See what I see,do as I do: Promoting joint attention and imitation in preschoolers with autism spectrum disorder

Abstract

Since imitation and joint attention are both important abilities for young children and since children with autism spectrum disorder show a range of problems in these domains, imitation and joint attention are important targets for intervention. In this study, we examined the possibility of promoting imitation and joint attention by means of a training programme specifically designed for low-intensity, non-residential treatment. Two matched groups of 18 children each participated in the study. The experimental group, receiving the training programme, improved significantly more on joint attention than the group receiving only treatment as usual. Only the experimental group obtained a significantly higher imitation score during the post-test compared to the pre-test. This study shows that it is possible to promote joint attention with a low-intensity treatment programme. The results concerning imitation are more modest. Future replications should involve measures of stability and generalization.

Keywords

autism spectrum disorder imitation intervention joint attention preschoolers rehabilitation

Introduction

In the first year of their life, before children are able to talk, they already communicate with their caregivers in a social, non-verbal way. Two abilities playing a prominent role in achieving this are imitation and joint attention. Imitation, defined as copying a model’s gestures, actions or words, seems to be present early in life (Heimann, 2002; Meltzoff and Moore, 1977; but see Anisfeld, 1996). In the infant, mutual imitation supports the continuation of mother–infant interaction (Kugiumutzakis, 1999). Imitation is also a strong means of communication in the interaction with peers: preverbal children will easily establish contact by imitating the play of the others (Nadel, 2002; Nadel-Brulfert and Baudonnière, 1982). By means of imitation, deferred imitation and recognition of being imitated, children increasingly gain experience about other people and the world, and gradually, they become aware of the mental states of other people, such as their intentions (Meltzoff and Moore, 1999, 2002). The early ability to imitate has also been linked to later language development: Charman et al. (2000) reported that object imitation at 20 months significantly predicted expressive language at 44 months in a group of typically developing children. In addition, Young et al. (2011) showed that increases in bodily, vocal and simple object imitation between 12 and 24 months were significantly related to expressive language and social engagement. In sum, imitation both serves a social and a learning function (Ingersoll, 2008).

In young children with autism spectrum disorder (ASD), the ability to imitate is usually more or less impaired (Williams et al., 2004). Especially non-meaningful gestures and novel object manipulations without an additional sensory effect seem to be difficult for children with ASD (Roeyers et al., 1998). An imitation impairment already discriminates infants who will later be diagnosed with ASD from infants without the disorder as early as 12 or 18 months of age (Robins et al., 2001; Young et al., 2011). The imitation problems in autism seem, however, to reflect a specific developmental delay rather than an absolute deficit: Sometimes simple imitation tasks do not discriminate between children with and without ASD, and older children with ASD have been reported to imitate (Charman and Baron-Cohen, 1994; Libby et al., 1997; Morgan et al., 1989). In addition, the growth trajectory of imitation between 12 and 36 months in children with ASD seems to parallel that of children without the disorder, albeit at a lower level (Young et al., 2011). It has been hypothesized that early imitation problems of children with ASD may lead to a weaker development of later social-communicative skills (Rogers and Pennington, 1991). Indeed, the presence of imitation seems to be predictive of more positive outcomes in children with ASD: Similar to typically developing children, a positive association has been found between imitation and subsequent language and play development in children with ASD (Charman et al., 2003; Stone et al., 1997; Stone and Yoder, 2001).

Next to imitation, a second pivotal skill in the early social-communicative development of infants is joint attention. Joint attention is the triadic coordination of attention between the child, a second person and a third event, object or person (Bakeman and Adamson, 1984). It implies more than just co-looking: both parties are aware of the focus of attention of the other person and therefore experience the mutuality of attention (Tomasello, 1995). In general, a distinction is made between imperative joint attention (or request), which serves an instrumental function, and declarative joint attention, which has the social function of sharing an interest in something with someone. In addition, a child can initiate (active) joint attention or follow the joint attention bid (passive joint attention) of someone else. In normal development, joint attention initially emerges in the form of gaze following (passive) in the second half of the first year of life (Butterworth, 1995; Corkum and Moore, 1998), or even earlier (Hoehl et al., 2009; Hood et al., 1998; but see Farroni et al., 2000). Shortly thereafter, by the age of 9 to 12 months, infants also actively try to involve others in bouts of joint attention, by gaze alternation, pointing and/or vocalizing (Tomasello, 2001). Sharing attention offers the child a possibility to share experiences and emotions with another person, meanwhile building and maintaining a relationship with that person. In addition, several authors have found a longitudinal association between early joint attention skills and later language and theory of mind (Markus et al., 2000; Morales et al., 1998; Tomasello and Farrar, 1986). For instance, Charman et al. (2000) found a significant association between joint attention skills at 20 months and theory of mind at 44 months. More recently, De Schuymer et al. (2011) reported longitudinal associations between gaze following at 9 and 14 months, initiating joint attention and behavioural requests at 14 months and receptive and expressive language at 30 months, both in typical and preterm children.

It is generally accepted that joint attention is, to a certain extent, impaired in children with ASD (see Bruinsma et al., 2004, and for reviews, see Charman, 1998). However, much depends on the form and the function of the behaviour and on the degree to which the child monitors and regulates the attention of the other person. For instance, declarative joint attention seems more impaired than imperative joint attention (e.g. Mundy et al., 1986, 1994; Sigman and Ruskin, 1999; Stone et al., 1997). The absence of pointing and showing discriminates infants with ASD from typically developing infants, already at their first birthday (Osterling and Dawson, 1994). Furthermore, an 18-month-old infant lacking declarative joint attention and symbolic play has a very high chance of being diagnosed with ASD (Baron-Cohen et al., 1992). Declarative, but not imperative, joint attention has been associated with ASD symptom severity (Mundy et al., 1994), concurrent language abilities (Bono et al., 2004; Charman et al., 2003; Dawson et al., 2004; Sigman and Ruskin, 1999) and social competence with peers (Meek et al., 2012).

In sum, imitation and joint attention are both present quite early in typical development, are both impaired or delayed in young children with ASD and seem to be positively associated with later development, both in typically developing children and children with ASD. Therefore, both abilities have been suggested to be important treatment goals in ASD (e.g. Ingersoll, 2008; Kasari et al., 2010; Mundy and Crowson, 1997; Poon et al., 2012; Rogers, 1999).

Several comprehensive programmes developed for young children with autism include imitation as an important treatment goal (e.g. the Walden Toddler Program; McGee et al., 2000). Once established, imitation is often used as a means to teach the children other skills. Ingersoll (2008) distinguished this learning function of imitation from its social function. She stressed that treatment of children with autism should also explicitly address this social function, since children who learn to imitate in a structured situation do not generalize this to spontaneous imitation in other settings (Whiten and Brown, 1998). However, it seems that spontaneous imitation in a social context is more strongly related to social reciprocity and symbolic play than imitation in a structured context, although both forms of imitation are related to vocabulary development (Ingersoll and Meyer, 2011).

Like imitation, joint attention (in different forms and functions) is often included as a treatment target in comprehensive treatment programmes, such as the Early Start Denver Model (Rogers and Dawson, 2009). In addition, several studies have reported results of more specific interventions, where joint attention was targeted as an isolated skill or one of a limited number of skills (e.g. Drew et al., 2002; Kasari et al., 2006). Although it is recommended to directly target joint attention in autism interventions in order to obtain an increase in joint attention behaviour (Mundy, 1995; Whalen and Schreibman, 2003), a recent review by White et al. (2011) suggests that joint attention can also increase when not directly targeted (e.g. as a collateral outcome in play interventions). This may be especially the case for responding to joint attention (Meindl and Cannella-Malone, 2011). Successful teaching or promoting of joint attention often occurs in the context of play, combining a developmental approach with behavioural techniques, such as reinforcement (White et al., 2011). Meindl and Cannella-Malone (2011) report that a large number of studies have used tangible reinforcers or access to preferred activities to teach initiation of joint attention. They note that this may result in requests instead of declarative joint attention behaviour, and instead recommend the use of social attention or social interaction as the main reinforcer.

In addition to the evidence that imitation and joint attention can, to a certain extent, be taught to or promoted in young children with ASD, there are also indications that the effect of these interventions may expand to other social-communicative abilities and language skills (Ingersoll and Schreibman, 2006; Jones et al., 2006; Kasari et al., 2008; Whalen et al., 2006). Therefore, following international tendencies, during the last decade, the importance of stimulating these abilities has been recognized in Belgium as well.

In Flanders (the Dutch-speaking part of Belgium), preschoolers with ASD often go to regular schools, unless they have a (severe) intellectual disability. In addition to their regular education, they are treated in non-residential, multidisciplinary rehabilitation centres. This treatment usually takes 3–5 h per week. Given the benefits of early diagnosis and intervention for both the children with ASD and their families (e.g. Corsello, 2005; Filipek et al., 1999; Wetherby et al., 2004), and given the pivotal role of imitation and joint attention in development, we developed a small-scale intervention programme with a focus on both imitation and joint attention, tailored to the needs and possibilities of these rehabilitation centres. During the past decade, several effective intervention programmes have been described (see White et al., 2011 for an overview of joint attention programmes, and Ledford and Wolery, 2011 for an overview of imitation programmes). However, we had several reasons for developing a new intervention programme. First, most of the internationally available treatment programmes for ASD involve many more hours per week and are designed for children who do not attend school. Second, some of the treatments used in previous research were not commercially or freely available or required a specific and time-consuming training. Third, we found no programme available that explicitly targeted imitation and joint attention in a limited number of sessions, and that could be done by general professionals. Therefore, we decided to develop new intervention sessions, taking into account both the practical situation in Flanders and the recommendations from international research. Our primary research goal was to test whether it is possible to promote imitation and joint attention in preschoolers with ASD in a limited time frame, with a relatively low treatment intensity.

Methods

Participants

A total of 48 children participated in the study. They were recruited through 10 local rehabilitation centres. The inclusion criteria for the study were as follows: having a chronological age between 3 and 7 years, a mental age below 7 years and a basic language understanding (sufficient to understand simple requests and instructions). The maximum mental age of 7 years was chosen because the intervention and its materials were most suited for preschoolers’ developmental levels. In addition, since imitation and joint attention seem to improve with increasing chronological and mental age, the need for imitation and joint attention training is the most urgent in or before the preschool years. All children had a diagnosis of ASD (either autism or pervasive developmental disorder–not otherwise specified (PDD-NOS) according to Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; American Psychiatric Association (APA), 2000) criteria). These diagnoses were not confirmed by the researchers but made by experienced multidisciplinary teams in the community. All parents signed an informed consent prior to the participation of their children. The children were matched pairwise on pre-test imitation/joint attention score (accepted range (AR) from +5 to −5 points), age (AR from +6 to −6 months) and full-scale IQ (FSIQ; AR from +5 to −5 points). In addition, we tried to match groups as well as possible on language age and gender. IQ and language age were abstracted from the children’s records. IQ of the children was assessed by the community teams by means of the Wechsler Preschool and Primary Scale of Intelligence–Revised (WPPSI-R; Wechsler, 1989, Flemish norms by Vander Steene and Bos, 1997), and language age was assessed by the community teams by means of the Flemish version of the Reynell Developmental Language Scales (RDLS, Flemish version by Schaerlaekens et al., 1993) and/or the Schlichting Test for Language Production (Schlichting et al., 1998). IQ and diagnostic assessment were done by the community teams within the year prior to the start of the intervention (mean time interval = 9.05 months, standard deviation (SD) = 3.28). Language assessment was done on average 6.55 months (SD = 3.79) before the start of the intervention. We estimated the child’s language age at the start of the intervention by calculating language age at the time of the language assessment/chronological age at the time of language assessment × chronological age at the time of pre-testing. Of each pair of children, one child was randomly assigned to the experimental group, and the other was included in the control group. In order to make sure that the experimental group was spread evenly over the participating centres, we controlled for centre during randomization. This was done by, as much as possible, assigning every other child from the same centre to the experimental group, resulting in about half of the children in every centre receiving the intervention. The majority of the therapists involved in the study treated both children in the intervention group and children in the control group. Parents were not explicitly told the group membership of their child, but they were informed of treatment status and possible progress after the study ended. However, since a substantial number of parents regularly followed the treatment sessions of their children, it was not possible to keep these parents completely blind for treatment content. Three children did not finish the entire training. In two children, the video of the pre- or post-test did not allow for reliable coding. An additional child obtained a ceiling score on the pre-test. These six children were not included in the final sample. In addition, the six children who were matched pairwise to these children were also excluded from the sample. Details about the 36 remaining children are shown in Table 1. Independent samples t-tests confirmed that the groups did not differ on the pre-test characteristics (all p > .50, except for verbal IQ (VIQ) with a p > .30).

Table 1.

Participant characteristics.

	Experimental group (n = 18)	Control group (n = 18)	t
Age (years)
M (SD)	5.72 (0.59)	5.74 (0.72)	.086
Range	4.70–6.80	4.07–6.92
Language age (years)
M (SD)	4.27 (1.12)	4.48 (0.70)	.583
Range	2.00–6.00	2.90–5.80
FSIQ
M (SD)	78.94 (15.49)	76.86 (16.79)	−.359
Range	50.00–103.00	50.00–105.00
VIQ
M (SD)	71.86 (13.55)	79.33 (14.55)	1.050
Range	53.00–91.00	53.00–101.00
PIQ
M (SD)	79.38 (16.19)	77.66 (16.36)	−.216
Range	52.00–97.00	56.00–110.00
Pre-test score JA
M (SD)	1.46 (0.60)	1.65 (0.60)	.928
Range	0.50–2.50	0.67–2.67
Pre-test score imitation
M (SD)	3.12 (.70)	3.16 (0.65)	.186
Range	2.07–4.20	2.02–4.07
Pre–post interval (days)
M (SD)	150.67 (39.80)	142.24 (40.76)	−.619
Range	103.00–201.00	92.00–203.00
Sex ratio (M:F)	14:4	13:5	χ ² = .148

SD: standard deviation; FSIQ: full-scale IQ; VIQ: verbal IQ; PIQ: performance IQ; JA: joint attention.

Procedure

After the children were assigned to their groups, a research assistant pre-tested them at the rehabilitation centre. The procedure of this pre-test is described below. During this appointment, the child’s usual therapist (psychologist, speech-language therapist or special educationalist) received the training package. This package consisted of the written training sessions, the materials needed for the exercises and an extensive manual for the trainer, in which the rationale for the training was explained, possible problems were explored and guidelines for the trainer’s behaviour and degrees of scaffolding were provided. The trainers were encouraged to contact the first author in case of questions. Most of the trainers also tried out previous versions of the training and/or attended a 1-day seminar on imitation and joint attention given by the first author. As soon as possible after the pre-test, trainers started the intervention with the child. The goal was to deliver two 30-min sessions per week to the child, although this was not always achieved due to issues such as absence or illness of the trainer. The missed sessions were never skipped but always postponed, resulting in a mean pre–post interval of 4.5–5 months, instead of the 3.5–4 months we estimated in advance. The sessions replaced two sessions of the regular treatment of the child. The children in the control group continued with their regular treatment. Since children receive comparable hours of treatment in the different rehabilitation centres and since all centres had children in the experimental as well as the control group, both groups of children received very similar amounts of treatment (i.e. 3–5 h per week, of which about 1 h of the specific imitation/joint attention training for the children in the experimental group). Given that established intervention models such as Applied Behaviour Analysis (e.g. Lovaas, 1987) or Treatment and Education of Autistic and related Communication Handicapped Children (TEACCH; Schopler et al., 1984) are not widespread enough in Flanders to serve as comparison treatment and due to other methodological (same amount of treatment and same therapists in both groups), practical (no possibilities to increase the number of treatment hours in both groups) and ethical reasons (no forced waiting list), we judged that this design would be the most appropriate one in the current conditions, yielding the most valid and reliable results. After the last session was completed, the child’s imitation and joint attention skills were assessed by a research assistant naïve to group membership of the child, following the same protocol that was used during the pre-testing. Control children were post-tested after a comparable time interval (see Table 1 for details).

Pre–post assessment

Although there are several protocols available for testing imitation and/or joint attention, to our knowledge, they all measure only part of the abilities we targeted in the intervention, or they cannot be scored in sufficient detail to allow for a measure of short-term change. Therefore, we designed a new pre–post assessment, combining and adapting existing methods. The tasks were aimed at assessing a wide range of imitation and joint attention abilities, both elicited and spontaneous. The assessment was always conducted in a quiet room at the child’s rehabilitation centre, with experimenter 1 and the child sitting at a table, facing each other. A second experimenter was present to videotape the interaction and to activate the toys used for eliciting declarative joint attention. The assessment was semi-standardized, meaning that the experimenter always conducted the tasks in the same way and with the same materials, but that some adjustments in timing and order of the tasks were allowed when needed. Table 2 summarizes the items administered during the test, as well as the scoring criteria. The tests were administered and simultaneously scored by three trained research assistants, naïve to group membership. Each test was videotaped, so that scoring could be verified afterwards. A total of 11 test protocols were scored by all three coders (based on the video). Inter-rater reliability, expressed in kappa (Cohen, 1960), varied between .60 and .93 for the joint attention scores and between .72 and .85 for imitation scores. These are good to excellent scores (Cohen, 1960). The number of spontaneous imitations could not be scored reliably. This was mainly due to the difficulty of describing unequivocal coding rules that applied to the large variety of imitated actions.

Table 2.

Pre–post items and scoring criteria.

Item (number of times item was administered during the test)	Description	Scoring	Max
Reaction to ambiguous behaviour: blocking (3), teasing (3) (based on Phillips et al., 1992)	Administered during free play interaction. Blocking: placing hands over child’s hands for 5 s or until eye contact. Teasing: offering and then withdrawing a toy for 5 s or until eye contact	Eye contact + verbal or non-verbal communication = 1 point	6
Gaze following (2) (based on Mundy et al., 1993)	Making eye contact with child and then looking (trials 1 and 2) or pointing (trials 3 and 4) to toy (at 90° and distance of 1.5–2 m from child)	Looking at target in first trial = 3 points, second trial = 2 points, third trial = 1 point	6
Initiating requests (3) (based on Stone et al., 1997)	Giving puzzle with missing pieces or closed container with toys to the child and waiting for 15 s or until request	Eye contact + verbal or non-verbal communication = 1 point	3
Following requests (2) (based on Mundy et al., 1993)	Asking child for a toy (‘can I have the …’ in trials 1 and 2, ‘give me the …’ in trials 3 and 4)	Following request at first trial = 1 point	2
Initiating declarative joint attention (2) (based on Charman et al., 1997)	Distant toy is activated by second experimenter. Not looking at the toy for 15 s or until joint attention bid of the child	Eye contact + verbal or non-verbal communication = 1 point	2
Spontaneous joint attention (any)	Any example of spontaneous (not elicited) joint attention is scored, may be declarative or requesting	1 point for every declarative or requesting initiative combined with eye contact	any
Gestural imitation (8) (based on Uzgiris and Hunt, 1975)	Variation of visible/invisible, familiar/unfamiliar gestures, for example, putting fingers in ears (invisible familiar). Demonstrating 3 times, encouraging imitation, waiting for 15 s or until imitation	Correct imitation = 1 point	8
Verbal imitation (10) (based on Uzgiris and Hunt, 1975)	Variation of sound, words, non-words and short sentences. Demonstrating 3 times, encouraging imitation, waiting for 15 s or until imitation	Correct imitation = 1 point	10
Object imitation (6) (based on Meltzoff, 1985, 1988 and Roeyers et al., 1998)	Variation of actions on objects: with visual effect, auditory effect or no added effect. Demonstrating 3 times, encouraging imitation, waiting for 20 s or until imitation	Correct imitation = 1 point	6
Symbolic imitation (3 scripts with 5 actions and 6 vocalizations each) (based on Libby et al., 1997)	Script with miniatures (giving doll a drink), replacement objects with no clear function (doll, bath, soap and towel represented by blocks) and replacement objects with other function (shoe as bed, truck as baby, yoyo as pillow and so on)	1 point per imitated action	15
		1 point per imitated vocalization	18
Spontaneous imitation (any)	Any spontaneous imitation is coded	1 point for every spontaneous imitation (gestural, vocal or object)	any

Max: maximal score that could be obtained during one test.

Recommended order of the items: free play with blocking (3 times) and teasing (3 times), gaze following 1, initiating requests 1, initiating declarative 1, following requests 1, gaze following 2, initiating requests 2, initiating declarative 2, following requests 2, gestural imitation, verbal imitations, object imitation, symbolic imitation.

The intervention

The intervention was developed over a time-course of several years. During the development of the training, there were two large process evaluations. These evaluations involved the trainers filling in a form for each session and an overall evaluation. The intervention was evaluated on feasibility, degree of difficulty, appeal, duration, variation, completeness and trainer support. In addition, by means of open-ended questions, we invited the trainers to provide us with as much feedback as possible. After every evaluation, the training was adapted to meet as many of the comments and questions as possible (e.g. changing the order of the sessions and exercises within the sessions, making exercises more enticing to the child and suggesting other materials). The current intervention consists of 24 sessions, each lasting about 30 min. Every session includes imitation as well as joint attention exercises. The sessions gradually increase in difficulty, based on the spontaneous development of both typically developing children and children with ASD. Gestural and vocal imitation and object imitation are presented alternately, in order to keep the child motivated, with a particular focus on the social function of imitation. Symbolic imitation is promoted in the last few sessions, but only for children who obtained a sufficiently high level of procedural imitation. Joint attention tasks evolve from following and initiating requests to following and eventually initiating declarative joint attention. However, because initiating joint attention is the most difficult behaviour to elicit and the possibilities to elicit it are usually quite short (e.g. an unexpected event), the possibility to initiate joint attention is included in most of the sessions. In all joint attention exercises, the reward is as natural and as social as possible. An overview of the abilities targeted in the 24 sessions is provided in Table 3. However, many of the tasks and games allowed for the stimulation of other abilities as well. This was also explicitly mentioned in the descriptions. The exercises involve a combination of behavioural and developmental techniques, such as modelling familiar acts in the first sessions and gradually moving on to more unfamiliar or novel actions (Responsive Imitation Training; Ingersoll et al., 2007; Ingersoll and Schreibman, 2006), following the child’s lead (Pivotal Response Training; Koegel and Koegel, 2006), embedding the exercises in typical activities (Incidental Teaching; McGee et al., 1999) and an emphasis on reciprocity (Responsive Teaching; Mahoney and Perales, 2003). A lot of attention is paid to motivating the children, for instance, by designing exercises that can be done with different materials (paint, pencils, crayons, play-dough etc.), depending on the child’s preference, by making the activities rewarding by themselves (e.g. mixing colours in a specific order and then being allowed to paint with it), by imitating the child and encouraging the child to model an action, by embedding the imitation of actions that are less rewarding by themselves within a favourite board game and by using a variety of reinforcers such as social (praise, adapted to preference and tolerance of the child), tangible (stickers) and natural consequences (access to preferred toys or activities). Examples of the exercises and games are provided in Table 4. Finally, we built in a fair amount of flexibility in timing, amount of support provided to the child, interaction style, amount of repetition, materials and so on, in order for the therapists to adapt the training to the needs and interests of each child and to make sure these experienced therapists would also feel comfortable with the training themselves. Together with the intervention sessions, the trainers received general instructions on the goal of the intervention, where for every skill, the hierarchy of possible support given and possible reinforcement were described. This is summarized in Table 5.

Table 3.

Intervention: overview of skills addressed in different sessions.

Session number	Skills addressed
1	Bodily imitation, initiating requests, gaze and point following, initiating declarative joint attention
2	Imitation of facial expressions (including social imitation), vocal imitation, initiating declarative joint attention, following requests, gaze and point following
3	Bodily imitation (including social imitation), initiating requests, gaze and point following, initiating declarative joint attention
4	Object imitation (with visual and/or auditory effect), initiating requests, initiating declarative joint attention, gaze (and point) following
5	Gestural and vocal imitation (including social imitation), initiating requests, initiating declarative joint attention, gaze (and point) following
6	Object imitation (auditory effect), initiating requests
7	Object imitation (without additional sensory effects), following requests, initiating declarative joint attention, gaze (and point) following
8	Social imitation (gestural, vocal, objects), initiating declarative joint attention, gaze (and point) following
9	Object imitation (visual effect), initiating requests, initiating declarative joint attention, gaze (and point) following
10	Object imitation (without additional sensory effects), initiating declarative joint attention, gaze (and point) following, initiating requests
11	Object imitation (auditory effect), initiating declarative joint attention
12	Object imitation (without additional sensory effect), initiating declarative joint attention, gaze following
13	Social imitation (gestural, vocal, objects), initiating declarative joint attention
14	Initiating declarative joint attention (two games), initiating requesting
15	Object imitation (immediate and deferred), initiating declarative joint attention
16	Object imitation, symbolic imitation (realistic miniatures), initiating declarative joint attention (two games)
17	Symbolic imitation (realistic miniatures), gaze following, initiating declarative joint attention (two games)
18	Symbolic imitation (scripts with doll), initiating declarative joint attention (two games)
19	Symbolic imitation (neutral replacement objects), initiating requests, initiating declarative joint attention
20	Symbolic imitation (neutral replacement objects), initiating requests, initiating declarative joint attention
21	Object imitation (without additional sensory effect), initiating requests, initiating declarative joint attention
22	Symbolic imitation (script with puppets), initiating declarative joint attention (two games)
23	Imitation (trainer’s choice of games from previous sessions), initiating requests, initiating declarative joint attention
24	Imitation (trainer’s choice of games from previous sessions), initiating declarative joint attention (two games)

Table 4.

Intervention: examples of games/tasks.

Primary skill	Example
Following requests	Session 2: Farm game. Trainer and child play with farm and animals for about 8–10 min. At a rate of about once per minute, the trainer asks for a specific animal. The trainer uses the different support steps as described in Table 5 and reinforces the response of the child by making the sound of the animal and giving something back.
Initiating requests	Session 3: Blowing bubbles. Trainer and child have a bubble blower. Trainer blows bubbles, waits. If child indicates it wants more bubbles, trainer blows more bubbles. Trainer encourages child to try for himself. If the trainer thinks the child will not be frustrated, she gives a bubble liquid without soap to the child, and the child has to indicate something is wrong.
Gaze and point following	Session 1 and following: Teddy. Before the child enters the room, Teddy is placed somewhere within the visual field of the child. If the child did not notice Teddy by the end of the session, the trainer attends the child to it by gazing and pointing (see also Table 5 for possible levels of support). The Teddy exercise gradually evolves into an initiating joint attention exercise (where we expect the child to spontaneously look for and show Teddy).
Initiating declarative joint attention	Session 2 and following: Teddy (see above for initial description). When the routine of looking for Teddy is installed, Teddy can suddenly wear a scarf or funny hat. Trainer waits until child notices, or uses support techniques from Table 5.Session 18: Alarm or song. Before the child enters the room, the trainer arranges for a cell phone alarm to go off or a song starting to play somewhere in the middle of the session. The trainer does not react to the sound, unless the child makes some attempt to involve her in it.
Imitation (including social imitation)	Session 8: Game of goose. The game has several special spaces. If the trainer or the child land on that space, they have to imitate something the other person models. This can be gestural, vocal, with objects or symbolic, depending on the imitation level of the child. Cards with things that can be modelled are available, but are not obligatory. If the child/trainer imitates correctly, he or she can roll the dice again. The game stimulates role-taking in imitation and also stimulates the child to think of creative actions to model, making imitation for the trainer more difficult (if they understand the competitive nature of the game). The difficulty level of the game is adapted to the level of the child

Table 5.

Intervention: overview of general instructions.

Section	Description
Introduction	Definition and description of imitation and joint attention behaviour, needed and provided materials, general instructions for flexibility in level of support (as much as needed to obtain the desired or ‘best possible’ behaviour from the child, start every next try with decreased support and increase when needed)
Following requests	Goal of the exercises (obtain object from child), possible levels of support (e.g. naming, pointing, asking, reaching), possible reinforcement (e.g. saying thank you, giving something back)
Initiating requests	Goal of the exercises (having child ask for help or object with verbal or non-verbal communication together with eye contact), possible levels of support (e.g. repeating what child wants to do (e.g. colouring), non-specific questions (‘Can I help?’), asking specific questions (‘Do you need a pencil?’), waiting, helping, giving), possible reinforcement (usually inherent to the task/game, for example, getting a pencil).
Gaze and point following	Goal of the exercises (having the child spontaneously following your attention), possible levels of support (e.g. calling child, touching child, gaze shift, pointing, showing object), possible reinforcement (e.g. giving object of attention, preferred game)
Initiating declarative joint attention	Goal of the exercises (having the child direct your attention to something of interest (not requesting) by means of verbal or non-verbal communication and eye contact), possible levels of support (e.g. waiting, repeating sound or action), asking non-specific questions (e.g. ‘did I just hear something?’), possible reinforcement (e.g. preferred activity, game, social reinforcement) + ideas for eliciting declarative joint attention in many of the other tasks/games, for example, by saying or doing something unexpected (e.g. replacing a ball by a block in a ball game, putting on a funny hat while the child is not looking and so on)
Imitation	Goal of the exercises (having the child do the same as you do, eventually also delayed), possible levels of support (e.g. explicit instructions, physical support), possible reinforcement (e.g. winning points in a game, collecting cards, social reinforcement)
Social imitation	Explaining the social function of imitation, possible ways to stimulate this (e.g. encouraging the child to take roles and to model, imitating the child regularly with clear expression of positive affect, reinforcing the child for modelling)

Treatment as usual

The control group received the treatment as usual. Recently, a survey was conducted in the Flemish centres treating young children with autism (Van der Paelt et al., 2012). The most frequently mentioned treatment goals were communication, social skills, language, play and motor skills. Imitation was mentioned as a treatment goal in only 6% of the respondents, and joint attention in only 2%. No specific programmes targeting imitation and/or joint attention were used. In this study, the therapists delivering the treatment to the control group were explicitly asked not to use exercises or techniques from the intervention package and not to change the focus and goals of the treatment as usual based on their knowledge of the intervention package. They were explained the reasons for not doing so, and they were given the option to use the intervention, or parts of it, after the post-testing of the control group was done.

Data analysis

In addition to the item scores displayed in Table 2, two composite scores were calculated. The Total Joint Attention score is the sum of the scores on reaction to ambiguous behaviour, gaze following, initiating requests and initiating declarative joint attention (see Table 2 for a description of these tasks). Following requests was left out of this total score, since a majority of the children obtained a maximal score on this item (23 out of 36 children on pre-test). Since we wanted all items to contribute equally to the total score, we summed the proportional scores (score of the child/maximal score of the item) instead of the absolute values. Consequently, this joint attention test score has a maximum value of 4. Similarly, a Total Imitation score was computed as a sum of gestural, vocal and object imitation, the number of actions imitated during symbolic imitation and the number of vocalizations imitated during symbolic imitation (see Table 2). Again, we calculated and summed the proportional scores of the five imitation variables, resulting in a maximum value of 5.

To answer the question whether the experimental group has made more progress than the control group, we performed repeated measures analyses of variance (ANOVAs) with pre–post score as the within-subjects factor and group as the between-subjects factor. However, since the control group also received autism-specific treatment, we additionally analysed progress for each group separately. This was done by paired-samples t-tests. Given that Bonferroni and related corrections strongly increase the chance of making Type II errors and that the Bonferroni method is primarily designed for testing the general null hypothesis (for an overview of these arguments, see Nakagawa, 2004; Perneger, 1998), and since in our study we test specific hypotheses and the chance of making Type II errors is already high due to our small and heterogeneous sample and due to small to medium expected effect sizes, we decided not to perform a correction for multiple testing. Finally, we also explored which subject characteristics possibly contributed to the child’s progress. This was done by Pearson’s correlations.

Results

Did the experimental group make more progress than the control group?

The group × pre–post interaction factor was significant for the Total Joint Attention score, indicating that the experimental group made more progress than the control group (F(1, 34) = 9.341, p < .01). Within the specific joint attention tasks, this was especially the case for gaze following (F(1, 34) = 6.507, p < .05) and initiating requests (F(1, 34) = 3.976, p < .05). Initiating declarative joint attention significantly decreased in both groups between pre- and post-test (F(1, 34) = 6.629, p < .05).

No significant group × pre–post interaction was found on the domain of imitation. There was, however, a main effect of time, with a significant pre–post gain for both groups combined, for Total Imitation, object imitation and number of imitated actions during symbolic imitation (F(1, 34) = 16.635, p < .001; F(1, 34) = 5.073, p < .05; and F(1, 34) = 15.667, p < .001, respectively). See Table 6 for exact scores and results of the ANOVAs.

Table 6.

Pre–post scores, results of the repeated measures ANOVAs and effect sizes ( $η_{p}^{2}$ ) of the interaction effect.

	Experimental group (n = 18)		Control group (n = 18)		F(1, 34) group	F(1, 34) pre–post	F(1, 34) interaction	$η_{p}^{2}$
	Pre-test	Post-test	Pre-test	Post-test
Total Joint Attention	1.46 (.60)	1.81 (.73)	1.65 (.60)	1.24 (.56)	1.264	.070	9.341**	.22
Ambiguous behaviour	.21 (.21)	.24 (.23)	.17 (.18)	.08 (.15)	3.937	.466	1.863	.05
Gaze following	.61 (.19)	.82 (.22)	.69 (.16)	.67 (.29)	.442	3.851	6.507*	.16
Initiating requests	.33 (.36)	.57 (.38)	.31 (.31)	.30 (.25)	2.808	2.921	3.976*	.11
Initiating declarative JA	.31 (.39)	.17 (.30)	.47 (.44)	.19 (.30)	1.188	6.629*	.737	.02
Spontaneous declarative JA	.89 (.96)	1.72 (2.19)	.67 (1.33)	.78 (1.00)	2.096	3.023	1.768	.05
Total Imitation	3.12 (.70)	3.64 (.61)	3.16 (.65)	3.42 (.54)	.227	16.635***	1.870	.05
Gestural imitation	.67 (.19)	.69 (.13)	.67 (.25)	.67 (.29)	.026	.077	.077	.00
Verbal imitation	.85 (.27)	.95 (.10)	.90 (.16)	.88 (.25)	.026	.891	1.746	.05
Object imitation	.54 (.19)	.62 (.24)	.55 (.15)	.67 (.15)	.433	5.073*	.168	.01
Symbolic imitation: actions	.69 (.25)	.88 (.13)	.70 (.31)	.86 (.11)	.009	15.667***	.089	.00
Symbolic imitation: vocalizations	.37 (.30)	.51 (.29)	.35 (.30)	.35 (.34)	1.056	1.895	2.080	.06

JA: joint attention

p ≤ .05

p < .01

***

p < .001.

Did each group make progress?

The experimental group showed a significant gain in the Total Joint Attention score (t(17) = −2.066, p ≤ .05). This was also seen on the item scores of gaze following, initiating requests and spontaneous declarative joint attention (t(17) = −2.848, p < .05; t(17) = −2.404, p < .05; and t(17) = −2.093, p ≤ .05, respectively). Children who received the targeted intervention also had significantly higher Total Imitation scores after the training, compared to before (t(17) = −3.976, p < .001). This was largely due to an improvement in the number of symbolic imitation actions (t(17) = −3.169, p < .01).

The children in the control group had a lower joint attention score on the post-test compared to the pre-test (t(17) = 2.252, p < .05), but this was largely due to the significant decrease in their initiating declarative joint attention (t(17) = 2.263, p < .05). The other joint attention scores for the most part remained stable. The control group showed no significant improvement on the Total Imitation score (t(17) = −1.860, p = NS), but they did, however, obtain higher scores on the second testing for object imitation and number of symbolic imitation actions (t(17) = −2.600, p < .05 and t(17) = −2.470, p < .05, respectively).

Which participant characteristics may contribute to progress?

To explore the relation between child characteristics and individual progress, we calculated Pearson correlations between the children’s age, verbal mental age, FSIQ, VIQ, performance IQ (PIQ) and imitation and joint attention pre-test score on the one hand, and their progress in imitation and joint attention (expressed as a difference score between post- and pre-test) on the other. In the experimental group, VIQ was significantly positively correlated with progress in imitation (r = .767, p < .05). There were no other significant correlations.

In addition, we explored the progress of subgroups of children based on their FSIQ or age. In the experimental group, the children with an average FSIQ (85–115; n = 8) on average improved .52 points on joint attention and .61 points on imitation. The children with a mild intellectual disability (FSIQ = 50–70; n = 7) improved on average .31 points on joint attention and .32 points on imitation. This difference was not significant (F(2, 12) = .409, $η_{p}^{2}$ = .064). There were only three children with borderline intellectual functioning (FSIQ = 71–84), so there were no statistical tests possible with this group. Similarly, the children younger than 5.5 years (n = 7) and older than 5.5 years (n = 11) in the experimental group improved equally much (F(2, 15) = .225, $η_{p}^{2}$ = .029).

Discussion

Since imitation and joint attention are both important abilities for young children and since children with ASD have well-documented problems on these domains, imitation and joint attention are important targets for intervention. In this study, we examined the possibility of promoting imitation and joint attention by means of a training programme specifically designed for low-intensity, non-residential treatment. In the intervention, special attention was given to the social role of imitation. Joint attention was directly targeted, and inasmuch as possible rewarded in a social and natural way.

The training seemed to be especially successful in promoting joint attention. Not only did the children in the training group show a substantial improvement in joint attention skills, this improvement was also significantly larger than in the control group. Gaze following and requesting seemed to be most responsive to this treatment. This corresponds with the typical development of children with ASD, where gaze following and initiating requests seem to be the least impaired in older, more able children (Charman, 1998; Warreyn et al., 2007). It may also have been easier to learn these skills, since there was almost always some instrumental reward following the behaviour of the child, mostly obtaining a desired object (either by requesting or by following gaze to see where the object was hidden). Perhaps the most difficult to teach is initiating declarative joint attention, since this requires the initiative of the child and serves a social, not instrumental, function. Previous studies also show more substantial and more stable intervention effects for responding (or following) joint attention behaviour compared to initiating, and for requesting compared to declarative behaviour (e.g. Hwang and Hughes, 2000; Kasari et al., 2010). At first sight, our training did not seem very successful in promoting declaratives, since this behaviour decreased significantly from pre- to post-treatment. However, it is very likely that the children did show less elicited declarative behaviour the second time because they were not surprised anymore by the suddenly appearing objects (used to elicit declaratives). If we had used new objects for the post-test, their reaction may have been more enthusiastic. In addition, the increased number of spontaneous (non-elicited) declaratives in the experimental group does suggest a positive training effect.

The results concerning imitation progress were modest: the training group did not show more improvement than the control group. However, the children receiving the training did show a significantly higher Total Imitation score at the post-test compared to the pre-test, while the Total Imitation score of the control children remained stable. The control children, on the contrary, also showed improvement on some aspects of imitation. Although we could not control for the contents of the treatment as usual that the control children received, a recent survey conducted in Flanders showed that this is not very often an explicit treatment target (Van der Paelt et al., 2012). The progress the control group made on the domain of imitation may therefore reflect a side effect of the treatment as usual, or the natural development of these children. The observation that the experimental group, in contrast to the control group, had a significantly higher Total Imitation score at the post-test compared to the pre-test may suggest that the intervention was somewhat effective in promoting imitation. However, since the group × pre–post interaction was not significant, we cannot conclude that it was more effective than treatment as usual.

Since every training session included imitation as well as joint attention exercises and since improvement in one domain may lead to (collateral) improvement in the other (e.g. Whalen et al., 2006), it is not possible to disentangle the effects of the imitation exercises from those of the joint attention exercises. Both abilities may not be independent. However, in our study, pre- and post-scores of imitation and joint attention did not correlate significantly (pre-score: r = −.044, p = NS; post-score: r = .206, p = NS), neither did the children’s progress in the two abilities (experimental group: r = .310, p = NS; control group: r = −.254, p = NS).

We found few interrelations between child characteristics and progress. The positive correlation between VIQ before the training and imitation improvement is consistent with previous research. However, while imitation is usually seen as a positive predictor for later language development, this study shows that the influence may also be applicable in the opposite direction. There were no substantial correlations between age and improvement and, apart from the above mentioned, between IQ and improvement. In addition, we found that children in the average FSIQ group and children in the mild intellectual disability group benefited equally much from the intervention, as did the older children compared to the younger children. Given the limited number of children these analyses were performed on, we must acknowledge the possibility that the absence of significant age and IQ effects is due to a limited power. However, the correlations found between age and IQ on the one hand and imitation and joint attention improvement on the other hand were weak (all r < .30) and the effect sizes were small. This suggests that the training may be useful for the entire age range and IQ range studied (4–7 years, FSIQ = 50–103), although replication with a larger sample is recommended. Based on our findings, we cannot recommend the intervention to children with IQs above average, since there were no such children included in our sample. Nevertheless, if these children also show impairments in imitation and joint attention, we see no contraindications to use the intervention with these children on an individual basis. Given the positive results of the study, a Dutch transcript of the training session and instructions is available upon request and has been informally distributed in rehabilitation centres. It will also become commercially available.

Limitations

In this study, we were not able to assess the stability of the children’s progress. There was also no assessment of generalization to other persons or situations. We chose to include parent questionnaires in the pre–post measures, but our attempts to do so resulted in questionable data. Given that we were not able to keep all parents blind to group allocation of their child (the most obvious reason for that is that some parents regularly follow the treatment sessions of their child), it was impossible to ask parents for an unbiased evaluation of their child’s social-communicative skills. In addition, in previous pilots of the intervention, we noticed that asking parents to fill in one or more questionnaires resulted in a significant loss of participants. Therefore, this time, in order to obtain an adequately large sample, we also allowed children in the intervention if their parents did not want to fill in questionnaires. However, since the trainer (the child’s own therapist) and the tester (a research assistant) were two independent persons and the trainer was not present at post-testing, we do have some indication that the learned abilities were generalized to interactions with at least one other person. A second limitation of the study was the absence of control over the treatment of the control group. Since it would not be ethical to deprive the control children of their regular treatment, we limited ourselves to asking about treatment goals. The most frequently named goals were imitation, communication, basic social skills and behaviour management. Allowing the regular treatment to continue did, however, teach us that our training can be a valuable addition to the currently available programmes. Third, we did not confirm the children’s ASD diagnoses with the Autism Diagnostic Interview–Revised (ADI-R) (Rutter et al., 2003) or the Autism Diagnostic Observation Schedule (ADOS; Lord et al., 1999). However, all children had a community diagnosis made by an experienced multidisciplinary team according to DSM-IV-TR criteria. In addition, since both groups had ASD, it was not necessary to discriminate the groups by means of ADI-R/ADOS data. However, the lack of ADI-R/ADOS data also did not allow us to match both groups for levels of symptom severity, or to investigate the relation between autism severity and the progress the children made. Since symptom severity is not often assessed in common practice in Flanders, no other measures for these variables were available. Fourth, the current sample is not very large, probably limiting the statistical power of the analyses. In addition, as described above, 12 of the initial 48 children were not included in the sample, resulting in a relatively high attrition rate. Finally, we included no formal measure of treatment fidelity in this study. Again, during piloting of previous version of the intervention package, we noticed that due to time constraints, therapists were likely to drop out of the study if they were required to fill in documents or answer questions about their understanding and application of the training sessions and background. By corresponding with the trainers by phone and by email, by delivering the most specific materials and by explicitly asking for feedback, we were confident that the trainers followed our instructions for delivery of the intervention quite well.

A final consideration we need to make is whether, although statistically significant, the changes observed in the experimental group are clinically significant. On average, they improved .34 on the Total Joint Attention score and .52 on the Total Imitation score. This refers to an increase of 23% and 17%, respectively, compared to the pre-test score. In terms of actual behaviour, the children in the intervention group increased their number of elicited joint attention acts from 6.53 to 8.41 (+1.88), their number of spontaneous declarative joint attention acts from .89 to 1.72 (+ .83) and their number of correct imitations from 34.11 to 41.12 (+7.01). The time frame in which the joint attention acts were coded was about 10 min, and the time frame for imitation acts was also about 10 min. Given these numbers, we believe that this increase, especially in joint attention behaviour, is clinically relevant.

Conclusion

Altogether, we believe that the results of this study are promising. We have shown that it is possible to promote joint attention in a quite limited number of sessions, over a rather short period of time. The results concerning imitation were less pronounced: although the experimental group showed a significant improvement on this domain, the improvement was not significantly larger than that in the control group. Since health care puts more and more emphasis on evidence-based, cost-effective treatment, we believe that the proposed approach makes a valuable contribution to treatment of children with ASD. Future research examining long-term effects, generalization and possibly other treatment goals is desirable.

Footnotes

Acknowledgements

The authors wish to thank the centres, children and therapists for their participation and valuable feedback, and Sig vzw for their help in recruiting. The work of the authors was supported by the ESF COST Action BM1004 Enhancing the Scientific Study of Early Autism (ESSEA)

Funding

During the time of this research, the first author (P.W.) was supported by grants from the Marguerite-Marie Delacroix Foundation and VVA, the Flemish Parent Association.

References

American Psychiatric Association (APA) (2000) Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR). 4th ed., text rev. Washington, DC: American Psychiatric Association.

Anisfeld

(1996) Only tongue protruding modelling is matched by neonates. Developmental Review 16: 149–161.

Bakeman

Adamson

(1984) Coordinating attention to people and objects in mother-infant and peer-infant interaction. Child Development 55: 1278–1289.

Baron-Cohen

Allen

Gillberg

(1992) Can autism be detected at 18 months? The needle, the haystack and the CHAT. The British Journal of Psychiatry 161: 839–843.

Bono

Daley

Sigman

(2004) Relations among joint attention, amount of intervention and language gain in autism. Journal of Autism and Developmental Disorders 34: 495–505.

Bruinsma

Koegel

(2004) Joint attention and children with autism: a review of the literature. Mental Retardation and Developmental Disabilities Research Reviews 10: 169–175.

Butterworth

(1995) Origins of mind in perception and action. In: Moore

Dunham

(eds) Joint Attention: Its Origins and Role in Development. Hillsdale, NJ: Lawrence Erlbaum Associates, pp. 29–40.

Charman

(1998) Specifying the nature and course of the joint attention impairment in autism in the preschool years. Autism 2: 61–79.

Charman

Baron-Cohen

(1994) Another look at imitation in autism. Development and Psychopathology 6: 403–413.

10.

Charman

Baron-Cohen

Swettenham

. (2000) Testing joint attention, imitation, and play as infancy precursors to language and theory of mind. Cognitive Development 15: 481–498.

11.

Charman

Baron-Cohen

Swettenham

. (2003) Predicting language outcome in infants with autism and pervasive developmental disorder. International Journal of Language & Communication Disorders 38: 265–285.

12.

Charman

Swettenham

Baron-Cohen

. (1997) Infants with autism: an investigation of empathy, pretend play, joint attention, and imitation. Developmental Psychology 33: 781–789.

13.

Cohen

(1960) A coefficient of agreement for nominal scales. Educational and Psychological Measurement 20: 37–46.

14.

Corkum

Moore

(1998) The origins of joint visual attention in infants. Developmental Psychology 34: 28–38.

15.

Corsello

(2005) Early intervention in autism. Infants and Young Children 18: 74–85.

16.

Dawson

Toth

Abbott

. (2004) Early social attention impairments in autism: social orienting, joint attention, and attention to distress. Developmental Psychology 40: 271–283.

17.

De Schuymer

De Groote

Beyers

. (2011) Preverbal skills as mediators for language outcome in preterm and full term children. Early Human Development 87: 265–272.

18.

Drew

Baird

Baron-Cohen

. (2002) A pilot randomised control trial of a parent training intervention for preschool children with autism. Preliminary finding and methodological challenges. European Child & Adolescent Psychiatry 11: 266–272.

19.

Farroni

Johnson

Brockbank

. (2000) Infant’s use of gaze direction to cue attention: the importance of perceived motion. Visual Cognition 7: 705–718.

20.

Filipek

Accardo

Baranek

. (1999) The screening and diagnosis of autistic spectrum disorders. Journal of Autism and Developmental Disorders 29: 439–484.

21.

Heimann

(2002) Notes on individual differences and the assumed elusiveness of neonatal imitation. In: Meltzoff

Prinz

(eds) The Imitative Mind: Development, Evolution, and Brain Bases. New York: Cambridge University Press, pp. 74–84.

22.

Hoehl

Reid

Parise

. (2009) Looking at eye gaze processing and its neural correlates in infancy – implications for social development and autism spectrum disorder. Child Development 80: 968–985.

23.

Hood

Willen

Driver

(1998) Adult’s eyes trigger shifts of visual attention in human infants. Psychological Science 9: 131–134.

24.

Hwang

Hughes

(2000) Increasing early social-communicative skills of preverbal preschool children with autism through social interactive training. The Journal of the Association for Persons with Severe Handicaps 25: 18–28.

25.

Ingersoll

(2008) The social role of imitation in autism: implications for the treatment of imitation deficits. Infants and Young Children 21: 107–119.

26.

Ingersoll

Meyer

(2011) Examination of correlates of different imitative functions in young children with autism spectrum disorders. Research in Autism Spectrum Disorders 5: 1078–1085.

27.

Ingersoll

Schreibman

(2006) Teaching reciprocal imitation skills to young children with autism using a naturalistic behavioral approach: effects on language, pretend play, and joint attention. Journal of Autism and Developmental Disorders 36: 487–505.

28.

Ingersoll

Lewis

Kroman

(2007) Teaching the imitation and spontaneous use of descriptive gestures in young children with autism using a naturalistic imitation intervention. Journal of Autism and Developmental Disorders 37: 1446–1456.

29.

Jones

Carr

Feeley

(2006) Multiple effects of joint attention intervention for children with autism. Behavior Modification 30: 782–834.

30.

Kasari

Freeman

Paparella

(2006) Joint attention and symbolic play in young children with autism: a randomized controlled intervention study. Journal of Child Psychology and Psychiatry, and Allied Disciplines 47: 611–620.

31.

Kasari

Gulsrud

Wong

. (2010) Randomized controlled caregiver mediated joint engagement intervention for toddlers with autism. Journal of Autism and Developmental Disorders 40: 1045–1056.

32.

Kasari

Paparella

Freeman

. (2008) Language outcome in autism: randomized comparison of joint attention and play interventions. Journal of Consulting and Clinical Psychology 76: 125–137.

33.

Koegel

(2006) Pivotal Response Treatments for Autism. Baltimore, MD: Paul H. Brookes.

34.

Kugiumutzakis

(1999) Genesis and development of early infants mimesis to facial and vocal models. In: Nadel

Butterworth

(eds) Imitation in Infancy: Cambridge Studies in Cognitive Perceptual Development. New York: Cambridge University Press, pp. 36–59.

35.

Ledford

Wolery

(2011) Teaching imitation to young children with disabilities: a review of the literature. Topics in Early Childhood Special Education 30: 245–255.

36.

Libby

Powell

Messer

. (1997) Imitation of pretend play acts by children with autism and Down syndrome. Journal of Autism and Developmental Disorders 27: 365–383.

37.

Lord

Rutter

DiLavore

. (1999) Autism Diagnostic Observation Schedule Manual. Los Angeles, CA: Western Psychological Services.

38.

Lovaas

(1987) Behavioral treatment and normal educational and intellectual functioning in young autistic children. Journal of Consulting and Clinical Psychology 55: 3–9.

39.

McGee

Morrier

Daly

(1999) An incidental teaching approach to early intervention for toddlers with autism. Journal of the Association for Persons with Severe Handicaps 24: 133–146.

40.

McGee

Morrier

Daly

(2000) The Walden preschool. In: Harris

Handleman

(eds) Preschool Education Programs for Children with Autism. 2nd ed. Austin, TX: PRO-ED, pp. 157–190.

41.

Mahoney

Perales

(2003) Using relationship-focused intervention to enhance the social-emotional functioning of young children with autism spectrum disorders. Topics in Early Childhood Special Education 23: 77–89.

42.

Markus

Mundy

Morales

. (2000) Individual differences in infant skills as predictors of child-caregiver joint attention and language. Social Development 9: 302–315.

43.

Meek

Robinson

Jahromi

(2012) Parent-child predictors of social competence with peers in children with and without autism. Research in Autism Spectrum Disorders 6: 815–823.

44.

Meindl

Cannella-Malone

(2011) Initiating and responding to joint attention bids in children with autism: a review of the literature. Research in Developmental Disabilities 32: 1441–1454.

45.

Meltzoff

(1985) Immediate and deferred imitation in fourteen- and twenty-four-month-old infants. Child Development 56: 62–72.

46.

Meltzoff

(1988) Infant imitation and memory: nine-month-olds in immediate and deferred tests. Child Development 59: 217–225.

47.

Meltzoff

Moore

(1977) Imitation of facial and manual gestures by human neonates. Science 198: 75–78.

48.

Meltzoff

Moore

(1999) A new foundation for cognitive development in infancy: the birth of the representational infant. In: Scholnick

Nelson

Miller

al.

(eds) Conceptual Development: Piaget’s Legacy. Mahwah, NJ: Erlbaum Press, pp. 53–78.

49.

Meltzoff

Moore

(2002) Imitation, memory, and the representation of persons. Infant Behavior & Development 25: 39–61.

50.

Morales

Mundy

Rojas

. (1998) Following the direction of gaze and language development in 6-month-olds. Infant Behavior and Development 21: 373–377.

51.

Morgan

Cutrer

Coplin

. (1989) Do autistic children differ from retarded and normal children in Piagetian sensorimotor functioning? Journal of Child Psychology and Psychiatry, and Allied Disciplines 30: 857–864.

52.

Mundy

(1995) Joint attention and social-emotional approach behaviour in children with autism. Development and Psychopathology 7: 63–82.

53.

Mundy

Crowson

(1997) Joint attention and early social communication: implications for research on interventions with autism. Journal of Autism and Developmental Disorders 6: 653–676.

54.

Mundy

Delgado

Block

. (1993) A Manual for the Abridged Early Social Communication Scales. Coral Gables, FL: University of Miami.

55.

Mundy

Sigman

Kasari

(1994) Joint attention, developmental level, and symptom presentation in autism. Development and Psychopathology 6: 389–401.

56.

Mundy

Sigman

Ungerer

. (1986) Defining the social deficits of autism: the contribution of non-verbal communication measures. Journal of Child Psychology and Psychiatry, and Allied Disciplines 27: 657–669.

57.

Nadel

(2002) Imitation and imitation recognition: functional use in preverbal infants and nonverbal children with autism. In: Meltzoff

Prinz

(eds) The Imitative Mind: Development, Evolution, and Brain Bases. New York: Cambridge University Press, pp. 42–62.

58.

Nadel-Brulfert

Baudonnière

(1982) The social function of reciprocal imitation in 2-year-old peers. International Journal of Behavioral Development 5: 95–109.

59.

Nakagawa

(2004) A farewell to Bonferroni: the problems of low statistical power and publication bias. Behavioral Ecology 15: 1044–1045.

60.

Osterling

Dawson

(1994) Early recognition of children with autism: a study of first birthday home videotapes. Journal of Autism and Developmental Disorders 24: 247–257.

61.

Perneger

(1998) What’s wrong with Bonferroni adjustments. British Medical Journal 316: 1236–1238.

62.

Phillips

Baron-Cohen

Rutter

(1992) The role of eye contact in goal detection: evidence from normal infants and children with autism or mental handicap. Development and Psychopathology 4: 375–383.

63.

Poon

Watson

Baranek

. (2012) To what extent do joint attention, imitation, and object play behaviors in infancy predict later communication and intellectual functioning in ASD? Journal of Autism and Developmental Disorders 42: 1064–1074.

64.

Robins

Fein

Barton

. (2001) The Modified Checklist for Autism in Toddlers: an initial study investigating the early detection of autism and pervasive developmental disorders. Journal of Autism and Developmental Disorders 31: 131–144.

65.

Roeyers

Van Oost

Bothuyne

(1998) Immediate imitation and joint attention in young children with autism. Development and Psychopathology 10: 441–450.

66.

Rogers

(1999) Intervention for young children with autism: from research to practice. Infants and Young Children 12: 1–16.

67.

Rogers

Dawson

(2009) Play and Engagement in Early Autism: The Early Start Denver Model. Volume I: The Treatment. New York: Guilford Press.

68.

Rogers

Pennington

(1991) A theoretical approach to the deficits in infantile autism. Development and Psychopathology 3: 137–162.

69.

Rutter

Le Couteur

Lord

(2003) Autism Diagnostic Interview Revised Manual. Los Angeles, CA: Western Psychological Services.

70.

Schaerlaekens

Zink

Van Ommeslaeghe

(1993) Reynell Taalontwikkelingsschalen (Handleiding). Nijmegen: Berkhout.

71.

Schlichting

JEPT

Van Eldik

MCM

Spelberg

. (1998) Schlichting Test voor Taalproductie. Nijmegen: Berkhout.

72.

Schopler

Mesibov

Shigley

. (1984) Helping autistic children through their parents: the TEACCH model. In: Reynolds

Gutkin

(eds) The Handbook of School Psychology. New York: Wiley, pp. 629–643.

73.

Sigman

Ruskin

(1999) Continuity and change in the social competence of children with autism, Down syndrome, and developmental delay. Monographs of the Society for Research in Child Development 64: 1–108.

74.

Stone

Yoder

(2001) Predicting spoken language level in children with autism spectrum disorders. Autism 5: 341–361.

75.

Stone

Ousley

Yoder

. (1997) Nonverbal communication in two- and three-year-old children with autism. Journal of Autism and Developmental Disorders 27: 677–696.

76.

Tomasello

(1995) Joint attention as social cognition. In: Moore

Dunham

(eds) Joint Attention: Its Origins and Role in Development. Hillsdale, NJ: Lawrence Erlbaum Associates, pp. 103–129.

77.

Tomasello

(2001) Cultural transmission: a view from chimpanzees and human infants. Journal of Cross-Cultural Psychology 32: 135–146.

78.

Tomasello

Farrar

(1986) Joint attention and early language. Child Development 57: 1454–1463.

79.

Uzgiris

Hunt

(1975) Assessment in Infancy: Ordinal Scales of Psychological Development. Champaign, IL: University of Illinois Press.

80.

Van der Paelt

Warreyn

Roeyers

(2012) Interventie bij jonge kinderen met een autismespectrumstoornis: inventarisatie van praktijkkennis in Vlaanderen. Report by order of SEN vzw. Unpublished manuscript.

81.

Vander Steene

Bos

(1997) WPPSI-R: Wechsler Preschool and Primary Scales of Intelligence: Vlaams-Nederlandse Aanpassing. Lisse: Swets & Zeitlinger.

82.

Warreyn

Roeyers

Van Wetswinkel

. (2007) Temporal coordination of joint attention behavior in preschoolers with autism spectrum disorder. Journal of Autism and Developmental Disorders 37: 501–512.

83.

Wechsler

(1989) Wechsler Preschool and Primary Scale of Intelligence (rev. ed.). New York: The Psychological Corporation.

84.

Wetherby

Woods

Allen

. (2004) Early indicators of autism spectrum disorders in the second year of life. Journal of Autism and Developmental Disorders 34: 473–491.

85.

Whalen

Schreibman

(2003) Joint attention training for children with autism using behavior modification procedures. Journal of Child Psychology and Psychiatry, and Allied Disciplines 44: 456–468.

86.

Whalen

Schreibman

Ingersoll

(2006) The collateral effects of joint attention training on social initiations, positive affect, imitation, and spontaneous speech for young children with autism. Journal of Autism and Developmental Disorders 36: 655–664.

87.

White

O’Reilly

Streudsand

. (2011) Best practices for teaching joint attention: a systematic review of the intervention literature. Research in Autism Spectrum Disorders 5: 1283–1295.

88.

Whiten

Brown

(1998) Imitation and the reading of other minds: perspectives from the study of autism, normal children, and non-human primates. In: Braten

(ed.) Intersubjective Communication and Emotion in Ontogeny: A Sourcebook. Cambridge: Cambridge University Press, pp. 260–280.

89.

Williams

JHG

Whiten

Singh

(2004) A systematic review of action imitation in autistic spectrum disorder. Journal of Autism and Developmental Disorders 34: 285–299.

90.

Young

Rogers

Hutman

. (2011) Imitation from 12 to 24 months in autism and typical development: a longitudinal Rasch analysis. Developmental Psychology 47: 1565–1578.