Journal of Autism and Developmental Disorders, Vol. 32, No.

2, April 2002 ( 2002)

The Behaviors of Parents of Children with Autism

Predict the Subsequent Development of Their
Childrens Communication
Michael Siller1,2 and Marian Sigman1

The present study focused on behaviors that caregivers of children with autism show during
play interactions, particularly the extent to which the caregivers behavior is synchronized
with the childs focus of attention and ongoing activity. The study had two major findings.
First, caregivers of children with autism synchronized their behaviors to their childrens attention and activities as much as did caregivers of children with developmental delay and caregivers of typically developing children, matched on language capacities. Second, caregivers
of children with autism who showed higher levels of synchronization during initial play interactions had children who developed superior joint attention and language over a period of
1, 10, and 16 years than did children of caregivers who showed lower levels of synchronization initially. These findings suggest a developmental link between parental sensitivity and
the childs subsequent development of communication skills in children with autism. Implications for parent training interventions are discussed.
KEY WORDS: Autism; attention; communication; parent; language.

INTRODUCTION

of another person to an object or event by either pointing, showing, or alternating their gaze between an object and another persons eyes (Loveland & Landry,
1986; Mundy, Sigman, Ungerer, & Sherman, 1986; Sigman, Mundy, Sherman, & Ungerer, 1986; Mundy, Sigman, & Kasari, 1990; Sigman & Ruskin, 1999).
Deficits in verbal communication are even more
obvious in children with autism. Compared with typically developing children, most children with autism
develop language late and at significantly slower rates
(Le Couteur, Rutter, Lord, Rios, Robertson, Holdgrafer,
& McLennan, 1989; Lord & Rhea, 1997). What is striking about the language development of children with
autism is that some children acquire good language
skills and others remain largely nonverbal. Two longitudinal studies tracked the language development of
4-year-old children with autism who were considered
preverbal as defined by receptive language skills that
were equivalent to those of typical children younger
than 24 months. In one study, about 50% developed at

Communication is a major area of concern in children with autism. Their nonverbal communication is
characterized by a lack of joint attention, defined as
behaviors used to follow or direct the attention of another person to an event or object to share an interest
in that event or object. Previous research has demonstrated that children with autism respond to joint attention (i.e., follow an adults gaze or pointing gesture
to an object) less than do typically developing or developmentally delayed children. Similarly, they initiate joint attention less than do children in these other
groups in that they attempt less to direct the attention
1

Department of Psychiatry, University of California, Los Angeles,

Los Angeles, California, 90024.
2
Correspondence should be addressed to Michael Siller, Department
of Psychiatry, UCLA School of Medicine, 68-237 NPI, 760 Westwood Plaza, Los Angeles, CA 90024-1759; e-mail: [email protected]

77
0162-3257/02/0400-0077/0 2002 Plenum Publishing Corporation

78
least some language (receptive language age of 24
months or above) by 10 years of age (Lord & Schopler,
1989). In the second study, about 72% of the children
with autism showed at least some language skills when
they were tested at about 13 years of age (Sigman &
Ruskin, 1999).
Attempts to explain the variability of the language
outcome in children with autism have generally investigated the relations between the childrens initial abilities and their consequent level of language acquisition.
One relation that is widely established is that the childs
language skills during early childhood are predictive of
his or her consecutive language development (Sigman
& Ruskin, 1999). With respect to children with autism
who are still nonverbal at age 4, both longitudinal studies reported above showed that initial IQ did not differentiate the children who later acquired language
from those who did not. In contrast, early nonverbal
communication and functional play skills have been
linked to subsequent gains in language skills. Young
children with autism who responded more frequently
to bids for joint attention by others made larger gains
in language skills over a period of both 1 year (Mundy
et al., 1990) and 9 years (Sigman & Ruskin, 1999) than
did children with autism who initially responded less
to others bids for joint attention.
In typically developing infants, both joint attention behaviors and language emerge within a relatively
brief developmental period between 9 and 15 months
(Sugarman, 1984; Bakeman & Adamson, 1984; Murphy & Messer, 1977; Leung & Rheingold, 1981; Franco
& Butterworth, 1996; Carpenter, Nagell, & Tomasello,
1998). Research concerning typically developing children has shown that the social experiences that the infant engages in during interactions with the caregiver
play an important role in fostering these newly emerging social behaviors. Bakeman and Adamson (1984)
found that before infants are able to actively coordinate their attention between an object and the caregiver,
mothers act to ensure a joint focus of attention by homing in on the spontaneous object engagement of the
child. It has been suggested that situations in which infant and caregiver display similar intentional states toward an external object provide the necessary
information to enable the infant to acquire an understanding of the similarity between self and other.
From this understanding comes the awareness that other
people attend to and have intentions toward outside entities (Moore, 1996; Carpenter et al., 1998). The emergence of this concept is reflected in the infants gaze
alternation between the caregivers face and an object
of interest and is the basis for further attempts of the

Siller and Sigman

child to read the attentional focus of another person
(i.e., respond to joint attention) or actively direct the
attention of another person either nonverbally (i.e., initiate joint attention) or with linguistic means.
Although infants who develop joint attention seem
to be equipped with the cognitive and social skills required for language, it has been argued that their attentional capabilities are not up to the challenges they face.
Their language acquisition rather depends on particular
social environments and routines (language learning formats) in which the adults communicative intentions are
made especially salient for the children (Bruner, 1981,
1983; Ratner & Bruner, 1978; Ninio & Bruner, 1978).
Social encounters that have been shown to be important
for language learning are episodes where mother and
child attend to the same object and the child coordinates
attention between mother and object ( joint engagement).
The ability of the mother-child dyad to initiate and
maintain episodes of joint engagement has been shown
to be predictive for the childs language growth in both
typically developing and children with developmental
delay. For typically developing children, superior consecutive language development has been linked to the
amount of time that they spent as infants in joint engagement with their mothers (Tomasello & Todd, 1983;
Tomasello, Mannle, & Kruger, 1986), maternal use of
language to label objects within the shared focus of attention (Tomasello & Farrar, 1986; Carpenter et al.,
1998), and low frequencies of maternal utterances with
a leading prescriptive pragmatic function (Akhtar, Dunham, & Dunham, 1991). Similarly for children with
Down syndrome, the extent to which caregivers maintained the childs attention to child-selected toys predicted the childs subsequent language development
(Harris, Kasari, & Sigman, 1996).
The role of parents in the fostering of nonverbal
and verbal communication in children with autism has
not been investigated. The possible contributions of
parents to their childrens development have been neglected, probably because of a sensitivity of investigators, clinicians, and parents to previous, fallacious,
psychogenic theories of autism. However, given that
there are predictive relations between parental styles of
interaction and their childrens communication skills
for both typically developing children and children with
Down syndrome, it seems likely that such relations
should exist for children with autism.
Findings from the literature on typically developing children contrast to some extent with how parents
of children with autism seem to behave and with how
they are often advised to behave. The few studies that
have compared the behaviors of parents of children

Parent Behavior and Communication Development in Autism

with autism with the behaviors of parents of typically
developing children have documented that parents of
children with autism are more directive and controlling
(Kasari, Sigman, Mundy, & Yirmiya, 1988; Watson,
1989). The same finding is also often reported when
the behaviors of parents of developmentally delayed
children are compared with the behaviors of parents of
typically developing children (Cunningham, Reuler,
Blackwell, & Deck, 1981; Eheart, 1982; Jones, 1977).
Moreover, parent training interventions in autism often
explicitly instruct caregivers to structure their interactions so that the adult follows through on his or her
clear expectations about what the child should do and
how the child should do it (Schopler & Reichler, 1971;
Mesibov, Schopler, & Hearsey, 1994; Lovaas, 1978;
Buch, 1995; Hemsley, Howlin, Berger, Hersov, Holbrook, Rutter, & Yule, 1978; Howlin et al., 1987). Although children with autism have been shown to
demonstrate more on-task behavior, relevant use of material, and educational progress in a setting that involves high levels of task-directed structure (Bartak &
Rutter, 1973; Rutter & Bartak, 1973; Clark & Rutter,
1981), an association with the childs social responsiveness and language development has not been
shown. The demonstration that parental sensitivity to
the attention of their autistic children is linked to gains
in communicative abilities would be relevant to the design of parent training interventions.
In conceptualizing the communicative deficits in
autism from a developmental point of view, it would
be expected that caregivers who try to maintain their
childs spontaneous toy engagement by pointing to,
showing, or talking about objects the child is already
attending to will have children who later develop superior communication skills. At an early stage of development, caregiver behaviors of this type facilitate
the social understanding of the child and thereby the
developmental origin of joint attention behaviors.
Later, the same caregiver behaviors scaffold the constrained attentional capabilities of the child and therefore facilitate the acquisition of language. The study
reported here addressed one question and one hypothesis. The question was whether caregivers of children
with autism attempt to adopt the childs focus of attention and toy engagement in an equal manner as parents of typically developing children and parents of
children with developmental delays who are at a similar stage of language development. The hypothesis was
that caregivers of children with autism who spend a
higher proportion of the play engagement targeting objects that are already the focus of the childs attention,
trying to maintain the childs ongoing activity, will

79

have children with superior communication skills at

later ages.
METHODS
Participants
Twenty-five children with autism (20 males and
5 females), 18 children with developmental delay
(11 males and 7 females), and 18 children with typical
development (14 males and 4 females) were recruited
as participants. The children with autism were recruited
from local special education programs, private pediatricians, and psychiatrists and psychologists at the
UCLA Neuropsychiatric Institute between 1980 and
1985. The diagnosis of each child was made independently of the research by a group of clinicians who form
the diagnostic core of the Clinical Research Center in
Childhood Psychopathology. This core group was experienced in the evaluation of developmental disorders
and used American Psychiatric Association (DSM-III,
1980) criteria. For all the subjects who were seen for
follow-up testing during the 1990s, the diagnosis was
confirmed using the Autism Diagnostic InterviewRevised (ADI-R; Lord, Rutter, & Le Couteur, 1994; Le
Couteur, Rutter, Lord, Rios, Robertson, Holdrafer, &
McLennan, 1989). The children with developmental
delay included 9 children with Down syndrome and
9 children with unspecified etiologies. Group comparisons reported in this study were based on a subsample
of 18 children with autism (14 males and 4 females)
who were selected to match the typically developing
children and the children with developmental delay on
mental and language age. Data on chronological age,
mental age, language age, and IQ of the children and
the educational background of the mother obtained during the initial assessment are presented in Table I. For
the children with autism, descriptive statistics are provided for both the complete sample and the subsample
that was the basis for group comparisons.
To determine whether the three comparison
groups differed in chronological age, mental age, language age, IQ, and maternal education, a series of
analyses of variance (ANOVAs) were calculated. The
results showed no significant group differences for language age, mental age, or years of maternal education.
Therefore, we considered the groups as matched on
these measures. However, significant group differences were found for chronological age [F(2, 51)
30.52, p .000] and IQ [F(2, 51) 129.02, p .000].
For both chronological age and IQ, follow-up t-tests
showed that the children with autism and the children

80

Siller and Sigman

Table I. Descriptive Information About the Sample at Intake
Autistic sample

n
Chronological age
(in months)
Mental age
(in months)
IQ
Language age
(in months)
Maternal education
(in years)
a

mean
SD
mean
SD
mean
SD
mean
SD
mean
SD

Complete

Comparison

25
50.3
11.7
24.2
8.4
46.0
9.4
16.7
6.9
13.3
1.6

18
54.2a
11.1
26.8
8.5
46.8a
10.3
18.4
7.5
13.5
1.5

Developmentally
delayed sample
18
46.1a
18.3
24.9
6.8
54.3a
16.0
20.3
5.5
13.7
2.4

Typically
developing
sample
18
21.8
6.8
25.2
8.9
110.4
12.5
22.5
9.3
15.0
2.1

Significantly different from the typically developing group, p .001.

with developmental delay did not differ from each

other but that both differed significantly from the typically developing children ( p .001).
Procedures
The initial assessments for each child occurred
during two individual sessions scheduled at the UCLA
Medical Center. These assessment sessions occurred
within a 3-week period and included the administration
of the Early Social Communication Scale (ESCS;
Mundy et al., 1986), a test of developmental abilities
and a language assessment, and the videotaping of an
episode of caregiver-child interaction. The children in
the autistic sample were seen for follow-up testing at
three time points. Because not all of the initial 25 children returned for each follow-up assessment, the sample size varies for each time point. At the first
follow-up, approximately 1 year after the initial assessment (mean interval, 13.8 months; SD, 1.0 months),
18 children were administered the ESCS. Of these children, 17 were also administered a language scale. The
second follow-up was about 10 years after the initial
assessment (mean interval, 10 years 1.5 months; SD,
20.9 months), and the third follow-up was approximately 16 years after the initial assessment (mean interval, 16 years 1.2 months; SD, 16.9 months). At the
second and third follow-ups, a language scale was administered to 17 and 19 children, respectively.
To determine how representative the follow-up
samples are of the children who were originally recruited, a series of ANOVAs was calculated that compared the subjects who did not return with those who
did return on chronological age, mental age, IQ, lan-

guage age, and maternal education. The results showed

that the group of children who returned for testing 1 year
later had a lower mean chronological age [F(1, 23)
7.81, p .01] when recruited as well as a lower mean
mental age [F(1, 23) 6.64, p .02] and a lower mean
language age [F(1, 23) 13.31, p .001] than the group
of children who did not return. The 10-year follow-up
sample had more highly educated mothers than the
group who did not return [F(1, 23) 12.77, p .002].
There were no differences between the two groups in
the 16-year follow-up.
Assessment of Nonverbal Communication Skills
The ESCS is designed to assess nonverbal communication behaviors of young children. In this procedure, the child and experimenter sat facing each other
at a small table. A set of toys was in view but out of
reach by the child. The experimenter presented and /or
activated the toys on the table one at a time. When presented, these toys were not within the grasp of the child.
Verbal interactions were kept to a minimum during the
ESCS. The child-experimenter interaction was videotaped to record the front upper body view of the child
and the upper body profile of the experimenter using a
split-screen video system. The period of interaction
with each child was approximately 25 minutes. Behavior ratings were collected from the ESCS videotapes
by trained observers. This measure has been used in a
variety of studies before, and reliability has been
shown. The mean generalizability coefficient of this
measure is .90, with a range between .73 and .98
(Mundy et al., 1990). The ESCS gives behavior ratings
for two groups of joint attention behaviors: child initi-

Parent Behavior and Communication Development in Autism

ations for joint attention and child responses to bids for
joint attention. Child initiations for joint attention include behaviors used to direct the attention of the experimenter to an event or object as an act of indicating
(alternates gaze between an active mechanical toy or a
toy in his hand and the testers face, points toward a
toy within reach or a poster, or shows a toy to the
tester). The child response to bids for joint attention is
evaluated by the tester pointing to the left, right, and
behind the child and is measured as the percentage of
instances in which the child followed the point by turning his head and eyes in the designated direction.
Assessments of Developmental and Language Skills
The test of developmental abilities used at the initial recruitment was the Cattell Infant Intelligence Scale
(Cattell, 1940) or, if the child passed the highest items
on that scale, the Stanford-Binet Intelligence Scale
(Thorndike, 1972). The Cattell Scale, which is composed of most of the subscales of the Gesell Developmental Examination, was designed to be a downward
extension of the Stanford-Binet Scale. The Cattell Scale
was selected because it was initially used clinically in
our setting and because we wished to keep the developmental measure the same over successive samples.
The language measure used at the initial and first
follow-up assessment were the Reynell Developmental Language Scales (Reynell, 1977), which was selected because it was the only standardized language
measure that could be used with our entire sample when
we started our studies in autism. At the second and third
follow-ups, the Reynell Scales were administered to
subjects who could not obtain basal scores on the Childhood Evaluation of Language Fundamentals-Revised
(CELF-R; Semel, Wiig, & Secord, 1987). For one child
in the 16-year follow-up, language assessment was
done with the Mullen Scales of Early Learning (Mullen,
1995), and for another child, the Childhood Evaluation of Language Fundamentals-Preschool (CELFPreschool; Wiig, Secord, & Semel, 1992) was used. All
these measures provide age-equivalent scores for the
childs language skills.
Caregiver-Child Interactions
Interactions were carried out in a carpeted playroom of a university laboratory that included a sectional
couch placed in a semicircle and a set of standard toys
placed on the carpet in the center of the room. These
toys included a wooden form board, a plastic shapesorting puzzle, a doll, a doll bed and bottle, a car,

81

blocks, a cup, a rattle, and a toy basket. The caregivers

were asked to play with their children using any toy
they would like to use. The interactions were videotaped with a hand-held camera for 4 minutes. Because
the participants were free to move around and only one
camera was used, the faces of both members of the dyad
were not always visible simultaneously and, occasionally, either the child or the caregiver was out of view.
For this reason, four sequences of 30 seconds were selected for analysis for each dyad. These sequences were
selected to provide optimal views of the caregiver and
child.
All sequences were coded with respect to three
behavioral dimensions, each one in a separate pass
through the videotape. The coding was done using
an observational computer system (The Observer;
NOLDUS) that is able to read the time code of videotapes. For the coding, the tape was watched at full speed
until a change in behavior occurred. Then the tape was
rewound and replayed at slow speed to determine the
exact onset /offset time of the behavior. The observers
who coded the caregiver-child interaction were different individuals than those who administered or coded
the ESCS, the cognitive or language assessment. Most
of the coding was done by the first author, who was
kept blind to the dependent measures (language age and
joint attention), as well as to the individual diagnosis
of each child. The videotapes were coded in random
order.
The coding system was designed to provide an online protocol of both the childs toy-directed attention as
well as the caregivers toy-directed behaviors. With respect to the childs toy-directed attention, each time the
child started or stopped gazing at one of the target toys,
both the time code and a code for the target toy were
added into the coding protocol. Caregiver behaviors that
were coded consisted of indicating behaviors and verbalizations. Indicating behaviors were coded each time
the caregiver pointed to, showed, or offered one of the
toys to the child and included both onset time and target toy. For the caregivers verbal behavior, only those
verbalizations that referred to at least one of the target
toys were coded. In the first step, the caregivers verbal
behavior was transcribed and then divided into distinct
utterances. In the second step, the onset/offset time was
determined for each utterance. The identity of the target
toy that the utterance referred to was also noted. Finally,
each utterance was categorized as either demanding (demands a change in the childs ongoing activity) or undemanding (caregiver maintains the childs ongoing
activity by offering reinforcement or a comment), and
this was recorded. For an utterance to be coded as un-

82
demanding, it was not sufficient that the caregiver only
referred to the same toy to which the child was attending. It was necessary that the caregiver also maintained
the childs ongoing activity without demanding that the
child modified the way he played with the toy of his
interest. For example, if the child was engaged with
making the dump truck drive on the floor and the caregiver said, Can you dump the truck, the caregiver utterance would be considered as demanding a change in
the childs ongoing activity. On the other hand, if the
caregiver said, Oh boy, this truck is driving fast, it
would be considered as an undemanding caregiver
utterance.
Interrater reliability was established with two
undergraduate students based on the interactions of
10 caregiver-child dyads (four children with autism,
three with developmental delay, and three with typical
development). Reliability was calculated both for
onset/offset time within the different behavioral categories and for the target toy to which the behaviors
were directed. For onset/offset reliability, a tolerance
of 2 seconds was used, and percentage agreement indices were calculated. This seemed appropriate, because the interobserver differences in timing were very
small (in 88% of the cases, the differences were less
than 0.4 second), and therefore the possibility of chance
agreement seemed negligible. Percentage agreement indices for the onset /offset within the different behavioral categories had a mean of 77% and ranged between
74% and 81%. Kappa coefficients for the agreement on
the target toy had a mean of .93 and ranged between
.90 and .98. For the distinction between demanding and
undemanding caregiver utterances, a kappa coefficient
showed an agreement of .78.
For data analysis, the principal operation was
to determine the degree to which the caregivers
showed /pointed to or talked about objects to which the
child was already attending (caregiver synchronization). Using the coding protocol, the synchronization
or lack of synchronization was determined for the onset
of each caregiver behavior. Based on the frequency of
synchronized caregiver behaviors, two separate composite scores were calculated for the caregivers indicating and verbal behaviors, using formula I. This
formula was designed to control for differences in the
base rate of caregiver behaviors as well as the total
duration of child toy-directed attention. A third composite score was calculated by applying the same formula to caregiver verbalizations that were not only
synchronized with the childs attention but also categorized as undemanding. For nine children (three
with autism, two with developmental delay, and four

Siller and Sigman

with normal development), the composite score for the
caregiver indicating behavior could not be calculated
because the total frequency of the behavior was zero.

Formula I. Computation formula used for the three

measures of caregiver synchronization

Caregiver
=
Synchronization

Percentage of synchronized and/ or

undemanding caregiver behaviors
Percentage of child
toy- directed attention

RESULTS
Group Comparisons of Maternal
Synchronized Behaviors
Before we investigated group differences in maternal synchronized behaviors, preliminary analyses
were conducted to determine whether the groups of
children differed in the extent of their attention to toys,
because this would have influenced the caregivers opportunity to act synchronously. Preliminary analyses
were also conducted for the total frequencies of caregiver behaviors in the two domains: indicating behaviors and verbalizations. Based on a series of one-way
ANOVAs, no reliable group differences were found in
any of those variables (Table II). Children with autism
spent the same duration attending to toys as did children in the other two groups. In addition, the three
groups did not differ in the mean frequency of caregiver indicating behaviors and verbalizations.
Finally, a series of ANOVAs was calculated comparing caregiver synchronized indicating behaviors,
verbalizations, and undemanding verbalization across
the three groups (see Table II). The results showed only
one marginally significant group difference for the synchronization of the caregivers undemanding verbalizations [F(2, 51) 3.00, p .06]. No significant
group differences were found for the other two measures of caregiver synchronization. In summary, caregivers were similar across groups in their overall
frequency of behaviors as well as the synchronization
of their behaviors with ongoing child attention.
Prediction of the Childrens Gain in Skills From
Caregiver Synchronized Behavior
The second issue to be addressed was the hypothesis that the degree of caregiver synchronization would

Parent Behavior and Communication Development in Autism

83

Table II. Group Comparison of Caregiver and Child Measures

Child measure
Duration of toy-directed
attention (sec)
Frequency of caregiver behaviors
Indicating behaviors
Verbalizations
Synchronization of caregiver behaviors
Indicating behaviors
Verbalizations
Undemanding verbalizations

Autistic
sample

Developmentally
delayed sample

Typically developing
sample

mean
SD

92.8
19.1

95.2
18.3

95.6
14.8

mean
SD
mean
SD

3.7
3.6
28.4
14.2

4.3
2.9
28.6
10.4

3.7
2.7
26.1
11.2

mean
SD
mean
SD
mean
SD

be predictive of the childrens gain in nonverbal communication and language skills. Difference scores between initial and follow-up testing were used as
measures of the childs gain in skills. Five measures of
gain were calculated: 1-year gain in the percentage of
responding to joint attention, in the frequency of initiating joint attention, and in language skills; 10-year
gain in language skills; and 16-year gain in language
skills. The mean language gain over the course of
1, 10, and 16 years was 4.7 months (SD, 6 months),
29.2 months (SD, 29.2 months), and 36.0 months (SD,
47.7 months), respectively. Gains in nonverbal communication over 10 and 16 years were not calculated
because the ESCS was modified for the older children
so that the scores were no longer comparable to those
acquired by the children at the first two testing sessions. For two children whose initial percentage of response to joint attention was 100%, a measure of 1-year
gain in response to joint attention was not calculated.
As a preliminary step, Pearson product moment correlations were calculated between the five gain scores
and total frequencies of caregiver indicating and verbal
behaviors as well as the total duration of child toydirected attention. None of these correlations was significant. Therefore, the degree of attentiveness of the
children to toys did not predict their increasing nonverbal and verbal competence. In addition, the childrens
communicative gains were not predicted by the total frequency of the caregivers indicating or verbal behaviors.
To determine whether caregiver synchrony was associated with the childs future development of communication skills, Pearson product moment correlations
were calculated between the three measures of care-

.93
.60
.91
.15
.54
.26

.96
.33
.95
.12
.49
.22

.87
.52
.98
.21
.69
.28

giver synchronization and five measures of gain in child

skills; these correlations are reported in Table III.
One-Year Gains in Nonverbal Communication
The childrens 1-year gain in response to joint attention was significantly correlated with both the total
measure of caregiver synchronized verbalizations and
the measure of undemanding synchronized verbalizations. The childs 1-year gain in frequency of initiating
joint attention was significantly correlated with the score
for the caregivers synchronized indicating behavior and
the score for the caregivers undemanding synchronized
verbalizations. Because of a concern that the results
might be partly attributable to outliers, these results
were confirmed using Spearmans nonparametric test.
The association between the gain in responding to joint
attention and the score for total synchronized verbalizations, Spearmans Rho .68, p .004, remained
significant. This was also true for the association between the gain in initiating joint attention and the score
for synchronized indicating behaviors, Spearmans
Rho .66, p .006. The relations between the scores
for undemanding synchronized verbalizations and the
improvement in both responding to and initiating joint
attention were not significant when the nonparametric
test was used and may have been due to outliers.
The predictive relation between caregiver synchronized verbalizations and the childrens increase in
the percentage of responding to joint attention might
be attributable to initial skills in the children if any of
these skills predicted both the increase in the percentage of responding to joint attention and the total score

84

Siller and Sigman

Table III. Prediction of Child Gain Based on Caregiver Synchronization
Gain in joint attention
Gain in language
Synchronization of caregiver
behaviors
Indicating behavior

Verbalizations

Undemanding verbalizations

r
p
n
r
p
n
r
p
n

Response
(1 year)

Initiates
(1 year)

1 year

10 years

16 years

.11
.71
14
.63
.01
16
.57
.02
16

.67
.01
16
.46
.05
18
.50
.03
18

.15
.61
15
.46
.07
17
.33
.20
17

.23
.43
14
.43
.08
17
.67
.005
17

.40
.11
17
.44
.06
19
.79
.001
19

for synchronized caregiver verbalizations. To determine

whether this explanation was plausible, correlations between the four initial child skill measures (percentage
of responding to joint attention, frequency of initiating
joint attention, IQ, and language score) and the gain in
the percentage of responding to joint attention were calculated. There were no predictive relations between any
of these initial child skills and the gain in the percentage of responding to joint attention. Thus, the predictive association between the caregivers use of
synchronized verbalizations and the childrens gain in
responding to joint attention cannot be attributed to initial characteristics of the children.
It is possible that the association between the caregivers use of indicating behaviors and the childrens
gain in the frequency of initiating joint attention stems
from an initial association between the caregivers use
of indicating behaviors and the childs initial skills. If
this were true, the association might be attributable to
a continuity in communicative skills in the children
rather than to any influence of the caregivers behaviors. To test this hypothesis, correlations were calculated between the four initial measures of child skills
and the score of caregiver synchronous indicating behaviors. The results showed that none of the initial child
skills was significantly correlated with the caregiver
synchronized indicating behaviors.
Finally, both predictive relations (caregiver synchronized verbalizations and gain in responding to joint
attention; caregiver synchronized indicating behaviors
and gain in initiating joint attention) remained significant even when initial IQ, joint attention, or language
skills was statistically controlled for.
Language Gains
The Pearson product moment results show a significant correlation between the score for the caregiver

undemanding verbalizations and the 10- and 16-year gain

in language skills of the children. These associations
were also significant using the nonparametric test, Spearmans Rho .59, p .013 and Spearmans Rho .67,
p .002, respectively. Although the association between the score for total synchronized verbalizations
and gain in 10-year language did not reach significance
when assessed with Pearsons r, there was a significant
association when this was assessed with the Spearmans
nonparametric test (Rho .49, p .04). Overall, then,
the frequency of synchronized verbalizations, particularly those with an undemanding quality, significantly
predicted the childs future gain in language skills.
One explanation for the relation between synchronized caregiver vocalizations and childs future
gain in language might be that caregivers vocalize synchronously more with children who concurrently have
higher communicative and developmental capabilities
and that these initial capabilities account for the predictive relation between caregiver behaviors and child
improvement. To test this explanation, correlations
were computed between the four initial measures of
child skills and the two measures of caregiver verbal
synchronization, as well as the three measures of language gain. The results showed that none of the measures of caregiver synchronization were significantly
correlated with any of the initial measures of child skills.
However, we did find an association between the initial child measures and the measures of language gain.
The results showed that initial skills in response to joint
attention were significantly correlated with the 1-year
(r .54, p .05) and 10-year (r .47, p .05) gains
in language. Also, initial IQ was significantly correlated with the childs 10-year (r .56, p .05) and
16-year (r .53, p .05) gains in language. The fact
that neither of the two measures of caregiver synchronization was significantly correlated with any of the

Parent Behavior and Communication Development in Autism

initial child measures suggests that the relation between
the synchronization of undemanding caregiver utterances and the childs gain in language cannot be explained as continuity in the childs communication
skills. In addition, both predictive relations between
caregiver undemanding synchronized verbalizations
and the 10- and 16-year language gains remained significant even when initial IQ, joint attention, or language skills were statistically controlled for.

DISCUSSION
In summary, the study reported here focused on
behaviors that caregivers of children with autism show
during play interactions with their children. We were
particularly interested in the extent to which the caregivers verbal and nonverbal behaviors were synchronized with the childs focus of attention as well as his
or her ongoing activity. The study had two major findings. First, caregivers of children with autism synchronized their behaviors to their childrens attention
and activities as much as did caregivers of children with
developmental delay and caregivers of typically developing children, matched on language capacities.
Second, caregivers of children with autism who showed
higher levels of synchronization during initial play interactions had children who developed superior communication skills over a period of 1, 10, and 16 years
compared with children of caregivers who showed
lower levels of synchronization initially.
The fact that parents of children with autism achieve
an equivalent level of synchronization is remarkable
given how difficult it often is to determine what the child
with autism is attending to and intending to do. It is also
remarkable because the developmental profile of children with autism, the distribution of weaknesses and
strengths in different areas of development, is qualitatively different from that of typically developing children or children with developmental delay. One would
expect that the complexities involved would challenge
the caregivers ability to recognize and adapt to the developmental level of the child with autism. Our findings
suggest that caregivers of children with autism successfully adapt their interactive behavior to the language
level of their child. However, it is important to mention
that children with autism, when matched with control
groups on language abilities, show characteristic deficits
in preverbal communication skills, especially joint
attention. Whether caregivers of children with autism
also recognize and adapt to these more subtle characteristics of their children cannot be answered based on
our study. Given their difficulties in sharing attention

85

with others, optimal development of language skills in

children with autism might require higher levels of synchronization from the caregiver than is true for typically
developing and even developmentally delayed children,
who do not show deficits in joint attention.
The predictive relation between the extent of caregiver synchronization and the childrens future gain in
communication skills is striking. Generally, this association was found with both parametric and nonparametric tests and was therefore not due to a few extreme
cases. With regard to the childs nonverbal communication skills, it is interesting that the strongest predictor of the gain in initiating joint attention is the
caregivers own tendency to initiate joint attention herself, albeit in a synchronized way. Total caregiver indicating behavior does not predict gain in the frequency
with which children initiate joint attention unless the
caregiver chooses to show, point, or offer an object that
is already the focus of the childs attention. This evidence suggests that children with autism are able to
learn by modeling if the activity selected is part of their
attentional focus.
The childs gain in language skills, however, is associated with the quality of the caregivers verbal but
not his nonverbal behavior. Although overall the quality of the caregivers verbal behavior turned out to be
a strong predictor of the childs long-term gain in language skills (over a period of 10 and 16 years), shortterm gain in language skills (over a period of 1 year)
was not predicted in the same way. The lack of prediction over 1 year may have been due to the fact that
there was very little change over the course of a year
in language skills. The language measures used in this
study may not have been sensitive enough to capture
the small short-term gains, whereas the substantial
long-term gains were measured more reliably.
The strongest predictor of the childs future gain
in language skills in our study was caregiver utterances
that are not only synchronized with the childs focus of
attention but also undemanding in quality. This is somewhat surprising given that even authors who favor the
idea of giving the choice over the stimulus material to
the child expect the caregiver to demand activities from
the child that go beyond his or her spontaneous toy engagement (Kern, Vorndran, Hilt, Ringdahl, Adelman,
& Dunlap, 1998). There are two distinct features of undemanding caregiver utterances that might explain this
finding. First, undemanding utterances by definition not
only match the toy to which the child is attending but
also the toy-directed activity in which the child is engaged. Demanding utterances, however, describe, suggest, or demand an activity that is different from the
childs ongoing activity. If we think of the childs focus

86
of attention as a focus on a certain activity with an object rather than just a visual focus on a certain object, the match between caregiver utterance and the
childs focus of attention is better for undemanding synchronized utterances than it is for demanding synchronized utterances. Also, it has been reported in the
literature that children with autism are less compliant
and demonstrate atypical gaze and affect patterns when
confronted with interpersonal demands (Arbelle, Sigman,
& Kasari, 1994; Lemanek, Stone, & Fishel, 1993). It
might be easier for a child with autism to process utterances that do not involve interpersonal demands.
Another issue that needs to be addressed is the optimal level of synchronized caregiver behaviors. Our
findings showed a positive correlation between the
caregivers level of synchronization and the childs outcome. However, we must be very cautious not to
extrapolate our findings beyond the range of synchronization scores that were observed in our sample of 25
caregiver-child dyads. Our study does not picture perfect synchronization and a complete ban of verbal demands as a behavioral goal for caregivers. The only
conclusion that our study allows is that caregivers of
children with autism may have the tendency to show
too little synchronization and too many demanding verbalizations rather than the other way around.
Although it is tempting to conclude from our findings that caregiver behavior facilitates the childs development of communication skills, it is important to
take at least two alternative explanations into consideration. First, one could argue that the level of caregiver synchronization reflects an adaptation of the
caregiver to a certain child characteristic and that it is
this child characteristic that in itself determines the
childs future development. The second alternative explanation is based on Sandra Scarrs concept of evoked
gene-environment effects (Scarr & McCartney, 1983).
Following her interpretation, early child characteristics have an indirect effect on the childs outcome, mediated by the childs social environment. Early child
characteristics elicit certain adaptations from his caregiver. These adaptations in the social environment,
rather than the child characteristics itself, are considered to play an important role in determining the developmental outcome of the child. In fact, with respect
to typically developing children, there is some empirical evidence that early child characteristics ( joint attention, language skills) do affect the subsequent
development in the quality of caregiver-child interaction (Markus, Mundy, Morales, Delgado, & Yale,
2000). The concept of evoked gene-environment effects is interesting in that the relation between child

Siller and Sigman

characteristics and behavioral adaptations of the caregiver is viewed as an empirical phenomenon rather
than as a theoretical necessity. Decoupling this empirical link might be a powerful tool for parent training interventions.
The design of this study does not allow us to finally decide in favor of either one of the three causal
interpretations given above. However, four arguments
speak against the alternative explanations that attribute
a crucial causal role to initial child characteristics. First,
the measures of caregiver synchronization were specifically designed to minimize the impact of child behaviors by selectively focusing on the onset of
caregiver behaviors. In addition, the formula used to
calculate the measures of caregiver synchronization
controlled for the caregivers opportunity to act synchronous (defined by the duration of the childs toydirected attention). Second, the profound differences
in child characteristics across the diagnostic groups
were not associated with variations in the extent of
caregiver synchronization. Third, our statistical analysis showed that none of the initial child characteristicslanguage age, joint attention skills, IQ, or
duration of child-directed attention to toyswas able
to explain the predictive relation between the measures
of caregiver synchronization and the childs future gain
in communication skills. It is true that one could argue
for a variety of other initial child variables of this type
that were not identified as part of our study. However,
it is important to mention that initial language and joint
attention skills are the most powerful predictors of later
language outcome that have been identified in the
autism-related literature.
Finally, the causal interpretation of our findings is
consistent with research on typically developing children as well as with recent approaches of both research
and intervention in autism. Research on typically developing children showed that maternal synchronization
is predictive for the childs later language development
and that this relation can be observed from very early
ages (9 months) (Carpenter et al., 1998). This finding
suggests that the link between caregiver synchronization
and the childs language development is foundational
rather than the product of some later derived strategy.
Genetic research provides compelling evidence for
the importance of biological factors in autism (Bailey,
Phillips, and Rutter, 1996; Rutter et al., 1999), but it
also highlights that the relation between the autistic
genotype and its manifestation in behavioral deficits
is far from being deterministic (Le Couteur, Bailey,
Goode, Pickles, Gottesman, Robertson, & Rutter, 1996).
Therefore, findings that the environment does play a

Parent Behavior and Communication Development in Autism

significant role in the development of communication
skills in children with autism is consistent with a line
of research that has a complex view on the processes
that are involved in the initiation, persistence, and desistance of behavioral deficits in childhood psychopathology and a focus on the interaction between
biological and environmental factors (Rutter, 1996;
Rutter & Plomin, 1997, Plomin, 1994).
In addition, the interactive style that was identified
in our study as being beneficial for the development of
communication skills in children with autism (being sensitive to the focus of attention of the child and trying to
maintain his or her ongoing activity) is closely related
to the concept of child choice (for a review, see Kern
et al., 1998), which is central to a variety of current interventional techniques. Child choice was shown to promote the language development in children with
language delay (Yoder, Kaiser, Alpert, & Fischer, 1993;
Camarata & Nelson, 1992; Koegel, ODell, & Koegel,
1987). In addition, a variety of recent interventions that
specifically target communication skills in children with
autism consider the childs motivation and the sharing
of control over material and tasks as basic parameters of
intervention: Pivotal Response Intervention (Koegel,
Koegel, Harrower, & Carter, 1999; Koegel, Koegel,
Shoshan, & McNerney, 1999), Natural Language Learning Paradigm (Koegel, ODell, & Koegel, 1987), Incidental Teaching Approach (McGee, Morrier, & Daly,
1999; Laski, Charlop, & Schreibman, 1988; Schreibmann, Kaneko, & Koegel, 1991), and Functional Development Approach (Greenspan & Wieder, 1999).
Given the assumption that there is a causal link between sensitive care giving and the childs future development of communication skills, the questions arises
as to why it might be so important for caregivers of children with autism to be sensitive to their childrens focus
of attention. Why is it so important that the caregivers
behavior matches the childs focus of attention and ongoing activity? Three interrelated answers are suggested.
First, recent research highlights the attentional deficits,
especially in disengaging from an ongoing stimulus and
shifting attention to a new stimulus, in children with
autism (Wainwright & Bryson, 1996). Therefore, an interactive partner who is engaged with or talks about an
object that the child is already attending to, one who
comments on the childs ongoing activity, might provide
the child with social and language experiences while
compensating for the childs attentional deficits. From
a social-cognitive point of view, the shared intentional
state toward an external object between caregiver and
child might help the child to acquire the understanding
that other persons attend to and have intentions about

87

external objects or events, an understanding thought to

be the basis for the development of joint attention and
language. The third reason why a caregiver who follows
the childs lead might promote the childs communication is related to the functional nature of communication. Naturally occurring communication always has a
goal and is therefore closely related to the motivation of
the person who wants to communicate. Children with
autism who have specific difficulties in interacting with
other people might often experience these interactions
as frustrating and unsuccessful. Therefore, an interactive
partner who is sensitive to the childs interests might
provide the child with the experience that interacting and
sharing an interest with another person are fun and motivating in and of themselves.
The difficulty in identifying a causal network from
this correlational study suggests future directions for
research. An experimental design in which caregivers
are encouraged to increase the extent of their synchronous indicating behaviors and verbalizations would be
useful from several points of view. Such a study would
investigate whether it is possible to modify the synchronous behaviors of parents during interactions with
their children with autism. If such modification was possible, the causal links between synchronized parental
behavior and improvements in communication styles of
children with autism would be more fully drawn.
ACKNOWLEDGMENTS
Preparation of this article was supported by Program Project Grant HD-DCD3570, Grant HD17662
from the National Institute of Child Health and Human
Development and the National Institute of Deafness and
Communication Disorders, and by Grant NS25243
from the National Institute of Neurological Diseases
and Stroke.
We wish to thank the children who allowed us to
observe them growing up. We extend our appreciation
to their parents who committed so much time and effort to our project. We also wish to thank all the researchers who contributed to this project over the years,
especially Corina Williams and Aida Cristina Fernandez, who conducted the most recent follow-up visit.
REFERENCES
Akhtar, N., Dunham, F., & Dunham, P. J. (1991). Directive interactions and early vocabulary development: The role of joint
attentional focus. Journal of Child Language, 18, 41 49.
American Psychiatric Association (1980). Diagnostic and statistical
manual of mental disorders, Third Edition (DSM III). Washington, DC.

88
Arbelle, S., Sigman, M. D., & Kasari, C. (1994). Compliance with
parental prohibition in autistic children. Journal of Autism and
Developmental Disorders, 24, 693702.
Bailey, A., Phillips, W., & Rutter, M. (1996). Autism: Towards an
integration of clinical, genetic, neuropsychological, and neurobiological perspectives. Journal of Child Psychology and Psychiatry, 37, 89126.
Bakeman, R., & Adamson, L. B. (1984). Coordinating attention to
people and objects in mother-infant and peer-infant interaction.
Child Development, 55, 12781289.
Bartak, L., & Rutter, M. (1973). Special educational treatment of
autistic children: A comparative study: I. Design of study and
characteristics of units. Journal of Child Psychology and Psychiatry, 14, 161179.
Bruner, J. (1981). The pragmatics of acquisition. In W. Deutsch (Ed.),
The childs construction of language. New York: Academic Press.
Bruner, J. (1983). Childs talk: learning to use language. New York:
Norton.
Buch, G. A. (1995). Teaching parents and paraprofessionals how to
provide behavioral intensive early intervention for children with
autism and pervasive developmental disorder. Unpublished doctoral dissertation, University of California, Los Angeles.
Camarata, S. M., & Nelson, K. E. (1992). Treatment efficiency as a
function of target selection in the remediation of child language
disorders. Clinical Linguistics and Phonetics, 6, 167178.
Carpenter, M., Nagell, K., & Tomasello, M. (1998). Social cognition, joint attention, and communicative competence from 9 to
15 months of age. Monographs of the Society for Research in
Child Development, 63 (Serial No. 255).
Cattell, P. (1940). The measurement of intelligence of infants and
young children. New York: Basic Books.
Clark, P., & Rutter, M. (1981). Autistic childrens responses to structure and to interpersonal demands. Journal of Autism and Developmental Disorders, 11, 201217.
Cunningham, C., Reuler, E., Blackwell, J., & Deck, J. (1981). Behavioral and linguistic developments in the interactions of normal and retarded children with their mothers. Child Development,
52, 6270.
Eheart, B. K. (1982). Mother-child interactions with nonretarded and
mentally retarded preschoolers. American Journal of Mental Deficiency, 87, 2025.
Franco, F., & Butterworth, G. (1996). Pointing and social awareness:
Declaring and requesting in the second year. Journal of Child
Language, 23, 307336.
Greenspan, S. I., & Wieder, S. (1999). A functional developmental
approach to autism spectrum disorders. The Journal of the
Association for Persons with Severe Handicaps, 24, 147161.
Harris, S., Kasari, C., & Sigman, M. D. (1996). Joint attention and
language gains in children with Down syndrome. American
Journal on Mental Retardation, 100, 608 619.
Hemsley, R., Howlin, P., Berger, M., Hersov, L., Holbrook, D., Rutter, M., & Yule, W. (1978). Treating autistic children in a family context. In M. Rutter & E. Schopler (Eds.), Autism, a
reappraisal of concepts and treatment. New York: Plenum Press.
Howlin, P., Rutter, M., Berger, M., Hemsley, R., Hersov, L., &
Yule, W. (1987). Treatment of autistic children. Chichester,
England: Wiley & Sons.
Jones, O. L. M. (1977). Mother-child communication with pre-linguistic
Downs syndrome and normal infants. In H. Schaffer (Ed.), Studies in mother-infant interaction. New York: Academic Press.
Kasari, C., Sigman, M., Mundy, P., & Yirmiya, N. (1988). Caregiver
interactions with autistic children. Journal of Abnormal Child
Psychology, 16, 4556.
Kern, L., Vorndran, C. M., Hilt, A., Ringdahl, J. E., Adelman, B. E.,
& Dunlap, G. (1998). Choice as an intervention to improve behavior: A review of the literature. Journal of Behavioral Education, 8, 151169.
Koegel, L. K., Koegel, R. L., Harrower, J. K., & Carter, C. M. (1999).
Pivotal response intervention: I. Overview of approach. The

Siller and Sigman

Journal of The Association for Persons with Severe Handicaps,
24, 174185.
Koegel, L. K., Koegel, R. L., Shoshan, Y., & McNerney, E. (1999).
Pivotal response intervention: II. Preliminary long-term outcome
data. The Journal of The Association for Persons with Severe
Handicaps, 24, 186198.
Koegel, R. L., ODell, M. C., & Koegel, L. K. (1987). A natural
language teaching paradigm for nonverbal autistic children.
Journal of Autism and Developmental Disorders, 17, 187200.
Laski, K. E., Charlop, M. H., & Schreibman, L. (1988). Training parents to use the natural language paradigm to increase their autistic childrens speech. Journal of Applied Behavior Analysis, 21,
391400.
Le Couteur, A., Bailey, A., Goode, S., Pickles, A., Gottesman, I.,
Robertson, S., & Rutter, M. (1996). A broader phenotype of
autism: The clinical spectrum of twins. Journal of Child Psychology and Psychiatry, 37, 785801.
Le Couteur, A., Rutter, M., Lord, C., Rios, P., Robertson, S.,
Holdrafer, M., & McLennan, J. (1989). Autism diagnostic interview: A standardized investigator-based instrument. Journal
of Autism and Developmental Disorders, 19, 363387.
Lemanek, K. L., Stone, W. L., & Fishel, P. T. (1993). Parent-child
interactions in handicapped preschoolers: The relation between
parent behaviors and compliance. Journal of Clinical Child Psychology, 22, 6877.
Leung, E. H. L., & Rheingold, H. L. (1981). Development of pointing as a social gesture. Developmental Psychology, 17, 215220.
Lord, C., & Rhea, P. (1997). Language and communication in autism.
In D. J. Cohen & F. R. Volkmar (Eds.), Handbook of autism.
New York: Wiley & Sons.
Lord, C., Rutter, M., & Le Couteur, A. (1994). Autism Diagnostic
Instrument-Revised: A revised version of a diagnostic interview
for caregivers of individuals with possible pervasive developmental disorders. Journal of Autism and Developmental Disorders, 24, 659685.
Lord, C., & Schopler, E. (1989). Stability of assessment results of
autistic and non-autistic language-impaired children from
preschool years to early school age. Journal of Child Psychology and Psychiatry, 30, 575590.
Lovaas, O. I. (1978). Parents as therapists. In M. Rutter & E. Schopler
(Eds.), Autism: a reappraisal of concepts and treatment. New
York: Plenum Press.
Loveland, K., & Landry, S. (1986). Joint attention and language in
autism and developmental language delay. Journal of Autism
and Developmental Disorders, 16, 335349.
Markus, J., Mundy, P., Morales, M., Delgado, C. E. F., & Yale, M.
(2000). Individual differences in infant skills as predictors of
child-caregiver joint attention and language, Social Development, 9, 302315.
McGee, G. G., Morrier, M. J., & Daly, T. (1999). An incidental teaching approach to early intervention for toddlers with autism. The
Journal of The Association for Persons with Severe Handicaps,
24, 133146.
Mesibov, G. B., Schopler, E., & Hearsey, K. A. (1994). Structured
teaching. In E. Schopler & G. B. Mesibov (Eds.), Behavioral issues in autism. New York: Plenum Press.
Moore, C. (1996). Theories of mind in infancy. British Journal of
Developmental Psychology, 14, 19 40.
Mundy, P., Sigman, M., & Kasari, C. (1990). A longitudinal study
of joint attention and language development in autistic children. Journal of Autism and Developmental Disorders, 20,
115123.
Mundy, P., Sigman, M., Ungerer, J., & Sherman, T. (1986). Defining the social deficits of autism: the contribution of nonverbal
communication measures. Journal of Child Psychology and Psychiatry, 27, 657669.
Murphy, C. M., & Messer, D. J. (1977). Mothers, infants and pointing: A study of a gesture. In H. R. Schaffer (Ed.), Studies in
mother-infant interaction. London: Academic Press.

Parent Behavior and Communication Development in Autism

Ninio, A., & Bruner, J. (1978). The achievement and antecedents of
labelling. Journal of Child Language, 5, 115.
Plomin, R. (1994). The Emanuel Miller Memorial Lecture 1993: Genetic research and identification of environmental influences.
Journal of Child Psychology and Psychiatry, 35, 817834.
Ratner, N., & Bruner, J. (1978). Games, social exchange and the acquisition of language. Journal of Child Language, 5, 391401.
Reynell, J. (1977). Reynell Developmental Language Scales. Windsor, England: NFER Publishing.
Rutter, M. L. (1996). Developmental psychopathology: Concepts and
prospects. In M. F. Lenzenweger & J. J. Haugaard (Eds.), Frontiers of developmental psychopathology. New York: Oxford University Press.
Rutter, M., & Bartak, L. (1973). Special educational treatment of
autistic children: A comparative study: II. Follow-up findings
and implications for services. Journal of Child Psychology and
Psychiatry, 14, 241270.
Rutter, M., & Plomin, R. (1997). Opportunities for psychiatry from
genetic findings. British Journal of Psychiatry, 171, 209219.
Rutter, M. Silberg, J., OConner, T., & Simonoff, E. (1999). Genetics and child psychiatry: I. Advances in quantitative and molecular genetics. Journal of Child Psychology and Psychiatry, 40,
318.
Scarr, S., & McCartney, K. (1983). How people make their own environments: A theory of genotype-environment effects. Child
Development, 54, 424 435.
Schopler, E., & Reichler, R. J. (1971). Parents as cotherapists in the
treatment of psychotic children. Journal of Autism and Childhood Schizophrenia, 1, 87102.
Schreibman, L., Kaneko, W. M., & Koegel, R. L. (1991). Positive
affect of parents of autistic children: A comparison across two
teaching techniques. Behavior Therapy, 22, 479490.
Semel, E., Wiig, E. H., & Secord, W. (1987). Clinical Evaluation of
Language Fundamental-Revised. San Antonio: The Psychological Corporation.

89

Sigman, M., Mundy, P., Sherman, T., & Ungerer, J. A. (1986). Social interactions of autistic, mentally retarded and normal children with their caregivers. Journal of Child Psychology and
Psychiatry, 27, 647669.
Sigman, M., & Ruskin, E. (1999). Continuity and change in the social competence of children with autism, Down syndrome, and
developmental delays. Monographs of the Society for Research
in Child Development, 64, Serial No. 256.
Sugarman, S. (1984). The development of preverbal communication: Its contribution and limits in promoting the development
of language. In R. L. Schriefelbusch & J. Pickar (Eds.), The acquisition of communicative competence. Baltimore: University
Park Press.
Thorndike, R. (1972). Stanford-Binet Intelligence Scale: norms tables. Boston: Houghton Mifflin.
Tomasello, M., & Farrar, M. J. (1986). Joint attention and early language. Child Development, 57, 14541463.
Tomasello, M., Mannle, S., & Kruger, A. (1986). Linguistic environment of 1- to 2-year-old twins. Developmental Psychology,
22, 169176.
Tomasello, M., & Todd, J. (1983). Joint attention and lexical acquisition style. First Language, 4, 197212.
Wainwright, J. A., & Bryson, S. E. (1996). Visual-spatial orienting
in autism. Journal of Autism and Developmental Disorders, 26,
423 438.
Watson, L. R. (1998). Following the childs lead: Mothers interactions with children with autism. Journal of Autism and Developmental Disorders, 28, 5159.
Wiig, E. H., Secord, W., & Semel, E. (1992). Clinical Evaluation of
Language Fundamental-Preschool. San Antonio: The Psychological Corporation.
Yoder, P. J., Kaiser, A. P., Alpert, C., & Fischer, R. (1993). Following the childs lead when teaching nouns to preschoolers
with mental retardation. Journal of Speech and Hearing Research, 36, 158167.

Copyright of Journal of Autism & Developmental Disorders is the property of Springer Science & Business
Media B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email articles for
individual use.