, men att deras första fixering på det objekt den andre tittade på var kortare än hos andra barn. Detta resultat, som är i linje med resultatet från en annan nyligen publicerad studie, kan tyda på att barn med AST skiljer sig från andra barn i hur de bearbetar information från objekt som andra personer tillskriver betydelse genom sin blick. Det kan vara så att andras blick har mindre påverkan på informationsbearbetningen hos barn med autism jämfört med barn utan autism. I studien fann vi också att fixeringsläng-den hos typiska barn påverkades mer av den andres blick om objektet på skär-men uppfattades som relativt ointressant. Detta skulle kunna innebära att den

72

andres blick har störst betydelse om hen tittar på något som i sig inte är intres-seväckande. För barn med AST fanns ingen sådan skillnad mellan intressanta och icke-intressanta objekt. Studien visar att det inte bara är viktigt att under-söka huruvida barn med AST följer blick eller inte, utan att man genom mät-ning av fixeringslängder kan lära sig mer om hur barnen använder andras blick som ett stöd i bearbetningen av icke-social information i miljön.

Tillsammans bidrar studierna i denna avhandling till vår förståelse för den roll förmågan till delad uppmärksamhet spelar både i den tidiga utvecklingen hos barn med förhöjd sannolikhet för autism, och senare i livet för barn med AST-diagnos. På sikt kan denna förståelse förhoppningsvis bidra till utvecklingen av förbättrad diagnostik och tidiga interventionsmetoder.

73

Acknowledgements

Pursuing a PhD has been one of the most challenging things I have ever de-cided to do, but also one of the most meaningful. I have learned so much dur-ing these five years – about science, autism, writing, statistics, teaching, col-laboration, and about myself. There are of course a number of people who in different ways have been instrumental in this process - people without whom this thesis would never have come to be, and who deserve a big THANK YOU!

First of all, I would like to thank all the children and parents that have partic-ipated in the studies. I am especially thankful to the EASE-families, who have generously given us their time, spending many days in our labs. Some of them have even travelled across the country to have us blow bubbles and show strange videos to their kids! Thank you so much for trusting us with your pre-cious babies, and for your important contribution to science!

Then, I want to thank my supervisor, Terje Falck-Ytter. I have been truly lucky to have you as my supervisor, and you have been nothing but encouraging, inspiring and supportive. I have learned so much from you! I am also thankful to my second supervisor, Gustaf Gredebäck, for his enthusiasm, guidance, and support.

I want to thank Karin Brocki and Marcus Lindskog for their useful comments on an earlier version of this thesis and for their useful contributions on my full-time seminar; as well as Christine Fawcett for her valuable input at my half-time seminar.

Then there is the EASE-team, which over the years has been full of smart and fun people. Johan and Elisabeth, I have much appreciated sharing the majority of this journey with you. I want to thank Eric for sharing his clinical expertise, and Pär for his generous help and support regarding all technical matters. Thank you Sven for all your knowledge and expertise. And thank you Maria – who for quite some time was truly the backbone of EASE – for everything! Thank you also to Therese, Sheila, Elodie, Linn, Isabelle, Emma, Johanna, Joy and all the other amazing people who are or have been part of this great team!

74

Luckily, this journey has not been all about work and seriousness. Janna and Laura – you have both been the best office mates I could have wished for! I want to thank Matilda, Andreas, Johanna, Eva, Jonas, Tommie, Hanna, Mari, Carin, Kirsti, Kristin, Goncalo, Dorota, Erik, Kahl, Joakim, Mona, Viktor, Philip, Pernilla, Sara, Olga, Martyna, Elin, Elina, Marta, Claudia, Andrea and Benjamin for all the shared lunch breaks and beers. It has been so fun getting to know you all!

Last but not least, I want to thank my friends and family. I want to thank my parents, Anne and Ronny, for their love and support over the years; and my siblings, Nils and Lisabet, for fun childhood memories; and also my grand-mother Gerd. And finally I want to thank Tim, for sharing my life, challenging me, and for believing in me when I do not.

75

References

Akechi, H., Kikuchi, Y., Tojo, Y., Osanai, H., & Hasegawa, T. (2013). Brief Report: Pointing Cues Facilitate Word Learning in Children with Autism Spectrum Dis-order. Journal of Autism and Developmental Disorders, 43(1), 230-235.

Akechi, H., Senju, A., Kikuchi, Y., Tojo, Y., Osanai, H., & Hasegawa, T. (2011). Do children with ASD use referential gaze to learn the name of an object? An eye tracking study. Research in Autism Spectrum Disorders, 5(3), 1230-1242.

Akhtar, N., & Gernsbacher, M. A. (2007). Joint Attention and Vocabulary Develop-ment: A Critical Look. Language and Linguistics Compass.

Antes, J. R. (1974). The time course of picture viewing. Journal of Experimental Psy-chology, 103(1), 62-70.

Asperger, H. (1944). The ''autistic psychopathy'' in childhood. Archiv Fur Psychiatrie Und Nervenkrankheiten, 117(1), 76-136.

Bailey, A., Lecouteur, A., Gottesman, I., Bolton, P., Simonoff, E., Yuzda, E., & Rut-ter, M. (1995). Autism as a strongly genetic disorder - evidence from a British twin study. Psychological Medicine, 25(1), 63-77.

Baird, G., Simonoff, E., Pickles, A., Chandler, S., Loucas, T., Meldrum, D., & Char-man, T. (2006). Prevalence of disorders of the autism spectrum in a population cohort of children in South Thames: the Special Needs and Autism Project (SNAP). Lancet, 368(9531), 210-215.

Baker, M. J. (2000). Incorporating the Thematic Ritualistic Behaviors of Children with Autsim into Games: Increasing Social Play Interactions with Siblings. Jour-nal of Positive Behavior Interventions, 2(2).

Baldwin, D. A. (1991). Infants contribution to the achievement of joint reference. Child Development, 62(5), 875-890.

Barneveld, P. S., Swaab, H., Fagel, S., van Engeland, H., & de Sonneville, L. M. J. (2014). Quality of life: A case-controlled long-term follow-up study, comparing young high-functioning adults with autism spectrum disorders with adults with other psychiatric disorders diagnosed in childhood. Comprehensive Psychiatry, 55(2), 302-310.

Baron-Cohen, S. (1989). Perceptual role-taking and protodeclarative pointing in au-tism. British Journal of Developmental Psychology, 7, 113-127.

Baron-Cohen, S. (1995). Mindblindness: an essay on autism and theory of mind. Cam-bridge, MA: MIT Press.

Baron-Cohen, S., Ashwin, E., Ashwin, C., Tavassoli, T., & Chakrabarti, B. (2009a). Talent in autism: hyper-systemizing, hyper-attention to detail and sensory hyper-sensitivity. Philosophical Transactions of the Royal Society B-Biological Sci-ences, 364(1522), 1377-1383.

Baron-Cohen, S., Jolliffe, T., Mortimore, C., & Robertson, M. (1997). Another ad-vanced test of theory of mind: Evidence from very high functioning adults with autism or Asperger syndrome. Journal of Child Psychology and Psychiatry and Allied Disciplines, 38(7), 813-822.

76

Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child have a theory of mind. Cognition, 21(1), 37-46.

Baron-Cohen, S., Lombardo, M. V., Auyeung, B., Ashwin, E., Chakrabarti, B., & Knickmeyer, R. (2011). Why Are Autism Spectrum Conditions More Prevalent in Males? Plos Biology, 9(6).

Baron-Cohen, S., Scott, F. J., Allison, C., Williams, J., Bolton, P., Matthews, F. E., & Brayne, C. (2009b). Prevalence of autism-spectrum conditions: UK school-based population study. British Journal of Psychiatry, 194(6), 500-509.

Baron-Cohen, S., Wheelwright, S., & Jolliffe, T. (1997). Is there a "language of the eyes"? Evidence from normal adults, and adults with autism or Asperger Syn-drome. Visual Cognition, 4(3), 311-331.

Bates, E., Camaioni, L., & Volterra, V. (1975). The acquisition of performatives prior to speech. Merrill-Palmer quarterly of behavior and development, 21(3), 205-226.

Batki, A., Baron-Cohen, S., Wheelwright, S., Connellan, J., & Ahluwalia, J. (2000). Is there an innate gaze module? Evidence from human neonates. Infant Behavior & Development, 23(2), 223-229.

Bedford, R., Elsabbagh, M., Gliga, T., Pickles, A., Senju, A., Charman, T., & Johnson, M. (2012). Precursors to Social and Communication Difficulties in Infants At-Risk for Autism: Gaze Following and Attentional Engagement. Journal of Autism and Developmental Disorders, 42(10), 2208-2218.

Belmonte, M. K., Allen, G., Beckel-Mitchener, A., Boulanger, L. M., Carper, R. A., & Webb, S. J. (2004). Autism and abnormal development of brain connectivity. Journal of Neuroscience, 24(42), 9228-9231.

Benson, V., Castelhano, M. S., Au-Yeung, S. K., & Rayner, K. (2012). Eye move-ments reveal no immediate "WOW" ("which one's weird") effect in autism spec-trum disorder. Quarterly Journal of Experimental Psychology, 65(6), 1139-1150.

Bettelheim, B. (1967). Empty fortess: Infantile autism and the birth of self. New York: Free Press.

Beuker, K. T., Rommelse, N. N. J., Donders, R., & Buitelaar, J. K. (2013). Develop-ment of early communication skills in the first two years of life. Infant Behavior & Development, 36(1), 71-83.

Bigelow, A. E. (2003). The development of joint attention in blind infants. Develop-ment and Psychopathology, 15(2), 259-275.

Bockler, A., Knoblich, G., & Sebanz, N. (2011). Giving a helping hand: effects of joint attention on mental rotation of body parts. Experimental Brain Research, 211(3-4), 531-545.

Botero, M. (2016). Tactless scientists: Ignoring touch in the study of joint attention. Philosophical Psychology, 29(8), 1200-1214.

Bottema-Beutel, K. (2016). Associations between joint attention and language in au-tism spectrum disorder and typical development: A systematic review and meta-regression analysis. Autism Research, 9(10), 1021-1035.

Boyd, B. A., Conroy, M. A., Mancil, G. R., Nakao, T., & Alter, P. J. (2007). Effects of circumscribed interests on the social behaviors of children with autism spec-trum disorders. Journal of Autism and Developmental Disorders, 37(8), 1550-1561.

Boyd, B. A., Odom, S. L., Humphreys, B. P., & Sam, A. M. (2010). Infants and Tod-dlers With Autism Spectrum Disorder: Early Identification and Early Interven-tion. Journal of Early Intervention, 32(2), 75-98.

Brian, J., Bryson, S. E., Garon, N., Roberts, W., Smith, I. M., Szatmari, P., & Zwaigenbaum, L. (2008). Clinical assessment of autism in high-risk 18-month-olds. Autism, 12(5), 433-456.

77

Brock, J. (2012). Alternative Bayesian accounts of autistic perception: comment on Pellicano and Burr. Trends in Cognitive Sciences, 16(12), 573-574.

Brooks, R., & Meltzoff, A. N. (2005). The development of gaze following and its relation to language. Developmental Science, 8(6), 535-543.

Brooks, R., & Meltzoff, A. N. (2015). Connecting the dots from infancy to childhood: A longitudinal study connecting gaze following, language, and explicit theory of mind. Journal of Experimental Child Psychology, 130, 67-78.

Brothers, L. (1990). The social brain: a project for integrating primate behavior and neurophysiology in a new domain. Concepts Neurosci, 1, 27-51.

Bruner, J. S. (1975). From communication to language - psychological perspective. Cognition, 3(3), 255-287.

Bryson, S. E., Zwaigenbaum, L., McDermott, C., Rombough, V., & Brian, J. (2008). The autism observation scale for infants: Scale development and reliability data. Journal of Autism and Developmental Disorders, 38(4), 731-738.

Butterworth, G. (1998). What is special about pointing in babes? In F. Simion & G. Butterworth (Eds.), The development of sensory, motor and cognitive capacities in early infancy: From perception to cognition (pp. 171-190). Hove: Psychology Press.

Butterworth, G., & Jarrett, N. (1991). What minds have in common is space: Spatial mechanisms serving joint visual attention in infancy. British Journal of Develop-mental Psychology, 9(1), 55-72.

Camaioni, L., Perucchini, P., Bellagamba, F., & Colonnesi, C. (2004). The role of declarative pointing in developing a theory of mind. Infancy, 5(3), 291-308.

Carpenter, M., Nagell, K., & Tomasello, M. (1998). Social cognition, joint attention, and communicative competence from 9 to 15 months of age. Monographs of the Society for Research in Child Development, 63(4), V-143.

Cassel, T. D., Messinger, D. S., Ibanez, L. V., Haltigan, J. D., Acosta, S. I., & Buch-man, A. C. (2007). Early social and emotional communication in the infant sib-lings of children with autism spectrum disorders: An examination of the broad phenotype. Journal of Autism and Developmental Disorders, 37(1), 122-132.

Charman, T. (2003). Why is joint attention a pivotal skill in autism? Philosophical Transactions of the Royal Society of London Series B-Biological Sciences, 358(1430), 315-324.

Charman, T., Baron-Cohen, S., Swettenham, J., Baird, G., Cox, A., & Drew, A. (2000). Testing joint attention, imitation, and play as infancy precursors to lan-guage and theory of mind. Cognitive Development, 15(4), 481-498.

Chawarska, K., Klin, A., Paul, R., & Volkmar, F. R. (2007). Autism spectrum disorder in the second year: stability and change in syndrome expression. Journal of Child Psychology and Psychiatry and Allied Disciplines, 48(2), 128-138.

Chawarska, K., Klin, A., & Volkmar, F. (2003). Automatic attention cueing through eye movement in 2-year-old children with autism. Child Development, 74(4), 1108-1122.

Chawarska, K., Macari, S., & Shic, F. (2013). Decreased Spontaneous Attention to Social Scenes in 6-Month-Old Infants Later Diagnosed with Autism Spectrum Disorders. Biological Psychiatry, 74(3), 195-203.

Chawarska, K., Shic, F., Macari, S., Campbell, D. J., Brian, J., Landa, R., . . . Bryson, S. (2014). 18-Month Predictors of Later Outcomes in Younger Siblings of Chil-dren With Autism Spectrum Disorder: A Baby Siblings Research Consortium Study. Journal of the American Academy of Child and Adolescent Psychiatry, 53(12), 1317-1327.

Chevallier, C., Kohls, G., Troiani, V., Brodkin, E. S., & Schultz, R. T. (2012). The social motivation theory of autism. Trends in Cognitive Sciences, 16(4), 231-238.

78

Chiang, C.-H., Soong, W.-T., Lin, T.-L., & Rogers, S. J. (2008). Nonverbal Commu-nication Skills in Young Children with Autism. Journal of Autism and Develop-mental Disorders, 38(10), 1898-1906.

Christensen, L., Hutman, T., Rozga, A., Young, G. S., Ozonoff, S., Rogers, S. J., . . . Sigman, M. (2010). Play and Developmental Outcomes in Infant Siblings of Chil-dren with Autism. Journal of Autism and Developmental Disorders, 40(8), 946-957.

Clifford, S. M., Hudry, K., Elsabbagh, M., Charman, T., & Johnson, M. H. (2013). Temperament in the first 2 years of life in infants at high-risk for autism spectrum disorders. Journal of Autism and Developmental Disorders, 43(3), 673-686.

Cochet, H., & Byrne, R. W. (2016). Communication in the second and third year of life: Relationships between nonverbal social skills and language. Infant Behavior & Development, 44, 189-198.

Colombi, C., Liebal, K., Tomasello, M., Young, G., Warneken, F., & Rogers, S. J. (2009). Examining correlates of cooperation in autism Imitation, joint attention, and understanding intentions. Autism, 13(2), 143-163.

Colvert, E., Tick, B., McEwen, F., Stewart, C., Curran, S. R., Woodhouse, E., . . . Bolton, P. (2015). Heritability of Autism Spectrum Disorder in a UK Population-Based Twin Sample. Jama Psychiatry, 72(5), 415-423.

Constantino, J. N., Lajonchere, C., Lutz, M., Gray, T., Abbacchi, A., McKenna, K., . . . Todd, R. D. (2006). Autistic social impairment in the siblings of children with pervasive developmental disorders. American Journal of Psychiatry, 163(2), 294-296.

Constantino, J. N., & Todd, R. D. (2003). Autistic traits in the general population - A twin study. Archives of General Psychiatry, 60(5), 524-530.

Cornew, L., Dobkins, K. R., Akshoomoff, N., McCleery, J. P., & Carver, L. J. (2012). Atypical Social Referencing in Infant Siblings of Children with Autism Spectrum Disorders. Journal of Autism and Developmental Disorders, 42(12), 2611-2621.

Courchesne, E., Campbell, K., & Solso, S. (2011). Brain growth across the life span in autism: Age-specific changes in anatomical pathology. Brain Research, 1380, 138-145.

Courchesne, E., Pierce, K., Schumann, C. M., Redcay, E., Buckwalter, J. A., Ken-nedy, D. P., & Morgan, J. (2007). Mapping early brain development in autism. Neuron, 56(2), 399-413.

Csibra, G., & Volein, A. (2008). Infants can infer the presence of hidden objects from referential gaze information. British Journal of Developmental Psychology, 26, 1-11.

Curcio, F. (1978). Sensorimotor functioning and communication in mute autistic chil-dren. Journal of Autism and Childhood Schizophrenia, 8(3), 281-292.

d'Arc, B. F., Delorme, R., Zalla, T., Lefebvre, A., Amsellem, F., Moukawane, S., . . . Ramus, F. (2017). Gaze direction detection in autism spectrum disorder. Autism, 21(1), 100-107.

D'Entremont, B. (2000). A perceptual-attentional explanation of gaze following in 3-and 6-month-olds. Developmental Science, 3(3), 302-311.

D'Entremont, B., Hains, S. M. J., & Muir, D. W. (1997). A demonstration of gaze following in 3- to 6-month-olds. Infant Behavior & Development, 20(4), 569-572.

Dalton, K. M., Nacewicz, B. M., Johnstone, T., Schaefer, H. S., Gernsbacher, M. A., Goldsmith, H. H., . . . Davidson, R. J. (2005). Gaze fixation and the neural cir-cuitry of face processing in autism. Nature Neuroscience, 8(4), 519-526.

Daniels, A. M., & Mandell, D. S. (2014). Explaining differences in age at autism spectrum disorder diagnosis: A critical review. Autism, 18(5), 583-597.

79

Dawson, G. (2008). Early behavioral intervention, brain plasticity, and the prevention of autism spectrum disorder. Development and Psychopathology, 20(3), 775-803.

Dawson, G., Bernier, R., & Ring, R. H. (2012). Social attention: a possible early in-dicator of efficacy in autism clinical trials. Journal of Neurodevelopmental Dis-orders, 4.

Dawson, G., Munson, J., Estes, A., Osterling, J., McPartland, J., Toth, K., . . . Abbott, R. (2002). Neurocognitive function and joint attention ability in young children with autism spectrum disorder versus developmental delay. Child Development, 73(2), 345-358.

Dawson, G., Toth, K., Abbott, R., Osterling, J., Munson, J., Estes, A., & Liaw, J. (2004). Early social attention impairments in autism: Social orienting, joint atten-tion, and attention to distress. Developmental psychology, 40(2), 271-283.

De Graef, P., Christiaens, D., & d'Ydewalle, G. (1990). Perceptual effects of scene context on object identification. Psychological Research, 52(4), 317-329.

Deak, G. O., Flom, R. A., & Pick, A. D. (2000). Effects of gesture and target on 12-and 18-month-olds' joint visual attention to objects in front of or behind them. Developmental psychology, 36(4), 511-523.

Di Martino, A., Yan, C. G., Li, Q., Denio, E., Castellanos, F. X., Alaerts, K., . . . Milham, M. P. (2014). The autism brain imaging data exchange: towards a large-scale evaluation of the intrinsic brain architecture in autism. Molecular Psychia-try, 19(6), 659-667.

Elison, J. T., Paterson, S. J., Wolff, J. J., Reznick, J. S., Sasson, N. J., Gu, H., . . . Network, I. (2013). White Matter Microstructure and Atypical Visual Orienting in 7-Month-Olds at Risk for Autism. American Journal of Psychiatry, 170(8), 899-908.

Elsabbagh, M., Divan, G., Koh, Y. J., Kim, Y. S., Kauchali, S., Marcin, C., . . . Fom-bonne, E. (2012a). Global Prevalence of Autism and Other Pervasive Develop-mental Disorders. Autism Research, 5(3), 160-179.

Elsabbagh, M., Fernandes, J., Webb, S. J., Dawson, G., Charman, T., Johnson, M. H., & British Autism Study Infant, S. (2013). Disengagement of Visual Attention in Infancy is Associated with Emerging Autism in Toddlerhood. Biological Psychi-atry, 74(3), 189-194.

Elsabbagh, M., & Johnson, M. H. (2016). Autism and the Social Brain: The First-Year Puzzle. Biological Psychiatry, 80(2), 94-99.

Elsabbagh, M., Mercure, E., Hudry, K., Chandler, S., Pasco, G., Charman, T., . . . Team, B. (2012b). Infant Neural Sensitivity to Dynamic Eye Gaze Is Associated with Later Emerging Autism. Current Biology, 22(4), 338-342.

Esler, A. N., Bal, V. H., Guthrie, W., Wetherby, A., Weismer, S. E., & Lord, C. (2015). The Autism Diagnostic Observation Schedule, Toddler Module: Stand-ardized Severity Scores. Journal of Autism and Developmental Disorders, 45(9), 2704-2720.

Falck-Ytter, T. (2015). Gaze performance during face-to-face communication: A live eye tracking study of typical children and children with autism. Research in Au-tism Spectrum Disorders, 17, 78-85.

Falck-Ytter, T., Carlström, C., & Johansson, M. (2015a). Eye Contact Modulates Cog-nitive Processing Differently in Children With Autism. Child Development, 86(1), 37-47.

Falck-Ytter, T., Fernell, E., Hedvall, A. L., von Hofsten, C., & Gillberg, C. (2012). Gaze Performance in Children with Autism Spectrum Disorder when Observing Communicative Actions. Journal of Autism and Developmental Disorders, 42(10), 2236-2245.

80

Falck-Ytter, T., Thorup, E., & Bölte, S. (2015b). Brief Report: Lack of Processing Bias for the Objects Other People Attend to in 3-Year-Olds with Autism. Journal of Autism and Developmental Disorders, 45(6), 1897-1904.

Farroni, T., Csibra, G., Simion, G., & Johnson, M. H. (2002). Eye contact detection in humans from birth. Proceedings of the National Academy of Sciences of the United States of America, 99(14), 9602-9605.

Farroni, T., Johnson, M. H., Brockbank, M., & Simion, F. (2000). Infants' use of gaze direction to cue attention: The importance of perceived motion. Visual Cognition, 7(6), 705-718.

Farroni, T., Massaccesi, S., Pividori, D., & Johnson, M. H. (2004). Gaze following in newborns. Infancy, 5(1), 39-60.

Fein, D., Barton, M., Eigsti, I. M., Kelley, E., Naigles, L., Schultz, R. T., . . . Tyson, K. (2013). Optimal outcome in individuals with a history of autism. Journal of Child Psychology and Psychiatry, 54(2), 195-205.

Flanagan, J. E., Landa, R., Bhat, A., & Bauman, M. (2012). Head Lag in Infants at Risk for Autism: A Preliminary Study. American Journal of Occupational Ther-apy, 66(5), 577-585.

Flom, R., Deak, G. O., Phill, C. G., & Pick, A. D. (2004). Nine-month-olds' shared visual attention as a function of gesture and object location. Infant Behavior & Development, 27(2), 181-194.

Foulsham, T., Walker, E., & Kingstone, A. (2011). The where, what and when of gaze allocation in the lab and the natural environment. Vision Research, 51(17), 1920-1931.

Freeth, M., Chapman, P., Ropar, D., & Mitchell, P. (2010). Do Gaze Cues in Complex Scenes Capture and Direct the Attention of High Functioning Adolescents with ASD? Evidence from Eye tracking. Journal of Autism and Developmental Disor-ders, 40(5), 534-547. doi:10.1007/s10803-009-0893-2

Frischen, A., & Tipper, S. P. (2004). Orienting attention via observed gaze shift evokes longer term inhibitory effects: Implications for social interactions, atten-tion, and memory. Journal of Experimental Psychology-General, 133(4), 516-533.

Frith, U. (1970). Studies in pattern detection in normal and autistic children. Journal of Experimental Child Psychology, 10(1), 120.

Galpin, A. J., & Underwood, G. (2005). Eye movements during search and detection in comparative visual search. Perception & Psychophysics, 67(8), 1313-1331.

Gangi, D. N., Ibanez, L. V., & Messinger, D. S. (2014). Joint Attention Initiation With and Without Positive Affect: Risk Group Differences and Associations with ASD Symptoms. Journal of Autism and Developmental Disorders, 44(6), 1414-1424.

Gangi, D. N., Messinger, D. S., Martin, E. R., & Cuccaro, M. L. (2016). Dopaminergic variants in siblings at high risk for autism: Associations with initiating joint at-tention. Autism Research, 9(11), 1142-1150.

Gardener, H., Spiegelman, D., & Buka, S. L. (2009). Prenatal risk factors for autism: comprehensive meta-analysis. British Journal of Psychiatry, 195(1), 7-14.

Geschwind, D. H., & Levitt, P. (2007). Autism spectrum disorders: developmental disconnection syndromes. Current Opinion in Neurobiology, 17(1), 103-111.

Geschwind, D. H., & State, M. W. (2015). Gene hunting in autism spectrum disorder: on the path to precision medicine. Lancet Neurology, 14(11), 1109-1120.

Gliga, T., Elsabbagh, M., Hudry, K., Charman, T., Johnson, M. H., & The, B. T. (2012). Gaze Following, Gaze Reading, and Word Learning in Children at Risk for Autism. Child Development, 83(3), 926-938.

81

Goldberg, M. C., Mostow, A. J., Vecera, S. P., Larson, J. C. G., Mostofsky, S. H., Mahone, E. M., & Denckla, M. B. (2008). Evidence for impairments in using static line drawings of eye gaze cues to orient visual-spatial attention in children with high functioning autism. Journal of Autism and Developmental Disorders, 38(8), 1405-1413.

Goldberg, W. A., Jarvis, K. L., Osann, K., Laulhere, T. M., Straub, C., Thomas, E., . . . Spence, M. A. (2005). Brief report: Early social communication behaviors in the younger siblings of children with autism. Journal of Autism and Developmen-tal Disorders, 35(5), 657-664. doi:10.1007/s10803-005-0009-6

Gordon, I., Eilbott, J. A., Feldman, R., Pelphrey, K. A., & Vander Wyk, B. C. (2013). Social, reward, and attention brain networks are involved when online bids for joint attention are met with congruent versus incongruent responses. Social Neu-roscience, 8(6), 544-554.

Gotham, K., Risi, S., Pickles, A., & Lord, C. (2007). The autism diagnostic observa-tion schedule: Revised algorithms for improved diagnostic validity. Journal of Autism and Developmental Disorders, 37(4), 613-627.

Gotts, S. J., Simmons, W. K., Milbury, L. A., Wallace, G. L., Cox, R. W., & Martin, A. (2012). Fractionation of social brain circuits in autism spectrum disorders. Brain, 135, 2711-2725.

Gredebäck, G., Fikke, L., & Melinder, A. (2010). The development of joint visual attention: a longitudinal study of gaze following during interactions with mothers and strangers. Developmental Science, 13(6), 839-848.

Green, J., Pickles, A., Pasco, G., Bedford, R., Wan, M., Elsabbagh, M., . . . Cheung, C. (2017). Randomised trial of a parent‐mediated intervention for infants at high risk for autism: longitudinal outcomes to age 3 years. Journal of Child Psychology and Psychiatry.

Gronborg, T. K., Schendel, D. E., & Parner, E. T. (2013). Recurrence of Autism Spec-trum Disorders in Full- and Half-Siblings and Trends Over Time A Population-Based Cohort Study. Jama Pediatrics, 167(10), 947-953.

Hallmayer, J., Cleveland, S., Torres, A., Phillips, J., Cohen, B., Torigoe, T., . . . Risch, N. (2011). Genetic Heritability and Shared Environmental Factors Among Twin Pairs With Autism. Archives of General Psychiatry, 68(11), 1095-1102.

Happe, F., & Frith, U. (2006). The weak coherence account: Detail-focused cognitive style in autism spectrum disorders. Journal of Autism and Developmental Disor-ders, 36(1), 5-25.

Happé, F., Ronald, A., & Plomin, R. (2006). Time to give up on a single explanation for autism. Nature Neuroscience, 9(10), 1218-1220.

Hayes, S. A., & Watson, S. L. (2013). The Impact of Parenting Stress: A Meta-anal-ysis of Studies Comparing the Experience of Parenting Stress in Parents of Chil-dren With and Without Autism Spectrum Disorder. Journal of Autism and Devel-opmental Disorders, 43(3), 629-642.

Hazlett, H. C., Gu, H. B., Munsell, B. C., Kim, S. H., Styner, M., Wolff, J. J., . . . Network, I. (2017). Early brain development in infants at high risk for autism spectrum disorder. Nature, 542(7641), 348-+.

Hazlett, H. C., Poe, M. D., Gerig, G., Styner, M., Chappell, C., Smith, R. G., . . . Piven, J. (2011). Early Brain Overgrowth in Autism Associated With an Increase in Cor-tical Surface Area Before Age 2 Years. Archives of General Psychiatry, 68(5), 467-476.

Hirotani, M., Stets, M., Striano, T., & Friederici, A. D. (2009). Joint attention helps infants learn new words: event-related potential evidence. NeuroReport, 20(6), 600-605.

82

Hirvikoski, T., & Blomqvist, M. (2015). High self-perceived stress and poor coping in intellectually able adults with autism spectrum disorder. Autism, 19(6), 752-757.

Hirvikoski, T., Mittendorfer-Rutz, E., Boman, M., Larsson, H., Lichtenstein, P., & Bolte, S. (2016). Premature mortality in autism spectrum disorder. British Journal of Psychiatry, 208(3), 232-+.

Holmqvist, K., Nyström, M., Andersson, R., Dewhurst, R., Jarodzka, H., & van de Weijer, J. (2011). Eye tracking: A comprehensive guide to methods and measures. New York: Oxford University Press.

Howard, M. A., Cowell, P. E., Boucher, J., Broks, P., Mayes, A., Farrant, A., & Rob-erts, N. (2000). Convergent neuroanatomical and behavioral evidence of an amygdala hypothesis of autism. NeuroReport, 11(13), 2931-2935.

Hultman, C. M., Sandin, S., Levine, S. Z., Lichtenstein, P., & Reichenberg, A. (2011). Advancing paternal age and risk of autism: new evidence from a population-based study and a meta-analysis of epidemiological studies. Molecular Psychiatry, 16(12), 1203-1212.

Ibanez, L. V., Grantz, C. J., & Messinger, D. S. (2013). The Development of Refer-ential Communication and Autism Symptomatology in High-Risk Infants. In-fancy, 18(5), 687-707.

Idring, S., Rai, D., Dal, H., Dalman, C., Sturm, H., Zander, E., . . . Magnusson, C. (2012). Autism Spectrum Disorders in the Stockholm Youth Cohort: Design, Prevalence and Validity. PLoS ONE, 7(7).

Iossifov, I., O'Roak, B. J., Sanders, S. J., Ronemus, M., Krumm, N., Levy, D., . . . Wigler, M. (2014). The contribution of de novo coding mutations to autism spec-trum disorder. Nature, 515(7526), 216-U136.

Irwin, D. E., & Zelinsky, G. J. (2002). Eye movements and scene perception: Memory for things observed. Perception & Psychophysics, 64(6), 882-895.

Jaarsma, P., & Welin, S. (2012). Autism as a Natural Human Variation: Reflections on the Claims of the Neurodiversity Movement. Health Care Analysis, 20(1), 20-30.

Johnson, M. H., Dziurawiec, S., Ellis, H., & Morton, J. (1991). Newborns preferential tracking of face-like stimuli and its subsequent decline. Cognition, 40(1-2), 1-19.

Johnson, M. H., Griffin, R., Csibra, G., Halit, H., Farroni, T., De Haan, M., . . . Rich-ards, J. (2005). The emergence of the social brain network: Evidence from typical and atypical development. Development and Psychopathology, 17(3), 599-619.

Johnson, M. H., Senju, A., & Tomalski, P. (2015). The two-process theory of face processing: Modifications based on two decades of data from infants and adults. Neuroscience and Biobehavioral Reviews, 50, 169-179.

Jones, E. J. H., Gliga, T., Bedford, R., Charman, T., & Johnson, M. H. (2014). Devel-opmental pathways to autism: A review of prospective studies of infants at risk. Neuroscience and Biobehavioral Reviews, 39, 1-33.

Jones, W., & Klin, A. (2013). Attention to eyes is present but in decline in 2-6-month-old infants later diagnosed with autism. Nature, 504(7480), 427-431.

Kanner, L. (1943). Autistic disturbances of affective contact. Kapp, S. K., Gillespie-Lynch, K., Sherman, L. E., & Hutman, T. (2013). Deficit, Dif-

ference, or Both? Autism and Neurodiversity. Developmental psychology, 49(1), 59-71.

Kasari, C., Sigman, M., Mundy, P., & Yirmiya, N. (1990). Afective sharing in the context of joint attention interactions of normal, autistic, and mentally-retarded children. Journal of Autism and Developmental Disorders, 20(1), 87-100.

Kim, K., & Mundy, P. (2012). Joint attention, social-cognition, and recognition memory in adults. Frontiers in Human Neuroscience, 6.

83

Kim, Y. S., Leventhal, B. L., Koh, Y. J., Fombonne, E., Laska, E., Lim, E. C., . . . Grinker, R. R. (2011). Prevalence of Autism Spectrum Disorders in a Total Pop-ulation Sample. American Journal of Psychiatry, 168(9), 904-912.

Klin, A., Danovitch, J. H., Merz, A. B., & Volkmar, F. R. (2007). Circumscribed in-terests in higher functioning individuals with autism spectrum disorders: An ex-ploratory study. Research and Practice for Persons with Severe Disabilities, 32(2), 89-100.

Klin, A., Jones, W., Schultz, R., Volkmar, F., & Cohen, D. (2002). Visual fixation patterns during viewing of naturalistic social situations as predictors of social competence in individuals with autism. Archives of General Psychiatry, 59(9), 809-816.

Kolevzon, A., Gross, R., & Reichenberg, A. (2007). Prenatal and perinatal risk factors for autism - A review and integration of findings. Archives of Pediatrics & Ado-lescent Medicine, 161(4), 326-333.

Kopp, F., & Lindenberger, U. (2011). Effects of joint attention on long-term memory in 9-month-old infants: an event-related potentials study. Developmental Science, 14(4), 660-672.

Kryzak, L. A., Bauer, S., Jones, E. A., & Sturmey, P. (2013). Increasing responding to others' joint attention directives using circumscribed interests. Journal of Ap-plied Behavior Analysis, 46(3), 674-679.

Kryzak, L. A., & Jones, E. A. (2014). The Effect of Prompts within Embedded Cir-cumscribed Interests to Teach Initiating Joint Attention in Children with Autism Spectrum Disorders. Journal of Developmental and Physical Disabilities.

Kylliäinen, A., & Hietanen, J. K. (2004). Attention orienting by another's gaze direc-tion in children with autism. Journal of Child Psychology and Psychiatry and Allied Disciplines, 45(3), 435-444.

Lai, M. C., Lombardo, M. V., & Baron-Cohen, S. (2014). Autism. Lancet, 383(9920), 896-910.

Laidlaw, K. E. W., Foulsham, T., Kuhn, G., & Kingstone, A. (2011). Potential social interactions are important to social attention. Proceedings of the National Acad-emy of Sciences of the United States of America, 108(14), 5548-5553.

Lam, K. S. L., & Aman, M. G. (2007). The repetitive behavior scale-revised: Inde-pendent validation in individuals with autism spectrum disorders. Journal of Au-tism and Developmental Disorders, 37(5), 855-866.

Landa, R., & Garrett-Mayer, E. (2006). Development in infants with autism spectrum disorders: a prospective study. Journal of Child Psychology and Psychiatry, 47(6), 629-638.

Landa, R. J., Gross, A. L., Stuart, E. A., & Bauman, M. (2012). Latent class analysis of early developmental trajectory in baby siblings of children with autism. Jour-nal of Child Psychology and Psychiatry, 53(9), 986-996.

Landa, R. J., Gross, A. L., Stuart, E. A., & Faherty, A. (2013). Developmental Tra-jectories in Children With and Without Autism Spectrum Disorders: The First 3Years. Child Development, 84(2), 429-442.

Landa, R. J., Holman, K. C., & Garrett-Mayer, E. (2007). Social and communication development in toddlers with early and later diagnosis of autism spectrum disor-ders. Archives of General Psychiatry, 64(7), 853-864.

Lasgaard, M., Nielsen, A., Eriksen, M. E., & Goossens, L. (2010). Loneliness and Social Support in Adolescent Boys with Autism Spectrum Disorders. Journal of Autism and Developmental Disorders, 40(2), 218-226.

Leekam, S., Baron-Cohen, S., Perrett, D., Milders, M., & Brown, S. (1997). Eye-di-rection detection: A dissociation between geometric and joint attention skills in autism. British Journal of Developmental Psychology, 15(9981), 77-95.

84

Leekam, S. R., Hunnisett, E., & Moore, C. (1998). Targets and Cues: Gaze following in Children with Autism. Journal of Child Psychology and Psychiatry, 39(7), 951-962.

Leekam, S. R., López, B., & Moore, C. (2000). Attention and joint attention in pre-school children with autism. Developmental psychology, 36(2), 261-273.

Levy, D., Ronennus, M., Yamrom, B., Lee, Y. H., Leotta, A., Kendall, J., . . . Wigler, M. (2011). Rare De Novo and Transmitted Copy-Number Variation in Autistic Spectrum Disorders. Neuron, 70(5), 886-897.

Lichtenstein, P., Carlstrom, E., Rastam, M., Gillberg, C., & Anckarsater, H. (2010). The Genetics of Autism Spectrum Disorders and Related Neuropsychiatric Dis-orders in Childhood. American Journal of Psychiatry, 167(11), 1357-1363.

Loftus, G. R., & Mackworth, N. H. (1978). Cognitive determinants of fixation loca-tion during picture viewing. Journal of Experimental Psychology-Human Percep-tion and Performance, 4(4), 565-572.

Lord, C. (1995). Follow-up of two-year-olds referred for possible autism. Journal of Child Psychology and Psychiatry and Allied Disciplines, 36(8), 1365-1382.

Lord, C., & Jones, R. M. (2012). Annual Research Review: Re-thinking the classifi-cation of autism spectrum disorders. Journal of Child Psychology and Psychiatry, 53(5), 490-509.

Lord, C., Luyster, R., Gotham, K., & Guthrie, W. (2012a). Autism Diagnostic Obser-vation Scedule - Toddler Module manual. Los Angeles, CA: Western Psycholog-ical Services.

Lord, C., Petkova, E., Hus, V., Gan, W. J., Lu, F. H., Martin, D. M., . . . Risi, S. (2012b). A Multisite Study of the Clinical Diagnosis of Different Autism Spec-trum Disorders. Archives of General Psychiatry, 69(3), 306-313.

Lord, C., Risi, S., Lambrecht, L., Cook, E. H., Jr., Leventhal, B. L., DiLavore, P. C., . . . Rutter, M. (2000). The autism diagnostic observation schedule-generic: a standard measure of social and communication deficits associated with the spec-trum of autism. J Autism Dev Disord, 30(3), 205-223.

Lord, C., Rutter, M., DiLavore, P., Risi, S., Gotham, K., & Bishop, S. (2012c). Autism diagnostic observation schedule, second edition. Torrance, CA: Western Psycho-logical Services.

Lord, C., Rutter, M., & LeCouteur, A. (1994). Autism diagnostic interview-revised - A revised version of a diagnostic interview for caregivers of individuals with pos-sible pervasive developmental disorders. Journal of Autism and Developmental Disorders, 24(5), 659-685.

Lugnegard, T., Hallerback, M. U., & Gillberg, C. (2011). Psychiatric comorbidity in young adults with a clinical diagnosis of Asperger syndrome. Research in Devel-opmental Disabilities, 32(5), 1910-1917.

Lundström, S., Chang, Z., Rastam, M., Gillberg, C., Larsson, H., Anckarsater, H., & Lichtenstein, P. (2012). Autism Spectrum Disorders and Autisticlike Traits Sim-ilar Etiology in the Extreme End and the Normal Variation. Archives of General Psychiatry, 69(1), 46-52.

Macari, S. L., Campbell, D., Gengoux, G. W., Saulnier, C. A., Klin, A. J., & Cha-warska, K. (2012). Predicting developmental status from 12 to 24 months in in-fants at risk for autism spectrum disorder: A preliminary report. Journal of Autism and Developmental Disorders, 42(12), 2636-2647.

Maljaars, J., Noens, I., Jansen, R., Scholte, E., & van Berckelaer-Onnes, I. (2011). Intentional communication in nonverbal and verbal low-functioning children with autism. Journal of Communication Disorders, 44(6), 601-614.

85

Mattila, M. L., Hurtig, T., Haapsamo, H., Jussila, K., Kuusikko-Gauffin, S., Kielinen, M., . . . Moilanen, I. (2010). Comorbid Psychiatric Disorders Associated with Asperger Syndrome/High-functioning Autism: A Community- and Clinic-based Study. Journal of Autism and Developmental Disorders, 40(9), 1080-1093.

McEvilly, M., Wicks, S., & Dalman, C. (2015). Sick Leave and Work Participation Among Parents of Children with Autism Spectrum Disorder in the Stockholm Youth Cohort: A Register Linkage Study in Stockholm, Sweden. Journal of Au-tism and Developmental Disorders, 45(7), 2157-2167.

Mercier, C., Mottron, L., & Belleville, S. (2000). A psychosocial study on restricted interests in high-functioning persons with pervasive developmental disorders. Au-tism, 4(4), 406-425.

Messinger, D. S., Young, G. S., Webb, S. J., Ozonoff, S., Bryson, S. E., Carter, A., . . . Zwaigenbaum, L. (2015). Early sex differences are not autism-specific: A Baby Siblings Research Consortium (BSRC) study. Molecular Autism.

Mohiuddin, S., & Ghaziuddin, M. (2013). Psychopharmacology of autism spectrum disorders: A selective review. Autism, 17(6), 645-654.

Moore, C., & Corkum, V. (1998). Infant gaze following based on eye direction. Brit-ish Journal of Developmental Psychology, 16(4), 495-503.

Morales, M., Mundy, P., Crowson, M. M., Neal, A. R., & Delgado, C. E. F. (2005). Individual differences in infant attention skills, joint attention, and emotion regu-lation behavior. International Journal of Behavioral Development, 29(3), 259-263.

Morales, M., Mundy, P., Delgado, C. E. F., Yale, M., Messinger, D., Neal, R., & Schwartz, H. K. (2000). Responding to joint attention across the 6-through 24-month age period and early language acquisition. Journal of Applied Develop-mental Psychology, 21(3), 283-298.

Morales, M., Mundy, P., & Rojas, J. (1998). Following the direction of gaze and lan-guage development in 6-month-olds. Infant Behavior & Development, 21(2), 373-377.

Mottron, L., Bouvet, L., Bonnel, A., Samson, F., Burack, J. A., Dawson, M., & Hea-ton, P. (2013). Veridical mapping in the development of exceptional autistic abil-ities. Neuroscience and Biobehavioral Reviews, 37(2), 209-228.

Mottron, L., Dawson, M., Soulieres, I., Hubert, B., & Burack, J. (2006). Enhanced perceptual functioning in autism: An update, and eight principles of autistic per-ception. Journal of Autism and Developmental Disorders, 36(1), 27-43.

Mullen, E. M. (1995). Mullen scales of early learning (AGS ed.). Circle Pines, MN: American Guidance Service Inc.

Mundy, P. (2016). Autism and joint attention: development, neuroscience, and clini-cal fundamentals: Guilford Publications.

Mundy, P., Block, J., Delgado, C., Pomares, Y., Van Hecke, A. V., & Parlade, M. V. (2007). Individual differences and the development of joint attention in infancy. Child Development, 78(3), 938-954.

Mundy, P., & Gomes, A. (1998). Individual differences in joint attention skill devel-opment in the second year. Infant Behavior & Development, 21(3), 469-482.

Mundy, P., Hogan, A., & Doehring, P. (1996). A Preliminary Manual for the Abriged Early Social Communication Scale (ESCS).

Mundy, P., & Jarrold, W. (2010). Infant joint attention, neural networks and social cognition. Neural Networks, 23(8-9), 985-997.

Mundy, P., Kim, K., McIntyre, N., Lerro, L., & Jarrold, W. (2016). Brief Report: Joint Attention and Information Processing in Children with Higher Functioning Au-tism Spectrum Disorders. Journal of Autism and Developmental Disorders, 46(7), 2555-2560.

86

Mundy, P., Sigman, M., & Kasari, C. (1994). Joint attention, delevelopmental level, and symptom presentation in autism. Development and Psychopathology, 6(3), 389-401.

Mundy, P., Sigman, M., Ungerer, J., & Sherman, T. (1986). Defining the social defi-cits of autism - The contribution of nonverbal-communication measures. Journal of Child Psychology and Psychiatry and Allied Disciplines, 27(5), 657-669.

Mundy, P., Sullivan, L., & Mastergeorge, A. M. (2009). A Parallel and Distributed-Processing Model of Joint Attention, Social Cognition and Autism. Autism Rese-arch, 2(1), 2-21.

Naber, F. B. A., Bakermans-Kranenburg, M. J., van Ijzendoorn, M. H., Dietz, C., van Daalen, E., Swinkels, S. H. N., . . . van Engeland, H. (2008). Joint attention de-velopment in toddlers with autism. European Child & Adolescent Psychiatry, 17(3), 143-152.

Nadig, A. S., Ozonoff, S., Young, G. S., Rozga, A., Sigman, M., & Rogers, S. J. (2007). A prospective study of response to name in infants at risk for autism. Archives of Pediatrics and Adolescent Medicine, 161(4), 378-383.

Nation, K., & Penny, S. (2008). Sensitivity to eye gaze in autism: Is it normal? Is it automatic? Is it social? Development and Psychopathology, 20(1), 79-97.

Nichols, K. E., Fox, N., & Mundy, P. (2005). Joint attention, self-recognition, and neurocognitive function in toddlers. Infancy, 7(1), 35-51.

Nyström, P., Falck-Ytter, T., & Gredebäck, G. (2016). The TimeStudio Project: An open source scientific workflow system for the behavioral and brain sciences. Be-havior Research Methods, 48(2), 542-552.

Okada, T., Sato, W., Murai, T., Kubota, Y., & Toichi, M. (2004). Eye gaze triggers visuospatial attentional shift in individuals with autism. Psychologia, 46(4), 246-254.

Ortega, F. (2009). The Cerebral Subject and the Challenge of Neurodiversity. Bioso-cieties, 4(4), 425-445.

Ozonoff, S., Goodlin-Jones, B. L., & Solomon, M. (2005). Evidence-based assess-ment of autism spectrum disorders in children and adolescents. Journal of Clini-cal Child and Adolescent Psychology, 34(3), 523-540.

Ozonoff, S., Iosif, A. M., Baguio, F., Cook, I. C., Hill, M. M., Hutman, T., . . . Young, G. S. (2010). A Prospective Study of the Emergence of Early Behavioral Signs of Autism. Journal of the American Academy of Child and Adolescent Psychiatry, 49(3), 256-266.e252.

Ozonoff, S., Macari, S., Young, G. S., Goldring, S., Thompson, M., & Rogers, S. J. (2008). Atypical object exploration at 12 months of age is associated with autism in a prospective sample. Autism, 12(5), 457-472.

Ozonoff, S., Young, G. S., Carter, A., Messinger, D., Yirmiya, N., Zwaigenbaum, L., . . . Stone, W. L. (2011). Recurrence risk for autism spectrum disorders: A baby siblings research consortium study. Pediatrics, 128(3), e488-e495.

Pannasch, S., Helmert, J., Herbold, A., Roth, K., & Walter, H. (2008). Visual fixation durations and saccade amplitudes: Shifting relationship in a variety of conditions. Journal of Eye Movement Research, 2(2), 1-19.

Parlade, M. V., Messinger, D. S., Delgado, C. E. F., Kaiser, M. Y., Van Hecke, A. V., & Mundy, P. C. (2009). Anticipatory smiling: Linking early affective communi-cation and social outcome. Infant Behavior & Development, 32(1), 33-43.

Paul, R., Fuerst, Y., Ramsay, G., Chawarska, K., & Klin, A. (2011). Out of the mouths of babes: vocal production in infant siblings of children with ASD. Journal of Child Psychology and Psychiatry, 52(5), 588-598.

Pellicano, E., & Burr, D. (2012). When the world becomes 'too real': a Bayesian ex-planation of autistic perception. Trends in Cognitive Sciences, 16(10), 504-510.

87

Perra, O., & Gattis, M. (2010). The control of social attention from 1 to 4 months. British Journal of Developmental Psychology, 28(4), 891-908.

Pickard, K. E., & Ingersoll, B. R. (2015). Brief Report: High and Low Level Initia-tions of Joint Attention, and Response to Joint Attention: Differential Relation-ships with Language and Imitation. Journal of Autism and Developmental Disor-ders, 45(1), 262-268.

Presmanes, A. G., Walden, T. A., Stone, W. L., & Yoder, P. J. (2007). Effects of Different Attentional Cues on Responding to Joint Attention in Younger Siblings of Children with Autism Spectrum Disorders. Journal of Autism & Developmen-tal Disorders, 37(1), 133-144.

Rane, P., Cochran, D., Hodge, S. M., Haselgrove, C., Kennedy, D. N., & Frazier, J. A. (2015). Connectivity in Autism: A Review of MRI Connectivity Studies. Har-vard Review of Psychiatry, 23(4), 223-244.

Redcay, E., & Courchesne, E. (2005). When is the brain enlarged in autism? A meta-analysis of all brain size reports. Biological Psychiatry, 58(1), 1-9.

Redcay, E., Dodell-Feder, D., Pearrow, M. J., Mavros, P. L., Kleiner, M., Gabrieli, J. D. E., & Saxe, R. (2010). Live face-to-face interaction during fMRI: A new tool for social cognitive neuroscience. NeuroImage, 50(4), 1639-1647.

Redcay, E., Kleiner, M., & Saxe, R. (2012). Look at this: the neural correlates of ini-tiating and responding to bids for joint attention. Frontiers in Human Neurosci-ence, 6.

Ristic, J., Mottron, L., Friesen, C. K., Iarocci, G., Burack, J. A., & Kingstone, A. (2005). Eyes are special but not for everyone: the case of autism. Brain research. Cognitive brain research, 24(3), 715-718.

Roberts, E. M., English, P. B., Grether, J. K., Windharn, G. C., Somberg, L., & Wolff, C. (2007). Maternal residence near agricultural pesticide applications and autism spectrum disorders among children in the California Central Valley. Environmen-tal Health Perspectives, 115(10), 1482-1489.

Rogers, S. J., Vismara, L., Wagner, A. L., McCormick, C., Young, G., & Ozonoff, S. (2014). Autism Treatment in the First Year of Life: A Pilot Study of Infant Start, a Parent-Implemented Intervention for Symptomatic Infants. Journal of Autism and Developmental Disorders, 44(12), 2981-2995.

Ronald, A., & Hoekstra, R. A. (2011). Autism Spectrum Disorders and Autistic Traits: A Decade of New Twin Studies. American Journal of Medical Genetics Part B-Neuropsychiatric Genetics, 156B(3), 255-274.

Rozga, A., Hutman, T., Young, G. S., Rogers, S. J., Ozonoff, S., Dapretto, M., & Sigman, M. (2011). Behavioral profiles of affected and unaffected siblings of children with autism: Contribution of measures of mother-infant interaction and nonverbal communication. Journal of Autism and Developmental Disorders, 41(3), 287-301.

Saemundsen, E., Magnusson, P., Georgsdottir, I., Egilsson, E., & Rafnsson, V. (2013). Prevalence of autism spectrum disorders in an Icelandic birth cohort. Bmj Open, 3(6).

Salomone, E., Charman, T., McConachie, H., & Warreyn, P. (2016). Child's verbal ability and gender are associated with age at diagnosis in a sample of young chil-dren with ASD in Europe. Child Care Health and Development, 42(1), 141-145.

Sandin, S., Hultman, C. M., Kolevzon, A., Gross, R., MacCabe, J. H., & Reichenberg, A. (2012). Advancing Maternal Age Is Associated With Increasing Risk for Au-tism: A Review and Meta-Analysis. Journal of the American Academy of Child and Adolescent Psychiatry, 51(5), 477-486.

88

Sasson, N. J., Elison, J. T., Turner-Brown, L. M., Dichter, G. S., & Bodfish, J. W. (2011). Brief report: Circumscribed attention in young children with autism. Jour-nal of Autism and Developmental Disorders, 41(2), 242-247.

Sasson, N. J., & Touchstone, E. W. (2013). Visual Attention to Competing Social and Object Images by Preschool Children with Autism Spectrum Disorder. Journal of Autism and Developmental Disorders, 1-9.

Sasson, N. J., Turner-Brown, L. M., Holtzclaw, T. N., Lam, K. S. L., & Bodfish, J. W. (2008). Children with autism demonstrate circumscribed attention during pas-sive viewing of complex social and nonsocial picture arrays. Autism Research, 1(1), 31-42.

Scaife, M., & Bruner, J. S. (1975). Capacity for joint visual attention in infant. Nature, 253(5489), 265-266. doi:10.1038/253265a0

Schietecatte, I., Roeyers, H., & Warreyn, P. (2012). Exploring the nature of joint at-tention impairments in young children with autism spectrum disorder: Associated social and cognitive skills. Journal of Autism and Developmental Disorders, 42(1), 1-12.

Schilbach, L. (2010). A second-person approach to other minds. Nature Reviews Neu-roscience, 11(6), 449.

Schilbach, L., Wilms, M., Eickhoff, S. B., Romanzetti, S., Tepest, R., Bente, G., . . . Vogeley, K. (2010). Minds made for sharing: Initiating joint attention recruits reward-related neurocircuitry. Journal of Cognitive Neuroscience, 22(12), 2702-2715.

Schroeder, J. H., Cappadocia, M. C., Bebko, J. M., Pepler, D. J., & Weiss, J. A. (2014). Shedding Light on a Pervasive Problem: A Review of Research on Bullying Ex-periences Among Children with Autism Spectrum Disorders. Journal of Autism and Developmental Disorders, 44(7), 1520-1534.

Sebat, J., Lakshmi, B., Malhotra, D., Troge, J., Lese-Martin, C., Walsh, T., . . . Wigler, M. (2007). Strong association of de novo copy number mutations with autism. Science, 316(5823), 445-449.

Senju, A., Csibra, G., & Johnson, M. H. (2008). Understanding the referential nature of looking: Infants' preference for object-directed gaze. Cognition, 108(2), 303-319.

Senju, A., Johnson, M. H., & Csibra, G. (2006). The development and neural basis of referential gaze perception. Social Neuroscience, 1(3-4), 220-234.

Shah, A., & Frith, U. (1983). An islet of ability in autistic-children - a research note. Journal of Child Psychology and Psychiatry and Allied Disciplines, 24(4), 613-620.

Shimada, S., & Hiraki, K. (2006). Infant's brain responses to live and televised action. NeuroImage, 32(2), 930-939.

Sigman, M., & Ruskin, E. (1999). Continuity and change in the social competence of children with autism, Down syndrome, and developmental delays. Monographs of the Society for Research in Child Development, 64(1).

Simonoff, E., Pickles, A., Charman, T., Chandler, S., Loucas, T., & Baird, G. (2008). Psychiatric disorders in children with autism spectrum disorders: Prevalence, comorbidity, and associated factors in a population-derived sample. Journal of the American Academy of Child and Adolescent Psychiatry, 47(8), 921-929.

Sodian, B., & Kristen-Antonow, S. (2015). Declarative Joint Attention as a Founda-tion of Theory of Mind. Developmental psychology, 51(9), 1190-1200.

Sorce, J. F., Emde, R. N., Campos, J., & Klinnert, M. D. (1985). Maternal emotional signaling - its effect on the visual-cliff behavior of 1-year-olds. Developmental psychology, 21(1), 195-200.

89

Sparrow, S., Balla, D., & Cicchetti, D. (1984). Vineland Adaptive Behavior Scales. Circle Pines, MN: American Guidance Service.

Spezio, M. L., Adolphs, R., Hurley, R. S. E., & Piven, J. (2007). Abnormal use of facial information in high-functioning autism. Journal of Autism and Develop-mental Disorders, 37(5), 929-939.

Stenberg, G., & Hagekull, B. (1997). Social referencing and mood modification in 1-year-olds. Infant Behavior & Development, 20(2), 209-217.

Stone, W. L., Coonrod, E. E., & Ousley, O. Y. (2000). Brief report: Screening tool for autism in two-year-olds (STAT): Development and preliminary data. Journal of Autism and Developmental Disorders, 30(6), 607-612.

Stone, W. L., Lee, E. B., Ashford, L., Brissie, J., Hepburn, S. L., Coonrod, E. E., & Weiss, B. H. (1999). Can autism be diagnosed accurately in children under 3 years? Journal of Child Psychology and Psychiatry and Allied Disciplines, 40(2), 219-226.

Striano, T., & Rochat, P. (2000). Emergence of Selective Social Referencing in In-fancy. Infancy, 1(2), 253-264.

Sucksmith, E., Roth, I., & Hoekstra, R. A. (2011). Autistic Traits Below the Clinical Threshold: Re-examining the Broader Autism Phenotype in the 21st Century. Neuropsychology Review, 21(4), 360-389.

Sullivan, M., Finelli, J., Marvin, A., Garrett-Mayer, E., Bauman, M., & Landa, R. (2007). Response to joint attention in toddlers at risk for autism spectrum disor-der: A prospective study. Journal of Autism and Developmental Disorders, 37(1), 37-48.

Swanson, M. R., & Siller, M. (2013). Patterns of gaze behavior during an eye tracking measure of joint attention in typically developing children and children with au-tism spectrum disorder. Research in Autism Spectrum Disorders, 7(9), 1087-1096.

Swettenham, J., Condie, S., Campbell, R., Milne, E., & Coleman, M. (2003). Does the perception of moving eyes trigger reflexive visual orienting in autism? Philo-sophical Transactions of the Royal Society B-Biological Sciences, 358(1430), 325-334.

Tomasello, M., Carpenter, M., Call, J., Behne, T., & Moll, H. (2005). Understanding and sharing intentions: The origins of cultural cognition. Behavioral and Brain Sciences, 28(05), 675-691.

Tomasello, M., & Farrar, M. J. (1986). Joint attention and early language. Child De-velopment, 57(6), 1454-1463.

Tomasello, M., Moore, C., & Dunham, P. (1995). Joint attention as social cognition. Joint attention: Its origins and role in development, 103-130.

Turner-Brown, L. M., Lam, K. S. L., Holtzclaw, T. N., Dichter, G. S., & Bodfish, J. W. (2011). Phenomenology and measurement of circumscribed interests in au-tism spectrum disorders. Autism, 15(4), 437-456.

Wakefield, A. J., Murch, S. H., Anthony, A., Linnell, J., Casson, D. M., Malik, M., . . . Walker-Smith, J. A. (1998). Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children (Retracted article. See vol 375, pg 445, 2010). Lancet, 351(9103), 637-641.

Van Boxtel, J. J. A., & Lu, H. (2013). Impaired global, and compensatory local, bio-logical motion processing in people with high levels of autistic traits. Frontiers in Psychiatry, 4(APR).

Van Hecke, A. V., Mundy, P., Block, J. J., Delgado, C. E. F., Parlade, M. V., Pomares, Y. B., & Hobson, J. A. (2012). Infant responding to joint attention, executive pro-cesses, and self-regulation in preschool children. Infant Behavior & Development, 35(2), 303-311.

90

Van Hecke, A. V., Mundy, P. C., Acra, C. F., Block, J. J., Delgado, C. E. F., Parlade, M. V., . . . Pomares, Y. B. (2007). Infant joint attention, temperament, and social competence in preschool children. Child Development, 78(1), 53-69.

Warreyn, P., Roeyers, H., & De Groote, I. (2005). Early social communicative behav-iors of preschoolers with autism spectrum disorder during interaction with their mothers. Autism, 9(4), 342-361.

Wass, S. V., Jones, E. J. H., Gliga, T., Smith, T. J., Charman, T., Johnson, M. H., & team, B. (2015). Shorter spontaneous fixation durations in infants with later emerging autism. Scientific Reports, 5.

Webster, S., & Potter, D. D. (2008). Brief report: Eye direction detection improves with development in autism. Journal of Autism and Developmental Disorders, 38(6), 1184-1186.

Wechsler, D. (2014). Wechsler Intelligence Scale for Children - Fifth Edition. San Antonio, TX: NSC Pearson.

Venezia, M., Messinger, D. S., Thorp, D., & Mundy, P. (2004). The development of anticipatory smiling. Infancy, 6(3), 397-406.

Wetherby, A. M., & Prizant, B. M. (2002). CSBS DP Manual: Communication and Symbolic Behavior Scales Developmental Profile. Baltimore, MD: Paul H Brookes Publishing Co.

Willfors, C., Carlsson, T., Anderlid, B. M., Nordgren, A., Kostrzewa, E., Berggren, S., . . . Bolte, S. (2017). Medical history of discordant twins and environmental etiologies of autism. Translational Psychiatry, 7.

Wingate, M., Kirby, R. S., Pettygrove, S., Cunniff, C., Schulz, E., Ghosh, T., . . . Autism Dev Disabilities, M. (2014). Prevalence of Autism Spectrum Disorder Among Children Aged 8 Years - Autism and Developmental Disabilities Moni-toring Network, 11 Sites, United States, 2010. Mmwr Surveillance Summaries, 63(2).

Volk, H. E., Lurmann, F., Penfold, B., Hertz-Picciotto, I., & McConnell, R. (2013). Traffic-Related Air Pollution, Particulate Matter, and Autism. Jama Psychiatry, 70(1), 71-77.

von dem Hagen, E. A. H., Stoyanova, R. S., Baron-Cohen, S., & Calder, A. J. (2013). Reduced functional connectivity within and between 'social' resting state net-works in autism spectrum conditions. Social Cognitive and Affective Neurosci-ence, 8(6), 694-701.

Wu, Z., Pan, J. T., Su, Y. J., & Gros-Louis, J. (2013). How joint attention relates to cooperation in 1- and 2-year-olds. International Journal of Behavioral Develop-ment, 37(6), 542-548.

Yirmiya, N., Gamliel, I., Pilowsky, T., Feldman, R., Baron-Cohen, S., & Sigman, M. (2006). The development of siblings of children with autism at 4 and 14 months: Social engagement, communication, and cognition. Journal of Child Psychology and Psychiatry and Allied Disciplines, 47(5), 511-523.

Yoder, P., Stone, W. L., Walden, T., & Malesa, E. (2009). Predicting social impair-ment and ASD diagnosis in younger siblings of children with autism spectrum disorder. Journal of Autism and Developmental Disorders, 39(10), 1381-1391.

Young, G. S., Rogers, S. J., Hutman, T., Rozga, A., Sigman, M., & Ozonoff, S. (2011). Imitation from 12 to 24 months in autism and typical development: A longitudinal rasch analysis. Developmental psychology, 47(6), 1565-1578.

Yu, C., & Smith, L. B. (2013). Joint Attention without Gaze Following: Human In-fants and Their Parents Coordinate Visual Attention to Objects through Eye-Hand Coordination. PLoS ONE, 8(11).

91

Zwaigenbaum, L., Bryson, S., Rogers, T., Roberts, W., Brian, J., & Szatmari, P. (2005). Behavioral manifestations of autism in the first year of life. International Journal of Developmental Neuroscience, 23(2-3).

Zwaigenbaum, L., Thurm, A., Stone, W., Baranek, G., Bryson, S., Iverson, J., . . . Sigman, M. (2007). Studying the emergence of autism spectrum disorders in high-risk infants: Methodological and practical issues. Journal of Autism and Devel-opmental Disorders, 37(3), 466-480.

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Social Sciences 145

Editor: The Dean of the Faculty of Social Sciences

A doctoral dissertation from the Faculty of Social Sciences,Uppsala University, is usually a summary of a number ofpapers. A few copies of the complete dissertation are keptat major Swedish research libraries, while the summaryalone is distributed internationally through the series DigitalComprehensive Summaries of Uppsala Dissertations from theFaculty of Social Sciences. (Prior to January, 2005, the serieswas published under the title “Comprehensive Summaries ofUppsala Dissertations from the Faculty of Social Sciences”.)

Distribution: publications.uu.seurn:nbn:se:uu:diva-327117

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2017