Social Timing in Preschoolers with Autism Spectrum Disorder

Master's Thesis, 2017

172 Pages


Table of Contents


Table of Contents

List of Tables

List of Figures


1 General introduction
1.1 Context
1.2 Problem statement
1.3 Aim and objectives
1.4 Scope
1.5 Research questions
1.6 Significance of this study
1.7 Definitions of terms and key concepts
1.8 Thesis structure

2 Social timing in young children with and without ASD
2.1 Introduction
2.2 Background to social timing
2.2.1 The role of timing in communication and development
2.2.2 Defining social timing
2.2.3 Related but different uses of 'synchrony'
2.3 Social timing from a developmental perspective
2.3.1 Which developments lead up to synchrony?
2.3.2 What is the significance of synchrony for later development?
2.4 Social timing in ASD
2.4.1 Social and communication deficits in children with ASD
2.4.2 Timing explanations of ASD
2.4.3 Timing in ASD
2.4.4 Social timing in ASD
2.4.5 Intrapersonal timing in ASD
2.4.6 Interpersonal timing in ASD
2.5 Summary and conclusion

3 Studying social timing
3.1 Introduction
3.2 Introduction to social timing methods
3.2.1 Fully automated methods
3.2.2 Non-computational methods
3.3 Studying social timing in ASD
3.3.1 Studies using automated analysis
3.3.2 Studies using micro-analysis
3.3.3 Studies using global analysis
3.4 Monadic Phases
3.4.1 Theoretical background
3.4.2 Participants of MP studies
3.4.3 Experimental set-up
3.4.4 Coding of Monadic Phases
3.4.5 Statistical analysis of Monadic Phases
3.4.6 Validity, inter-rater agreement and reliability
3.5 Summary and conclusion

4 Methods
4.1 Introduction
4.2 Choice and justification of methods
4.2.1 Population and sample
4.2.2 Data analytic approach
4.2.3 Adaptation of research instrument
4.3 Adaptation of Monadic Phases
4.3.1 Draft
4.3.2 Draft 2: Monadic Phases 2
4.3.3 Conclusion on Monadic Phase adaptation
4.4 Method
4.4.1 Participants
4.4.2 Setting and apparatus
4.4.3 Ethical approval
4.4.4 Procedures
4.4.5 Data analysis plan
4.5 Summary

5 Results
5.1 Introduction
5.2 Preliminary analyses
5.2.1 Suitability
5.2.2 Stationarity
5.3 Cyclicity and synchrony
5.3.1 Cyclicity
5.3.2 Coherence
5.3.3 Synchrony
5.4 Summary

6 Discussion
6.1 Introduction
6.2 Answering the research questions
6.2.1 Cyclicity
6.2.2 Synchrony
6.3 Unexpected findings
6.3.1 Music enhances social timing in ASD
6.3.2 No change over time
6.4 Evaluating the chosen approach
6.4.1 Terminology
6.4.2 Study design
6.4.3 Data analysis
6.4.4 Adaptation of coding scheme
6.4.5 Comparability with other studies

7 Conclusion
7.1 Introduction
7.2 Key findings
7.3 Contributions and limitations
7.4 Beyond this study


Appendix A: What or who drives timing in interaction?

Appendix B: Monadic Phases

Appendix C: Time-series analysis - Theory and practice

Appendix D: Musical Interaction Therapy

Appendix E: Coding protocol


Social timing plays a concurrent and long-term role in social interactions. Cyclicity (a person's coordination of speech, body movements etc.) and synchrony (the coordination between individuals) are especially important. Synchronous interactions in childhood affect later developments, such as language development and emotion regulation.

In ASD, timing and social timing are abnormal, which may adversely affect or even cause impairments. Evidence of interactional synchrony skills in ASD is sparse, therefore I sought to investigate cyclicity and synchrony skills in ASD.

Video-recordings of interaction with and without music between children with ASD and a caregiver (N = 14; 2 to 8 years) were analysed using an adaptation of the well-established Monadic Phase coding scheme. Time-series analysis enabled quantification of cyclicity, level of synchrony (coherence) and significant synchrony.

Cyclicity was present in most interactions (76-90%). Coherence scores ranged from .08-.39. Synchrony was present in 19% of time-series without and 60% of time-series with music. Music significantly enhanced presence of synchrony (p < .000) and indicated a trend for enhanced cyclicity (p = .058) and coherence (p = .063). No change over time was observed.

Therefore, preschoolers with ASD engaged in rhythmic social timing but consistency was low and diminished compared to neurotypical infants. Music enhanced social timing considerably. No change over time was likely due to fluctuations in children's willingness to engage. Findings are limited by the lack of interrater reliability and control group.

The aim of the thesis, to contribute to social timing evidence in ASD was achieved. The method successfully quantified social timing parameters, compared data to previous studies and showed that music enhanced social timing performance. Recommendations for further study include replication with a larger group, more time points, and control groups. This method could be applied to other settings to investigate the concurrent effect of music on social timing.

List of Tables

Table 2.1 Interactional synchrony definitions

Table 2.2 Different uses of the term 'synchrony'

Table 2.3 Overview over ASD deficits

Table 2.4 Temporo-spatial processing problems

Table 2.7 Isolated intrapersonal timing studies in ASD

Table 2.8 Non-isolated intrapersonal timing studies in ASD

Table 2.9 Vocal synchrony studies in ASD

Table 2.10 Motoric synchronisation studies in ASD

Table 2.11 Interactive flow studies in ASD

Table 2.12 Biobehavioural synchrony study in ASD

Table 2.13 Interactional synchrony studies in ASD

Table 3.1 Overview over social timing methods

Table 3.2 Evaluation of social timing studies in ASD

Table 4.1 Overview over draft changes: Monadic Phases to Mini MP2.0

Table 4.2 Behavioural categories and codes

Table 4.3 Phases

Table 4.4 MP2.0: Behavioural categories and codes

Table 4.5 MP2.0: Phases

Table 4.6 Demographic data of participants

Table 4.7 Synchrony parameters

Table 5.1 Summary of suitability scores

Table 5.2 Summary stationarity results

Table 5.3 Summary cyclicity

Table 5.4 Summary synchrony coherence

Table 5.5 Summary two-way repeated measures ANOVA

Table 5.6 Summary synchrony

Table 6.1 Cyclicity comparison data

Table 6.2 Synchrony comparison data

List of Figures

Figure 1.1 Thesis structure

Figure 2.1 Chapter overview

Figure 3.1 Chapter overview

Figure 4.1 Chapter overview

Figure 4.2 Time-series analysis flowchart

Figure 5.1 Chapter overview

Figure 5.2 Agreement between raters of suitability scores

Figure 5.3 Suitability score averages of Music vs Interaction over time

Figure 5.4 Suitability score averages of Music vs Interaction

Figure 5.5 Percentage of cyclic time-series on average by dyadic partner and condition

Figure 5.6 Synchrony coherence averages over time Music vs Interaction

Figure 5.7 Mean coherence values for time (Time1, Time2 and Time3) and condition (Music and Interaction).

Figure 5.8 Number of interactions with significant synchrony comparing Music and Interaction

Figure 5.9 Number of interactions with synchrony during Music vs Interaction per dyad

Figure 5.10 Mean time-lag to synchrony in Music vs Interaction

Figure 6.1 Chapter overview

Figure 7.1 Chapter overview


Abbildung in dieser Leseprobe nicht enthalten

1 General introduction

1.1 Context

Timing plays a vital role in communication. It allows perception and coordination of communication tools such as speech, facial expressions, and body language. For example, neurotypical individuals naturally sense when it is their turn to speak, and use pauses to emphasise or alter the meaning of what they are trying to convey. When timing is optimal and communication flows evenly, people remain largely unaware of its importance. On the contrary, erratic timing is noticed immediately. For example, when the audio or video transmission during a Skype-conversation lags behind, one can still carry on with his or her conversation but it becomes harder to read social cues and communication becomes tedious.

Scholars refer to this type of timing as 'social timing'. Its role is to structure and organise the timing of verbal and non-verbal communication tools. Social timing is further split into the temporal organisation of our own body language and speech ('intrapersonal') and that between self and other ('interpersonal').

Social timing in parent-infant interaction1 plays a key role for long-term development of communication and social skills (e.g. Feldman, 2007b; Feldman & Eidelman, 2004). The pioneers of developmental social timing studies focused on describing rhythmicity of pre-verbal communication in the 1970s. A decade later, advances in statistical methods allowed examination of intra- and interpersonal timing, which can be cyclical and synchronous respectively.

Evidence of emerging social timing is already present in neonates and young infants. Cyclical and synchronous social timing patterns emerge long before a child develops language, and are stimulated during parent-infant interactions and 'social games'. Further evidence for the importance of early social timing skills stems from studies on children or parents at risk. For example, preterm infants or mothers suffering post-partum depression show lower levels of interactional synchrony (Granat, Gadassi, Gilboa-Schechtman, & Feldman, 2017). In sum, social timing plays a big part in the development of mature communication and social interaction abilities.

Poor social timing in Autism Spectrum Disorder (ASD) may be responsible for communication and social interaction difficulties (Allman, 2011; Amos, 2013). Plenty of timing research found disruptions in people with ASD across several domains, including cognition, genetics, circadian rhythm and motor skills. This is supported by self-reports from people with ASD and their caregivers who attest to experiencing temporal events at a different pace than typically developing individuals.

Much less evidence exists for social timing difficulties in ASD, even though it seems like a promising approach to forming theories and optimising interventions. Existing studies are few in number and suffer from methodological limitations. Several scholars proposed hypotheses that attribute symptoms in ASD to timing and social timing problems, yet knowledge gaps still exist. In short, it is plausible that timing problems in ASD include social timing difficulties and that they may cause subsequent communication problems.

1.2 Problem statement

Despite decades of developmental social timing research, few studies have examined social timing in ASD, and even fewer have investigated cyclicity and synchrony parameters. Furthermore, existing studies suffer from limitations that make comparisons across studies and populations difficult, such as the use of varying approaches and the lack of establishing discrete social timing parameters. Finally, some of the evidence is based on case and at risk studies, thus limiting generalisation. In sum, more knowledge is needed about cyclicity and synchrony abilities in ASD, along with a method that overcomes previous drawbacks.

The main difference between previous social timing studies in typical development (TD), at risk children, and children with ASD is their age: infants in those studies (e.g. Feldman, 2006; Yirmiya et al., 2006) were 3 or 4 months old, whereas ASD is only diagnosed at around 2 or 3 years of age. While pre-verbal infants and children with severe forms of ASD both rely on pre-verbal communication, what they can do physically differs in many ways. For this reason, previously established methods cannot be applied directly to the study of social timing in young children with ASD. As a solution, a pre-existing method could be adapted so that desired features are preserved and comparisons between studies are enabled.

Undertaking this study is worthwhile for two reasons: firstly, it produces a method which can be subjected to further testing if promising, or discarded if it proves flawed. Secondly, any knowledge about social timing in ASD gained from this study may inform future research, theories, and clinical practice. If social timing is shown to be disrupted, and enhancing factors are uncovered, then strategies to foster social timing skills in early childhood can be developed or existing ones refined.

1.3 Aim and objectives

The aim is to learn more about social timing in ASD and contribute knowledge about cyclicity and synchrony performance in preschoolers with ASD. In order to achieve this aim, a method is required that addresses previous methodological shortcomings. The following objectives are defined to achieve this aim:

- Gain an understanding of social timing in early development in individuals with and without ASD.
- Identify a methodology that allows studying social timing in young children with ASD.
- Adapt the chosen method to the sample of this study and describe the method used.
- Present outcomes of statistical analysis of cyclicity and synchrony performance in ASD.
- Analyse and interpret findings in the context of previous studies.
- Provide recommendations based on the findings and critically evaluate study.

1.4 Scope

The emphasis of this study is on social timing during parent-child interaction in a real-life environment in preschoolers with ASD. As an additional feature, the children and adults in this sample received Musical Intervention Therapy (MIT; Wimpory, Chadwick & Nash, 1995; see Appendix D) in North Wales in the United Kingdom (UK). This allows comparison of interaction with music (hereafter referred to as 'Music') and without music ('Interaction') within an MIT setting.

This thesis is a pilot study because an existing method is adapted and used for the first time, which means findings are tentative until replication can confirm its validity and reliability. Generalisability only extends as far as the sample allows, which are mostly pre-verbal children with ASD and their caregiver (usually the mother). Thus, findings cannot be extended to make conclusions about fully verbal children with ASD, nor children who are much younger or older. Findings can only inform about interactions with a familiar adult caregiver, rather than peer interactions or strangers. This study is confined to a naturalistic setting as the data comes from video tapes recorded to support intervention. This means that variables were not as carefully controlled as expected in a laboratory setting. While these limitations may affect study outcomes, any insights gained and the resulting method are expected to be useful for further research and clinical practice.

1.5 Research questions

The following questions were addressed in this study:

- Do children with ASD show cyclicity during Music and Interaction?
- Is cyclicity more consistently present in Music compared to Interaction?
- Is Music associated with an increased level of synchrony (or 'coherence') compared to Interaction?
- Is there more significant synchrony in Music compared to Interaction?

1.6 Significance of this study

I hope that this study contributes to ASD research in two ways: firstly, I intend to advance the knowledge about social timing in young children with ASD. As existing evidence is limited, my goal is to add insights that follow rigorous scientific standards as far as possible. In particular, I aim to add knowledge about cyclicity and synchrony during social interaction, which is under-researched in ASD.

Secondly, the method designed to achieve this aim is intended to produce output that is comparable across studies and easily replicable. Furthermore, it is designed to enable studying social timing of other populations, such as older typically developing, developmentally delayed and at risk groups.

1.7 Definitions of terms and key concepts

ASD is marked by impairments in social interaction and communication as well as restricted, repetitive and stereotyped patterns of behaviour, interests and activities. Symptoms must be present in early development, and cause clinically significant impairment (American Psychiatric Association, 2000).

Social timing consists of intrapersonal and interpersonal timing. Intrapersonal timing refers to the individuals own rhythmicity2 during interaction, in other words how one temporally coordinates communication tools such as speech, gesture, facial expressions, and body movement. This rhythm has been referred to as cyclicity, the term I adopt here.

Interpersonal timing relates to the temporal coordination between individuals (Jaffe et al., 2001). One way in which such coordination may manifest is called (interactional) synchrony. In general, synchrony describes the relationship of events happening at the same time. The study of synchrony has been applied to a range of scientific fields, such as physics, mathematics, populations growth, weather change or genetics (De Jaegher, 2006; Feldman, 2007b). Of interest for this study is interactional synchrony, or the temporally organised flow of social interaction between people. Research has looked at synchrony in dyadic, triadic, and multi person interactions, at early infancy, childhood, adolescence and adulthood. In this thesis, I focus on parent-child dyads during pre-verbal playful interactions.

1.8 Thesis structure

This thesis consists of six further chapters including the literature review, findings and conclusion (see Figure 1.1 below).

Chapter 2 and 3 review the literature. Chapter 2 establishes the broader context by outlining the connection between timing, social timing, early development and ASD. It begins by explaining the role of timing in communication and development and defining social timing terms. Next follows an overview of how social timing unfolds in the first year of life in typical development. Finally, social timing in ASD is explored. Social and communication difficulties in ASD are presented, followed by a brief overview over timing hypotheses pertaining to ASD. Finally, existing evidence on timing and social timing abilities in ASD are addressed.

Chapter 3 investigates available methods to study social timing. I give an overview over existing methods and critically evaluate their merits and drawbacks. The Monadic Phase method stands out from other available ones and is presented in more detail.

My intended contribution is contained in Chapters 4, 5 and 6. Chapter 4 presents this study's methodology including a discussion of the choices that were made and how the research instrument was adapted. This is followed by a description of the chosen method to study social timing in ASD complete with procedures and data analysis plan.

Chapter 5 contains the results of the study. It begins with preliminary analyses and is followed by a presentation of the results that address cyclicity and synchrony performance.

Chapter 6 analyses and interprets research findings and sets them into the context of previous studies. I first answer the research questions, then address unexpected findings and finally evaluate the chosen approach.

Finally, Chapter 7, the conclusion, provides a brief summary of key findings, then addresses this study's contributions and limitations and suggests further research agendas and gives recommendations for both research and clinical practice.

Abbildung in dieser Leseprobe nicht enthalten

Figure 1.1 Thesis structure

2 Social timing in young children with and without ASD

2.1 Introduction

Chapter 1 established the context of this thesis; social timing is important for the development of social interaction and communication skills including language. Children with an ASD have difficulties with timing, yet relatively few studies have investigated social timing in ASD. One major problem was the lack of an appropriate method that enables replication and comparison of results across studies. Thus, a need for such a method and further study of social timing in ASD was identified.

The objective of Chapter 2 is to give the reader an understanding of social timing in young children with and without ASD. In order to do that, background information is provided on the connections between timing, development, social and communication skills.

This chapter is split into three sections (see Figure 2.1 below). The first provides an introduction to social timing. It explains the role of timing in communication and development, through cognition, movement, and the central nervous system. This is followed by definitions of the terms social timing, cyclicity and synchrony. Section two looks at social timing through a developmental perspective; how it develops in the first year of life and how it affects development beyond that time. The third section looks at social timing in ASD. It begins by summarising what social, communication and language deficits prevail in ASD. This is followed by hypotheses that attribute social and communication deficits in ASD to underlying timing deficits. Next, evidence of timing abilities in ASD is presented, for which a multi-disciplinary approach was adopted. Finally, social timing abilities in ASD are considered. A summary and conclusion is provided at the end of the chapter.

Abbildung in dieser Leseprobe nicht enthalten

Figure 2.1 Chapter overview

2.2 Background to social timing

Time plays an omnipresent role in each of our lives. Our innate sense of timing allows us to judge fairly accurately when it is time for tea, for example. Time perception lets us ‘feel’ time, in the sense of whether something goes by quickly or seems to be never-ending.

This ability stems from internal 'clocks' that help us keep track of time. Time’s subconscious nature can be revealed by studies of temporal processing, binding, or motor timing. Examples include measuring how quickly our brain can recalibrate its movements when encountering an unexpected obstacle, how efficiently audio-visual stimuli are integrated, and at what threshold a desynchronisation is noticed. The importance of timing in social interaction is discussed below, followed by research findings about timing in cognition, motor skills and in the brain with an emphasis on early childhood development.

2.2.1 The role of timing in communication and development

Social interactions are fast and subtle, requiring quick and complex integration of multiple sensory inputs. For example, a parent’s interaction approach may include leaning over his child, smiling, vocalising, and stroking the child’s head. This requires the infant’s nervous system to combine incoming visual, auditory, and tactile information to derive full meaning from this interaction event. Unsuccessful or delayed processing of just one modality could eliminate perceived simultaneity among these inputs, thereby disrupting the event’s significance (McPartland et al., 2004).

Appropriate timing in communication is important for conveying mood (Natale, 1978), empathy (Welkowitz & Feldstein, 1970), theory of mind abilities (Blakemore et al., 2003) and affiliation (Hove & Risen, 2009; Ramseyer & Tschacher, 2011). The connection between social timing and neuro-cognition was first made over three decades ago; Lester and colleagues (1985) suspected that social timing is affected by temporal expectancies in cognition and emotion driven by underlying cerebral timing.

Aspects of a timing mechanism are already manifest in pre-verbal children, as shown by studies on synchrony (e.g. Feldman, 2007b; Feldman & Eidelman, 2007). From a developmental perspective, temporal organisation is crucial for increasingly sophisticated pre-verbal interactions, such as the 'conversations' infants have with their caregivers before they can speak (Trevarthen & Aitken, 2001). Timing of social interactions shows a correspondence with later years' development in language, attachment, pretend play, theory of mind and empathy (reviewed by Feldman, 2007b). Successful social timing is characterised by coordinated and timely turn-taking even in preschoolers (Trevarthen & Aitken, 2001). The following sections give the reader an idea of how temporal aspects impact on social interaction skills and their development.


The ability to estimate time intervals is crucial for predicting when to respond, for example, when catching a falling object (Meck, 2005). In communication, this is important for keeping up a smoothly flowing conversation. Infants can perceive time both at birth and even before birth (DeCasper & Carstens, 1980, DeCasper & Fifer, 1980), and are able to distinguish rhythm and rate (for a review, see Lewkowicz, 2000). Evidence from an electroencephalogram study demonstrated evidence of scalar property3 in 6-month-olds (Allman et al., 2012). Within the first half year of life, infants perceive temporal structures, can detect subtle changes in auditory and visual stimuli, and distinguish between rhythms (Gratier, 2003; Malloch, 2000).

Temporal processing and multi-sensory integration

Incoming stimuli need to be processed in a timely fashion, or the response may become redundant. Most, if not all, sensory experiences are made up of various modalities. Stimuli registered closely in time, have likely co-occurred and need to be combined meaningfully, through a process called multi-sensory integration (MSI; Kwakye, Foss-Feig, Cascio, Stone, & Wallace, 2011). MSI research covers how simultaneously occurring stimuli are perceived, how our brains bind them, and within which time-windows binding occurs. It is generally agreed that (near) temporal synchrony is the most important factor for MSI to occur (de Boer-Schellekens, Eussen, & Vroomen, 2013). MSI performance matures as adolescence is reached, but depends on the type of stimuli to be combined, their complexity, and whether they are of a social nature or not. Finally, the order of incoming stimuli affects temporal parameters because visual and oral stimuli reach the brain at different speeds (de Boer-Schellekens et al., 2013).

Various studies have explored MSI at different ages. For example, 10-16-week-old infants presented with audio-visual synchronised and non-synchronised (off by 400 ms) nursery rhymes preferred the synchronised presentations (Dodd, 1979). Toddlers, with an average age of 2.4 years, preferred watching synchronous displays of audio-video information when given the choice during the preferential looking paradigm4 (Bebko, Weiss, Demark, & Gomez, 2006).

Infants between 2-8 months detected audio-visual asynchrony of simple non-social stimuli when both events were 350 ms apart (sound first) and 450 ms apart (sound last) (Lewkowicz, 1996). Adults needed considerably smaller gaps to detect asynchrony: 65 ms (sound first) and 112 ms (sound last; Lewkowicz, 1996). In a flash-beep illusion task, multiple beeps are presented with one flash of light. If the flash is close enough to the beeps, the participant perceives an additional flash. TD children, aged between 8 and 17 years, experienced the illusion within a 300 ms time-window of proximity (Foss-Feig et al., 2010).

Motor timing

Motor timing research includes, for example, studies of motor coordination, postural control, eye-blink responses, and finger tapping. Motor timing is of considerable importance in development because a baby’s first mode of communication with his caregivers is through volitional movements (Schmitz et al., 2003). Children with motoric difficulties show less active play with peers (Trawick-Smith, 2014). In turn, less active play and/or impoverished social skills that hindered play decreased opportunities to practise motor skills during play/interaction. In sum, hindered motor skills may negatively affect opportunities for social interaction and exacerbate symptoms of repetitive/stereotypical behaviours and mannerisms (Lloyd, MacDonald, & Lord, 2011).

Brain anatomy and neurological findings

The brain regions associated with the internal clock are the cerebellum, basal ganglia, supplementary motor area and prefrontal cortex (Cope, Grube, Singh, Burn & Griffiths, 2013). These areas are linked anatomically and functionally (Cope et al., 2013). While they interact, these parts also act slightly differently, for example, the cerebellum has been shown to adjust to timing information and the basal ganglia to task order predictability (Dreher & Grafman, 2002).

The cerebellum is mainly associated with motor coordination (Gowen & Miall, 2005), which affects executive function, learning and language (Courchesne et al., 2004; Schmahmann & Sherman, 1998). Integrity of basal ganglia and cerebellum was found to play a key role for motoric synchronisation (Claassen et al., 2013).

Most relevant for this thesis' topic is the cerebellum’s connection to timing perception in the millisecond range, which impacts on social cognition (Van Overwalle, Baetens, Mariën & Vandekerckhove, 2013). Courchesne (1994) highlights the cerebellum’s implications for development through its role in selective attention and rapid attention shifts. He argued that the locus of information, which can be on objects, actions, sounds, speech, feelings and so on, changes rapidly and unpredictably in interaction, thus requiring frequent attentional shifts. Young children usually begin to master joint attention and rapid shifting to subtle cues at around 12-15 months (Bakeman & Adamson, 1984; Meindl & Cannella-Malone, 2011). This ability is thought to be one of the key factors for the healthy development of communication (Bakeman & Adamson, 1984).

2.2.2 Defining social timing

The term ‘social timing’ incorporates two aspects: intrapersonal timing, and interpersonal timing. Intrapersonal timing refers to the coordination of communicative expression and comprehension by an individual (including auditory perception, pitch, body language, etc.). Intrapersonal timing can take on a quality known as cyclicity, an important concept discussed below.

Interpersonal timing is about the timing of interactive behaviour between two people. In other words, it describes the coordination of communication modes (body language, vocalisations, verbalisations, etc.) between individuals (Jaffe et al., 2001). Predictability is a key feature of coordinated interpersonal timing; each interaction partner’s timing must be predictable from that of the other (Jaffe et al., 2001). Interpersonal timing requires adequate perception and understanding of social and communicative signals, and the ability to adapt to the other interaction partner continuously (Delaherche et al., 2012). It can take on the quality of synchrony (see below). Synchrony appears to be a universal feature as it has been observed in human interactions cross-culturally, including in interactions of Mayan Indians, Kung Bushmen, Eskimos (as reviewed by Condon & Sander, 1974).

What is interactional cyclicity?

The concept of cyclicity in humans was first noticed in biological rhythms5 to describe events recurring at regular intervals, such as a woman’s menstrual cycle. While in some contexts cyclicity implies strict regularity, as in the case of music, in others it is regarded less stringently, for example in the study of interaction (Jaffe et al., 2001; Lewkowicz, 2000).

Interactional cyclicity is alternatively referred to as self-synchrony and describes the mathematically definable rhythm of one’s own coordination of verbal and non-verbal communicative modes (Condon & Sander, 1974; Petitto, Holowka, Sergio, & Ostry, 2001). It quantifies the integration of communicative behaviour of one individual, for instance how change in someone’s own speech corresponds with their change in body movements (Condon & Sander, 1974). Some view cyclicity as a prerequisite for interactional synchrony and argue that one needs to be able to coordinate his own communication modes before one can coordinate them with someone else's (Lester, Hoffman, & Brazelton, 1985, p.24). Yet others are less certain that it is essential, and only suspect that deficits in intrapersonal coordination may reduce interpersonal timing (Marsh et al., 2013).

In parent-child interaction, cyclicity is thought to provide a framework within which caregiver and child co-ordinate their rhythms, both constraining and enabling interactive behaviours (De Jaegher, 2006; Lester et al., 1985). The repetitive nature of cyclical behaviour provides temporal expectancies by organising the infant’s and caregiver’s cognitive and affective expectancies (Stern, 1971), thereby making future actions more predictable (e.g. Lester et al., 1985; Tronick, Als, & Brazelton, 1980). In other words, cyclicity is thought to facilitate timing of both the infant’s and caregiver’s actions.

Mother-infant pre-verbal 'conversations' can be parsed into cycles of attention and non-attention, or positive and negative affect. For example, when a mother soothes her crying infant, the child moves from a negative state through a neutral to a more positive state (Brazelton, Koslowski, & Main, 1974).

What is interactional synchrony?

Interactional synchrony was first defined as rhythmic coordination between two people6 (Condon & Ogston, 1971), for example when a listener’s coordinates his movements to the speaker’s speech. Since then, the term synchrony in the context of interpersonal interactions has been used by many different researchers, each using different words, concepts and methodologies to describe this concept (Bernieri, Davis, Rosenthal, & Knee, 1994; Delaherche et al., 2012). See Table 2.1 below for definitions. Chapter 3 looks at different approaches to studying synchrony.

Descriptions that have been used to express this concept include entrainment (Condon & Sander, 1974; Malloch, 2000), attunement (Stern, 1985), mutual influence (Cohn & Tronick, 1988), co-regulation (Fogel, 1993), contingency (Nichols, Gergely, & Fonagy, 2001) and coordination (Jaffe et al., 2001). The term synchrony is used interchangeably with closely related concepts or requirements, such as mutuality (Delaherche et al., 2012) or congruence (Green et al., 2010), which will be explored in more detail below (see 2.2.3 and Table 2.2 below).

Table 2.1 Interactional synchrony definitions

Abbildung in dieser Leseprobe nicht enthalten

Delaherche (2012) defined interactional synchrony as the 'dynamic and reciprocal adaptation of the temporal structure of behaviours between interactive partners' (p. 351). Her definition is in agreement with the most frequent elements found in other researcher's definitions, which are coordination, dynamic (pertaining to temporal), reciprocal, predictability and rhythmicity (see Table 2.1 above). Each of these elements will be addressed in turn.

Dynamic emphasises the importance of an interaction's flowing nature rather than the actions contained within it, which distinguishes interactional synchrony from simply mirroring the other person (Delaherche et al., 2012; Fitzpatrick, Diorio, Richardson, & Schmidt, 2013). Reciprocity highlights the exchanging of behaviour of both interaction partners. In other words, the change of one person’s behaviour drives change in that of the other and vice versa (Condon & Ogston, 1971). While this seems to follow the dynamic criterion logically, not all studies examining 'synchrony' have actually studied both interaction partner’s behaviours (e.g. Green et al. 2010). Behaviours is set in plural to indicate that communicative behaviours are mostly multimodal events, describing meaningful units of communicational behaviour (Delaherche et al., 2012). However, some researchers have chosen to study isolated modalities such as speech or movement, thus studying vocal or movement synchrony (Feldman & Eidelman, 2004; Feldman, Magori-Cohen, Galili, Singer, & Louzoun, 2011; Feldstein, Konstantareas, Oxman, & Webster, 1982; Jaffe et al., 2001). Finally, the interactive element covers the various forms interaction can take, including verbal and non-verbal communication or play (Delaherche et al., 2012).

Scholars have also highlighted the following qualities of synchronous interaction; predictability (e.g. Feldstein, Jaffe, Beebe, & Crown, 1993; Jaffe et al., 2001), a harmonious smooth flow (e.g. Bernieri et al., 1994; Harrist & Waugh, 2002) and adaptation to the other (Feldstein et al., 1982; Harrist & Waugh, 2002).

2.2.3 Related but different uses of 'synchrony'

The term ‘synchrony’ is used to describe similar concepts in the interaction literature (for a review see Harrist & Waugh, 2002; Warner, 2002) yet they differ from interactional synchrony because they do not contain all of the criteria established by Delaherche and colleagues (2012). Thus, they lack one or more of the following: the temporal dynamic aspect, reciprocity, and occurring within an interactive setting (which can be verbal, pre-verbal or play) between interaction partners. Below three different but related concepts are described (see Table 2.2 Different uses of the term 'synchrony' below).

Table 2.2 Different uses of the term 'synchrony'

Abbildung in dieser Leseprobe nicht enthalten

Interactive flow

Firstly, synchrony has been used to describe the concept of ‘emotional engagement and interactive flow’ during interaction (e.g. Hedenbro & Tjus, 2007), which is established through the rater’s global perception (Delaherche et al., 2012) but lacks a clear temporal element.


Secondly, the term 'synchrony' has been defined as engagement in the same action at the same time during social tasks, and is used interchangeably with the term ‘synchronisation’, which it will be referred to hereafter (e.g. Fitzpatrick et al., 2013; Koehne, Hatri, Cacioppo, & Dziobek, 2016; Srinivasan et al., 2015). Synchronisation is closely related to imitation; however, they differ in that imitation does not necessarily focus on the temporal aspect (Fitzpatrick et al., 2013). To clarify, synchronisation occurs when both dyad partner’s actions coincide temporally and modality-wise, such as tapping at the same time. Synchronisation is distinct from interactional synchrony because it lacks reciprocity, interaction and behaviours may not be of a communicative nature. It is not reciprocal because in synchronisation tasks the parent was asked to follow a rhythm rather than starting a rhythm and then adapting to his child. It is not interactive because the tasks are neither of a conversational nor playful nature.


Finally, synchrony has been used to refer to parental closeness or congruence in following their child’s attentional or behavioural focus. Congruence studies differ from interactional synchrony studies because they lack reciprocity. Instead, these studies focus on one interaction partner only during social interaction, typically the adult caregiver (Green et al., 2010; Hudry et al., 2013).

2.3 Social timing from a developmental perspective

The following section explores the role that social timing plays in the development of cognitive, social and communicative abilities during a child's first year. The first part looks at developments leading up to synchrony and its increased sophistication. The second part explores the effects of synchrony on long-term outcomes.

2.3.1 Which developments lead up to synchrony?

A ‘feeling’ for time in pre-verbal communications is instilled at an early age, long before language is acquired (Stern, 1977). Early interactions pave the way for increasingly mature communication skills. Parents help the infant along by practicing timed interchanges unconsciously through their interactions, such as when playing games like ‘peek-a-boo’. The build-up of suspense uses timing, and eye glances are exchanged to signify surprise and joy. The repetitive nature of such early play serves an important function: by building up the infant’s attention, the parent allows for an optimal level of information intake. Too little or too much information can cause the infant to grow tired or turn away, and be detrimental to learning (Brazelton et al., 1974). By carefully balancing familiar and novel stimuli, such interactions further serve to facilitate cognitive growth, information processing skills and motivation (Feldman, 2012c). The following paragraphs lead the reader through developments in the first year related to social timing and interaction skills.

Before birth

During pregnancy, physiological systems mature in both mother and unborn child to prepare them for interactional synchrony (Feldman, 2012a). In the first trimester, the mother's oxytocin level predicts the amount of postpartum behaviour that she will express after birth as well as her coordination with the baby (Feldman, Weller, Zagoory-Sharon, & Levine, 2007). During the last trimester of pregnancy, the infant’s neurological basis for coordinated interactions develops in the brain (Feldman, 2007b). Firstly, the 'biological clock' matures and the sympathetic nervous system gains control over heart rhythms. This is followed by structural and functional brain development, including the assembly of brain nuclei, rapid increase in synaptic growth, and maturation of neuro-chemical systems (Feldman, 2006).

One to two months

Maternal postpartum behaviour is genetically programmed and triggered right after birth, and serves a crucial role for the infant's care, survival and development (Feldman, 2012a; Feldman & Eidelman, 2007). It includes gazing at the infant, 'motherese' vocalisations, positive affect, and affectionate touch (Feldman, 2012a). These behaviours sensitise an infant to micro-level alterations of the mother's interactional tools (Feldman & Eidelman, 2007), thereby preparing the infant for communicational interchanges early on.

A newborn’s interest in faces is present right from birth (Johnson, Dziurawiec, Ellis, & Morton, 1991) and considered paramount for developing social interaction and cognition skills (Williams, Whiten, Suddendorf, & Perrett, 2001). In the first month of life, interactions revolve around the infant's basic attention to the mother's face (Lavelli & Fogel, 2005), and there is a readiness to respond to human speech and a preference for their mother’s voice (DeCasper & Fifer, 1980).

Condon and Sander (1974) found that American-born neonates, as young as 12 hours, synchronised their movements to adult speech using micro-analysis of video-tapes. Synchronisation was found when the adult speaker was present and addressing the neonate, but also when the voice came from an audio tape. Remarkably, both American English and Chinese language fragments elicited coordinated movements by the infant but disconnected vowel sounds and tapping sounds failed to do so (Condon & Sander, 1974).

Additionally, babies were found to be innately equipped to detect contingencies (Tarabulsy et al., 1996), which depended on the baby's autonomic maturity, indicated by cardiac vagal tone, and predictive of later parent-infant synchrony (Feldman & Eidelman, 2007). Infant and maternal preparedness provided the components for the baby’s first experiences of temporally matched interactions. Such early means of communication were indexed by the rhythmicity in neonate activity, such as crying, nursing, or sucking (Feldman, 2007b). Furthermore, mothers were shown to model interaction rhythms on such patterns, for example, the burst–pause pattern typical of early face-to-face interaction (Tronick, Als, & Brazelton, 1977).

To emphasise the importance of maternal behaviour, examples are considered whereby maternal mental health is compromised, such as in mothers suffering from depression, anxiety or those who experienced stressful preterm labour. These mothers showed more hostility, dominance and lower responsiveness to their child, engaged less often and experienced difficulties with providing optimal levels of stimulation and appropriate responding (Feldman et al., 2009; Zlochower & Cohn, 2001). Additionally, mothers affected by depression or anxiety engaged less in postpartum behaviour, which was a main predictor of parent-infant synchrony at 3 months (Feldman et al., 2009, Feldman & Eidelman, 2007). Less affective touch and problems with emotion regulation further exacerbated difficulties in reaching synchrony, which was disturbed in all of its aspects, including taking twice as long to reach synchrony compared to dyads with non-depressed mothers (Feldman et al., 2009; Granat et al., 2017).

Dyads with premature babies were doubly at risk for lower levels of synchrony (Feldman & Eidelman, 2007). Firstly, mothers of preterm babies tended to show less maternal behaviour, looked, smiled, vocalised and touched their children less frequently. Secondly, preterm infants were more irritable, less tolerant for changes in affective behaviour and showed more poorly organised cues for interaction. This in turn, made it harder for their mothers to 'read' their infants, and possibly led them to adopt disadvantageous behaviours, such as overactive stimulation, intrusiveness and interaction dominance (Feldman & Eidelman, 2007).

Three to four months

In an infant’s second month, mother-infant interaction undergoes a shift from simple attention to active and emotionally positive attention (Lavelli & Fogel, 2005). From the second to the third month, the child gains the ability to display temporally coordinated facial expressions and vocalisations, including cooing, gazing, and smiling (Lavelli & Fogel, 2005; Yale et al., 2003). Interactions mature further to involve recurring rhythmic cycles using different modalities to communicate, such as gaze, touch, affective expressions, bodily movements, and arousal indicators. Parent–infant pre-verbal 'conversations' start showing a temporal structure; behaviour between interaction partners starts to match, and turn-taking emerges (Feldman, 2007b; Trevarthen, 1974).

It was first suggested that the ability to coordinate timing in social situations emerged at 4 months of age (Jasnow, Crown, Feldstein, Taylor, & Beebe, 1988). However, Feldman has shown that 3 months marks the initial period of gaze synchrony (Feldman, 2007b; Feldman & Eidelman, 2007), and most studies on parent-infant interaction test synchrony at this age7. Touch synchrony, defined as the coordination of affectionate touch and shared gaze, increases significantly from 3 to 9 months with the development of the infant's fine-motor skills (Feldman, 2007b). Mothers and their babies were found to synchronise their heart rhythms within one-second time lags, with even closer matching during episodes of vocal and affect synchrony (Feldman et al., 2011).

Interactions are parent-specific. Interactions with mothers are more rhythmic and socially oriented, whereas interactions with fathers are more environmentally oriented, encouraging exploration and are jerkier, including high arousal and a less rhythmic contour (Feldman, 2003). Triadic synchrony, between both parents and the infant, emerges at this age. Infants have to coordinate their behaviour to each parent as well as to non-verbal cues between parents (Gordon & Feldman, 2008).

Five to 12 months

Infant's social interactions evolve further and by 6 months, mothers’ and infants’ interaction patterns are well-established, benefitting from intimate knowledge and the ability to anticipate each other's temporal expressions (Sigman, Dijamco, Gratier, & Rozga, 2005). Intentionality emerges and interactions become more reciprocal in nature (Feldman, Greenbaum, & Yirmiya, 1999). Facial expressions become more complex, as infants coordinate smiles, gaze and gestures to communicate about objects while maintaining joint attention (Sigman et al., 2005).

Infants enter the age of intersubjectivity at around 9 months (Stern, 1985), which has been hypothesised to affect the lead-lag structure, which shifts from infant dominance (or the parent following the child) to mutual influence (see Appendix A for a deeper discussion). Nine months also marks the end of the 'sensitive period' of synchrony development (Feldman, 2015). Dyads sustain mutual coordination across time (Feldman, 2007b). Time-lag to synchrony decreases from 3 to 9 months, reflecting increased familiarity with the partner’s interaction style and maturing of the relationship, both facilitating faster matching (Feldman, 2007b). Once established, the individual degree of synchrony (coherence) remains stable across the first year (Feldman, 2007b). Following the first stages of prefrontal maturation, the ability to self-regulate emerges, which shapes an individual's regulation skills throughout life (Feldman, 2012a).

Post-partum depression and anxiety, not due to premature birth, single parenthood, poverty, or teenage parenthood, negatively affects gaze and touch synchrony measured during dyadic interactions with 9-month old infants (Granat et al., 2017). Depression was associated with lower levels of synchrony, whereas anxiety predicted higher levels of synchrony linked with maternal intrusiveness, yet both indicated suboptimal synchrony (Granat et al., 2017).

Beyond the first year

Once established, non-verbal (gaze, touch, affect expressions) and verbal-symbolic synchrony remains a feature of close social interaction throughout life (Feldman, 2012a). For instance, interactions between romantic couples show both of these aspects during an intimate conversation. Levels of synchrony are reflected in vagal tones and oxytocin levels, mimicking early synchronous interactions between mother and child during bonding (Feldman, 2012b). Later in life, synchrony also manifests in switching pauses in spoken conversational language (e.g. Feldstein et al., 1982).

2.3.2 What is the significance of synchrony for later development?

Synchrony measured at 3 or 5 months promotes a range of developmental skills, including turn-taking, emotion regulation, symbolic use and language. It further promotes healthy attachment. Each is briefly addressed below.


Early pre-verbal interactions lay the groundwork for turn-taking and reciprocity, and both play a key role for increasingly sophisticated interactions (Dominguez, Devouche, Apter, & Gratier, 2016; Lorenz, Weiss, & Hirche, 2015). Rhythmic temporal patterns facilitated the ability to expect behaviour in pre-verbal turn-taking, which is a necessary step towards fluent verbal conversations (Jaffe et al., 2001). Mastering interactional synchrony requires the ability to anticipate and predict when it’s one’s turn to speak, as prolonged or shortened pauses may be awkward and even alter meaning (Feldstein et al., 1982).


Early attachment relationships are believed to provide models for all later social interactions, and thus play an important role for long-term outcomes (Main, Kaplan and Cassidy, 1985). Synchrony at 3 and 4 months predicted attachment security at 12 months (Isabella et al., 1989, as cited in Hane et al., 2005; Jaffe et al., 2001). Likewise, dyads who interacted in a well-timed, reciprocal and mutually enjoying manner at 3 and 9 months developed a secure attachment style later (Feldman, 2012c). In contrast, dyads with minimally involved, insensitive or intrusive mothers were later insecurely attached or showed an avoidant attachment style (Isabella & Belsky, 1991).

Emotion regulation

Maternal positive engagement preceded the infant becoming more positive at 3 and 6 months (Cohn & Tronick, 1988). As the infant matured, his growing independence allows him to self-regulate. At 9 months, maternal positive engagement was not necessary anymore for the infant to become more positive (Cohn & Tronick, 1988). Feldman (2007b) hypothesised that infants who did not experience coordinated interactions with a caregiver very early in life may suffer pervasive difficulties in their social, emotional, and self-regulatory development. This is because participation in well-structured interaction with a sensitive adult was seen to foster several skills; firstly, the ability to empathise with the emotional states of others, using symbols and generally function in society. In fact, Feldman (2007a) provided evidence that level of synchrony across the first year predicted level of empathy in adolescence. Synchrony and empathy were found to mediate a child’s self-regulatory abilities at 2, 4, and 6 years (Feldman, 2007a).

Further, synchrony with either parent at 3 months was related to fewer behavioural problems 2 years later, indicating that better self-regulation is enhanced following the achievement of synchrony (Feldman & Eidelman, 2004). Conversely, parents of triplets who provided the same amount of parenting but reduced coordination between their and their infant’s social cues, showed diminished levels of parent-infant synchrony at 3 months. Lower synchrony predicted lower attachment at 12 months and more behavioural problems at 24 months; deficits which could be recovered from 5 years later (Feldman & Eidelman, 2004).

Symbolic use and theory of mind skills

Correlations between synchrony at 5 months, symbolic play at 3 years and theory of mind skills at 4 years suggest a preparedness for the development of the abstract and creative facets of language, and the ability to see several perspectives (Feldman, 2007b).

Language development

Lester et al. (1985) speculated that evidence for synchrony’s effect on language may be provided by his study on cyclicity in term and pre-term infants. Premature infants, compared to full-term infants, were found to show less coordination with their mothers and to subsequently have more difficulties with language development (Lester et al., 1985).

In sum, these findings attest to the importance of synchronous interactions in the first year of life. Examples from dyads at risk for achieving optimal synchrony showed how easily it is disrupted and may affect other developments during this sensitive phase.

2.4 Social timing in ASD

ASD8 is marked by impairments in social interaction and communication, and restricted and repetitive patterns of behaviour (see Table 2.3 below). Impairments must be present in early development, and cause clinically significant impairment (American Psychiatric Association, 2000). The term ASD combines three previously separate diagnoses: Autistic Disorder, Asperger’s Syndrome and Pervasive Developmental Disorder - not otherwise specified (PDD-NOS; American Psychiatric Association, 2000).

Restricted, repetitive patterns of behaviour, interests or activities may include stereotyped or repetitive motor movements, rigidity involving routines or ritualising behaviour, highly focused interests and being overly sensitive to sensory input (American Psychiatric Association, 2013).

2.4.1 Social and communication deficits in children with ASD

Difficulties with social interactions and communication are one of the main features of ASD and needed for a diagnosis (American Psychiatric Association, 2000; 2013). Already present in infancy, these deficits impact on and exacerbate other symptoms, thereby affecting individuals throughout their life (American Psychiatric Association, 2000; Wimpory, Hobson, Williams, & Nash, 2000). The following paragraphs illustrate the nature of social and communication impairments in ASD with a focus on pre-verbal interaction.

Social deficits in ASD

A variety of social deficits are present in children with ASD that set them apart from typical developing (TD) peers and from children with developmental delay (DD). Firstly, children with ASD do not engage in spontaneous initiation of interaction or joint attention as frequently as TD peers. When they do, these interactions are classified as low-level interactions9 in contrast to TD children and those with DD (Loveland & Landry, 1986; Mundy, Sigman, Ungerer, & Sherman, 1986; Sigman, Mundy, Sherman, & Ungerer, 1986). Long-term effects of initiations in early childhood predicted later language gains (Siller & Sigman, 2008).

Deficits in social-emotional reciprocity are also commonly found in children with ASD. From a large sample of children diagnosed with ASD or PDD-NOS, a third were found to engage on a need-fulfilment basis only, however the majority of children (40%) showed both reciprocity and need-fulfilment activities (Greenspan & Wieder, 1997).

Turn-taking, the back-and-forth between interaction partners that structures non-verbal and verbal conversations, is often diminished or absent in ASD (Chiang, Soong, Lin, & Rogers, 2008; Mundy et al., 1986; Wimpory et al., 2000).

Children with ASD also show abnormal responding to speech stimuli. For example, infants with ASD regularly failed to respond to their name, which is considered an early warning sign, and consistently distinguished children with ASD from others (Baranek, 1999; Osterling & Dawson, 1994; Saint-Georges et al., 2010; Zwaigenbaum et al., 2005). Furthermore, preschool children with ASD preferred noises over their mother’s speech. This was reflected by differential brain activation for speech stimuli and may account for abnormal responding and impaired language development (Klin, 1991; Kuhl, Coffey-Corina, Padden, & Dawson, 2005).

Imitation plays an important role in learning new skills via copying, symbolic play and expressive language skills (Stone, Ousley, & Littleford, 1997). Copying of facial and bodily movements seems impaired in ASD, and more difficult than action imitation using objects (Biscaldi et al., 2015; Rogers, Hepburn, Stackhouse, & Wehner, 2003; Stone et al., 1997). Compared to TD peers, children with ASD imitate less frequently and accurately (Vivanti, Trembath, & Dissanayake, 2014) but performance depends on task, difficulty, and symptom severity (Biscaldi et al., 2015; Rogers et al., 2003).

Finally, individuals with ASD often suffer difficulties with relationships, for example with forming, maintaining and understanding relationships. In addition, they may experience problems with adjusting behaviour to situational circumstances (American Psychiatric Association, 2013). In early childhood, such deficits surface as a lack of imaginative play and an apparent disinterest in peers (American Psychiatric Association, 2013). Besides deficits in the social realm, individuals with ASD also experience deficits with communication, which are addressed below.

Communication deficits in ASD

Non-verbal communication impairments in ASD may manifest as abnormalities in eye contact, facial expressions, body language, and problems with making and understanding gestures. In other words, problems arise with non-verbal pragmatics10 of communication. More severely affected children with ASD might show a profound lack of drive to communicate, which is in stark contrast to young children not affected by ASD (Rapin & Dunn, 1997). Comprehension of gestures, facial expressions and tone of voice is also widely impaired in ASD (Rapin & Dunn, 1997). Finally, deficits in this category include poor integration of non-verbal and verbal communicative behaviours, such as integration of eye contact or gestures with speech (Silverman, Eigsti, & Bennetto, 2017).

Joint attention is also often difficult for people with ASD (Bruinsma, Koegel, & Koegel, 2004; Charman, 2003; Vivanti et al., 2014). Imperative joint attention11 was found to be relatively intact in children with ASD, however, there seem to be major difficulties with declarative joint attention12 (Loveland & Landry, 1986; McEvoy, Rogers, & Pennington, 1993; Mundy et al., 1986; Mundy, Sigman, & Kasari, 1990; Sigman et al., 1999; Stone et al., 1997).

Language deficits in ASD

Heterogeneity of language development and skills in ASD is large, and a language deficit is not required for a diagnosis of ASD today (American Psychiatric Association, 2013; Kjelgaard & Tager-Flusberg, 2001). If language develops, onset may be later and the developmental trajectory differs compared to typical language development (Mitchell et al., 2006). Loss of language abilities, is rare, but has been recorded in ASD (Pickles et al., 2009).

Pragmatics, comprehension and formulation of discourse is often impaired (Rapin & Dunn, 1997). Comprehension abilities may vary. In one study with children suffering from an auditory processing deficit, receptive skills were absent in more than half the sample13 (55%), while the other half (41%) had some receptive skills including understanding of single words and simple directions (Greenspan & Wieder, 1997). Only 4% showed an understanding of more complex directions (Greenspan & Wieder, 1997).

Speech of people with ASD often differs in prosody (either very high-pitched and squeaky, or robotic and monotonous), phonology14 and syntax15 (Kargas, López, Morris, & Reddy, 2016; Lyakso, Frolova, & Grigorev, 2016; Nakai, Takashima, Takiguchi, & Takada, 2014). Echolalia, the repetition of a word or phrase, is often seen in very young children with ASD and often does not fade as the child’s language abilities mature (Rapin & Dunn, 1997).

Language development may be predicted by pre-verbal communication skills in early childhood, functional play skills and responsiveness to joint attention requests (Mundy et al., 1990; Sigman et al., 1999). Non-verbal IQ score may play a role in subsequent language development (Bopp, Mirenda, & Zumbo, 2009). Conversation content is often ritualistic and revolves around special topics.

In summary, there is a range of deficits and abnormalities that characterise social, communication and language abilities in individuals with ASD. For an overview see Table 2.3 below.

Table 2.3 Overview over ASD deficits

Abbildung in dieser Leseprobe nicht enthalten

2.4.2 Timing explanations of ASD

Timing abnormalities may explain diagnostic features seen in ASD (e.g. Allman, 2011; Amos, 2013; Grossberg & Seidman, 2006; Wimpory, Nicholas, & Nash, 2002). Perceptual, learning, memory and central coherence problems could be ascribed to difficulties in integrating complex sensory information (Bertone, Mottron, Jelenic, & Faubert, 2005). Social and communication difficulties may be due to atypical social timing patterns. For instance, poor social timing skills act to reduce social bonding in early childhood, joint attention and social reciprocity (Jaffe et al., 2001). Restricted and repetitive behaviours and interests may serve by ‘segmenting time’ and thus be a coping mechanism aimed at organising temporal stimuli (Amos, 2013; Spiker, Lin, Van Dyke, & Wood, 2012). Finally, problems with skills that involve putting oneself in someone else’s point of view such as pretend play, empathy, theory of mind, might require the ‘mental time travel’ that people with ASD find challenging and confusing (Amos, 2013; Feldman, 2012c).

This section introduces timing explanations of ASD. Cognitive hypotheses which function at a descriptive level in that they explain perceptual and comprehensive processes are illustrated first. It is shown how temporal anomalies at the cognitive level may affect processing, comprehension and behaviour in ASD. Next, neural causes which possibly underlie cognitive changes are discussed. Finally, genetically driven timing explanations of ASD are investigated and their possible role as the ultimate cause of ASD as they may determine neural and cognitive changes.

Cognitive timing theories of ASD

In line with Weak Central Coherence16 theory (Happé & Frith, 2006), it has been suggested that autistic features may be caused by a deficit in temporal binding due to a reduced integration of neural networks (Brock, Brown, Boucher, & Rippon, 2002). The problem seems to lie with binding between networks (Brock et al., 2002). In contrast, binding within networks is thought to be intact or even enhanced, explaining an enhanced local focus and impaired global processing.

Difficulties with language processing may be due to reduced contextual influence. Impaired attentional shifting could be ascribed to reduced connectivity between frontal lobes and posterior regions, further causing behavioural inflexibility in joint attention. Theory of mind impairments could be associated with poor integration of multiple sources of information (Brock et al., 2002). A real-life example of what poor integration may mean for an individual with ASD is provided by Amos (2013), who explains that her own son with ASD copes by watching television muted with captions, so that he only needs to attend to visual instead of audio-visual information. In sum, a temporal binding deficit may underlie ASD symptoms and could explain social and communication deficits, restricted behaviours and why some skills are seemingly enhanced in ASD. However, it remains unclear where this binding deficit stems from.

A different proposition suggests that ASD is a temporo-spatial processing disorder (TSPD) affecting multi-sensory flows caused by multi-system Brain Disconnectivity-Dissynchrony (MBD; Gepner & Féron, 2009). MBD describes increases/decreases of functional connectivity and neural synchrony within/between brain regions. In general, temporo-spatial processing problems affect (1) detection and integration of visual motion; (2) coding and parsing language; and (3) anticipation and programming postural adjustments (see Table 2.4 below).

Table 2.4 Temporo-spatial processing problems

Abbildung in dieser Leseprobe nicht enthalten

In response to overwhelming environmental demands, people with ASD may compensate by focusing on static visual stimuli or auditory singularities instead. This may explain enhanced abilities in areas such as spatial memory or pitch sensitivity (Gepner & Féron, 2009). Similarly, repetitive behaviours like switching the lights on and off may be used as coping strategies to slow down the world (cited in Gepner & Féron, 2009). In support, deliberate slowing down of stimuli increased levels of performance in severely affected children with ASD on measures such as face and emotion recognition, voluntary imitation and sentence comprehension (Gepner & Féron, 2009).

In sum, Gepner and Féron (2009) concluded that the environment may be too fast for people with ASD. They support this notion using quotes by able autistic people such as Donna Williams who explained that: "the constant change of most things never seemed to give [her] any chance to prepare [herself] for them." The following subsection addresses brain alterations as a possible underlying cause of cognitive timing problems in ASD.

Neural deviations may underlie timing anomalies in ASD

Timing anomalies in ASD may be due to neural deviations. For example, cerebellar damage and loss of Purkinje cells causes problems with rapid attention shifting. Attrition in both of these structures may be responsible for social and communication deficits in ASD (Courchesne et al., 1994).

The cerebellum is described to act as a ‘master computational system’ (p. 861; Courchesne et al., 1994). Purkinje cells connect the cerebellum to other areas in the brain. Together they support brain structures involved in motoric, attentional, arousal, sensory, memory systems to produce timely, accurate and appropriate responses. Without the cerebellum’s support these systems still function but sub-optimally.

In support of this idea, Courchesne and colleagues (1994) found that the impairment is limited to rapid attentional shifts, specifically those under 2.5 seconds. At slower speeds, performance of children with ASD was comparable to age-matched controls (Courchesne et al., 1994). Slowed attention shifting can interfere with social functioning. For example, if a child is unable to adequately shift his focus between rapidly incoming information from gestures, verbalisations, postural, tactile and facial cues, important information may be missed. In response to information overload, the child might withdraw from the interaction (Courchesne, 2004). In short, there is evidence that cerebellar damage hinders processing of rapidly incoming sensory information, thus negatively affecting fast paced interactions.

In addition, structural differences in the inferior olive of individuals with ASD may disrupt its role in the brain, resulting in impairments of processing and reacting to rapid stimulus sequences (Welsh et al., 2005). The authors present evidence for both the inferior olive's role, and why this might affect language development in early childhood. According to Welsh, the infant’s brain needs to develop a processing speed that allows him to adapt to the fast communication speed of adult language in order to have fluid conversations.

First ‘conversations’ between infant and caregiver are slow, and the adult typically follows the infant’s lead. As the infant matures, the speed of conversational flow increases and influence becomes mutual (Feldman, 2007b). A global decrease of motor and cognitive speed would thus hinder early communication, with profound consequences for language acquisition (Welsh et al., 2005). This idea appears closely linked with the impaired rapid attention shifting deficit that Courchesne et al. (1994) attributed to cerebellar damage. In fact, the inferior olive is a major input hub to the cerebellum (Xu, Liu, Ashe & Bushara, 2006).

It is possible that both structures are impaired jointly or separately, in both cases causing similar symptoms. In sum, it is conceivable that neural alterations underlie cognitive changes in ASD, which manifest in temporal deficits. However, neural explanations of ASD do not address where structural brain differences stem from. Genetics research suggests answers that may reveal the underlying cause of structural and cognitive anomalies in ASD.

Genetics-the ultimate cause ?

Researchers have hypothesised that social timing may have a genetic basis (Wimpory et al., 2002). Evidence for this comes from fruit fly studies but equivalents were found in humans and mice (Konopka & Benzer, 1971; Sun et al., 1997). The genes that regulate timing are known as 'clock genes'. They are responsible for circadian timing and thereby regulate metabolic function. Clock genes are thought to influence other aspects of timing and social timing, such emotional and contextual memory and communication, for example, courtship patterns. Circadian timing may play a part in social interaction by supplying 'timekeeping cues' (Barnard & Nolan, 2008).

Dysfunctional circadian timing has been linked to ASD where it may manifest through sleep disturbances (Barnard & Nolan, 2008; Cortesi, Giannotti, Ivanenko, & Johnson, 2010). In ASD, sleeping problems and social problems appear closely linked (Richdale & Prior, 1995). In sum, timing appears to be influenced by processes involving specialised genes.

Genetic anomalies in clock genes and methylation changes that are specific to ASD may cause social timing deficits (Wimpory et al., 2002). Methylation status of a gene decides whether it is expressed or not. In ASD there is reasonable evidence to suspect that methylation status is anomalous (Wimpory et al., 2002). Two genes in particular (per1 and npas2) appear to be significantly17 associated with ASD (Nicholas et al., 2007). Per1 is implicated in lack of cerebellar Purkinje cells. Npas2, a clock-related gene, is thought to affect sleep-wake cycles, and possibly plays a role in memory formation (Barnard & Nolan, 2008). Both genes are linked to the cerebellum, forebrain and limbic system. Given the role the cerebellum and Purkinje cells play in timing, alterations within those structures may be responsible for timing, contextual and memory anomalies in ASD. In sum, neural anomalies in ASD may be based on genetic differences (albeit indirectly). Further, the affected neural structures play a role in functions that support social and communication. Thus, anomalies in these brain structures plausibly lead to problems in social timing, which may manifest early in parent-infant interactions (Nicholas et al., 2007).

2.4.3 Timing in ASD

The following section examines evidence of differential timing in ASD. Various research domains are considered, including cognition, circadian rhythm, motor timing, and neurology.

Interval timing

Interval timing in ASD may be attenuated compared to TD controls, however findings are mixed (Jones, 2017). Amongst time estimation studies, two reported that timing was intact in ASD (Mostofsky, Goldberg, Landa, & Denckla, 2000; Wallace & Happé, 2008). Two other studies supported this, but found that the ASD group was less sensitive at estimating time (Allman, DeLeon, & Wearden, 2011; Falter et al., 2012). In contrast, one other study found a clear impairment in time perception (Brodeur, Gordon Green, Flores, & Burack, 2014). Likewise, temporal bisection (Gil, Chambres, Hyvert, Fanget, & Droit-Volet, 2012; Jones, Lambrechts, & Gaigg, 2017) and time production (Wallace & Happé, 2008) appeared intact.

Out of six studies that investigated reproduction of time intervals, four found an impairment (Brenner et al., 2014; Karaminis et al., 2016; Martin, Poirier, & Bowler, 2009; Szelag, Kowalska, & Galkowski, 2004). Mixed findings with equal performance in intermediate intervals and poor performance for intervals at extreme ends were also reported (Maister & Plaisted-Grant, 2011). Only one study reported intact timing abilities with a tendency for superiority in the ASD group (Wallace & Happé, 2008). Most authors observed large variability within the ASD groups. The vast heterogeneity within ASD and methodological differences may account for mixed findings. Attributing a causal role to interval timing deficits may be premature, however, existing research points to the likelihood that timing deficits interact with and modulate symptoms in ASD (Allman & Falter, 2015; Falter & Noreika, 2011).

Temporal processing of multiple inputs and synchrony

The majority of studies testing multi-sensory integration found a deficit in ASD dependent on stimulus presentation (Foss-Feig et al., 2010; van der Smagt, van Engeland, & Kemner, 2007). Audio-visual integration was intact in youth with ASD, however, ERP analysis revealed differential activation of neural networks in the ASD group, with time windows corresponding to differences in lexical-semantic processing (Megnin et al., 2011). De Boer-Schellekens et al. (2013) observed that people with ASD were less sensitive to asynchrony, but performance improved at larger latencies. This supports other evidence showing that performance of timing task improves when timing is slowed down.

Temporal order judgements of aural and visual stimuli revealed no significant differences between children with ASD and a TD control group (Kwakye et al., 2011). Detection of audio-visual synchrony with speech and non-speech stimuli showed no overall group difference, however, results indicated an impaired level of responding in tasks using speech (Bebko et al., 2006). Similarly, judgements of simultaneity and asynchrony of visual stimuli showed that the ASD group had lower synchrony thresholds, suggestive of an impaired ability to integrate stimuli (Falter, Elliot & Bailey, 2012). Finally, synchronised button pressing was done earlier and with greater variability by male adults with Asperger's compared to a matched NT control group (Gowen & Miall, 2005).

Temporal processing of auditory information

Research into auditory temporal processing points to an overall impairment when using social stimuli (Chevallier, Noveck, Happé, & Wilson, 2011; Dawson, Meltzoff, Osterling, Rinaldi, & Brown, 1998; Kargas et al., 2016; Lerner, McPartland, & Morris, 2013; Russo, Zecker, Trommer, Chen, & Kraus, 2009) but not when using non-social stimuli (Dawson et al., 1998; Fujikawa-Brooks, Isenberg, Osann, Spence, & Gage, 2010; Jones et al., 2009). Auditory training in children with ASD removed deficits post training (Russo, Hornickel, Nicol, Zecker, & Kraus, 2010).

Temporal processing of visual information

Several studies in youth and adults with ASD have revealed slowed processing of facial information as indicated by N170 latency (for example McPartland, Dawson, Webb, Panagiotides, & Carver, 2004), including one finding that the ASD group processed objects faster than faces (Webb, Dawson, Bernier, & Panagiotides, 2006). Another ERP study by the same lab tested facial emotion processing in children with ASD using both child and adult face stimuli. While there were no differences for recognising emotions from child faces, significant group differences were found for adult faces, with longer latencies for early perceptual processing, indicated by N170 amplitudes, rather than emotion recognition deficits (Lerner, McPartland, & Morris, 2013). It should be noted that these deficits were behaviourally minor, and that within the ASD groups large variability was found with some individuals performing very well and others very poorly. In a study looking at spontaneous and voluntary (participants were asked to mimic facial expressions) facial mimicry in response to pictures of happy, sad, angry and neutral faces, the authors found that while children with ASD did mimic facial expressions spontaneously, it took them longer to do so. No difference was found for voluntary mimicry, thus the ASD group performed just as well as TD controls (Oberman, Winkielman & Ramachandran, 2009). The authors propose a potential issue with automatic engagement of sensory-motor mechanisms involved in timing of social interactions, consistent with neuropsychological evidence.

Circadian rhythm

Timing difficulties in ASD are also evident in circadian rhythms. For instance, sleep disturbances are well-recorded in ASD (Cortesi et al., 2010; Elia et al., 2000; Limoges, 2005; Richdale & Prior, 1995). This is further supported by alterations in melatonin patterns (Nir et al., 1995) and evidence from genetic studies (Hu et al., 2009; Nicholas et al., 2007).

Motoric timing

Studies on motoric timing indicated general differences in gait, muscle tone, and balance in ASD (Teitelbaum et al., 2004; Teitelbaum, Teitelbaum, Nye, Fryman, & Maurer, 1998). Performance may be normal in simple tasks (tapping as fast as possible) but added complexity revealed differences in speed and execution strategies (Gowen & Miall, 2005).

Neural correlates

Finally, insights from studies on neural correlates add to the evidence that timing is disrupted in ASD. Most notably, the cerebellum appears to be developing differently (Courchesne et al., 2001; Courchesne et al., 2004) with implications for timing processing and motor abnormalities.

In sum, converging evidence shows that a timing deficit is present in ASD. While exceptions exist, performance was almost always found to be diminished in ASD. Timing studies in ASD revealed three overarching themes;

- Processing of social stimuli is impaired while non-social stimuli remain largely intact
- Increased task complexity diminishes performance in ASD more so than in TD people
- Rapid presentation of stimuli further impairs task performance in ASD
It follows that social interaction, which contains all of these features, must be especially difficult for individuals with ASD. The following subsection addresses this question.


1 Traditionally, research has focused primarily on mother-infant interaction. Some studies (e.g. Feldman, 2003) examined father-infant versus mother-infant interactions, and found that interactions differed in content but not temporal parameters. However, gender matched (i.e. father-son or mother-daughter) interactions showed higher levels of synchrony, more balance in leading-and-following and shorter time needed to reach synchrony.

2 The term rhythm, when used in connection to social interaction, is meant to be flexible and allows for variations. This is in contrast to the musical term, which is characterised by mathematical regularity and repetitiveness.

3 The scalar property is a fundamental characteristic of interval timing and manifests through mean accuracy and variance when perceiving or reproducing time intervals (Falter, Noreika, Wearden, & Bailey, 2012). Mean accuracy means that on average, perceived or reproduced time intervals are equal or nearly equal to the true length of the time interval. The variance principle requires that the timing sensitivity remains constant, even as the time to be perceived or reproduced, varies. Scalar property is thought to reflect the analogue mental representations that obey Weber’s law, which are also found in other forms of perception (Allman, Pelphrey, & Meck, 2012).

4 In this preferential looking paradigm participants saw non-verbal, simple verbal or complex verbal video segments. Two screens showed a video segment each, accompanied by a single identical sound track available to the participant. One of these audio-video displays was asynchronous by 3 seconds, whereas the other was synchronised. Participants' gaze was video-recorded and coded for time spent looking at each screen. The screen most looked at was assumed to be the preferred one.

5 Biological rhythms are the circadian 24-hour rhythms within which physiological functions are regulated, including for example, sleep, body temperature, and hormone production.

6 This thesis focuses on dyadic interaction, i.e. interaction between two people. However, some studies also examined triadic interaction.

7 The age of 4 months is especially suitable because at this stage of development infants are interested and able to take part in face-to-face interactions with their caregivers but not quite able yet to grasp for objects and move about the room, so that all their attention is directed to human interaction (Hane, Feldstein, & Dernetz, 2003).

8 Note that throughout this dissertation, the term ASD is used to refer to all manifestations of the condition as defined in the DSM-V. Where appropriate, for instance when reporting research, there is a distinction made between AD, AS and/or HFA.

9 Low level initiations include making eye contact with other while holding a toy, while high level initiations include pointing/showing or giving toy.

10 Pragmatics of language refer to the implicit rules for using language communicatively and appropriately, so that conversation is meaningful and engaging (Rapin & Dunn, 1997). Verbal pragmatics refers to such things like turn-taking, staying on topic, providing conversational partners with appropriate information to clarify meaning and so on.

11 Requesting help in attaining an object or event from a caregiver by using gestures, pointing, showing, and following someone’s gaze to where they are looking (Doussard-Roosevelt, Joe, Bazhenova, & Porges, 2003).

12 Directing a caregiver’s attention to an object, action, or entity. Declarative joint attention may include pointing to, showing, or giving of objects (Gabig, 2013).

13 These findings were based on a clinical records review of 200 children diagnosed with autism or PDD-NOS according to the DSM-IV. Children were between 22 months and 4 years (Greenspan & Wieder, 1997).

14 Phonology refers to the auditory code, or speech sounds that make up words. If deficient, words can be distorted and hard to understand (Rapin & Dunn, 1997).

15 Syntax refers to the grammatical rules that constitute well-formed, clear sentences (Rapin & Dunn, 1997).

16 Weak Central Coherence posits that individuals with ASD tend to process locally rather than globally unlike neurotypicals, who are biased to a global processing style (Happé & Frith, 2006).

17 The authors note that correction for multiple testing would render results non-significant, due to the small sample (N = 90) and effect sizes (Wimpory et al., 2002).

Excerpt out of 172 pages


Social Timing in Preschoolers with Autism Spectrum Disorder
University of Wales, Bangor
Catalog Number
ISBN (eBook)
ISBN (Book)
Es handelt sich bei dieser Arbeit um eine MPhil Dissertation für die es keine Noten gibt.
autism, asd, social timing, social, social behaviour, social behavior, synchrony, social synchrony, time series analysis, developmental, preschoolers, children
Quote paper
Anne-Katrin Muth (Author), 2017, Social Timing in Preschoolers with Autism Spectrum Disorder, Munich, GRIN Verlag,


  • No comments yet.
Look inside the ebook
Title: Social Timing in Preschoolers with Autism Spectrum Disorder

Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free