Joint Attention Predicts Adaptive Behavior Development In Infants with and without ASD

Background:

Children with autism and their siblings who do not develop autism exhibit reduced adaptive behaviors and joint attention skills relative to typically developing controls (Freeman et al., 1991; Rozga et al., 2010; Toth et al., 2007; Yoder et al., 2009). Szatmari and colleagues (2009) proposed that a single mechanism such as joint attention may underlie language and adaptive-skill impairment and variability in severity of autism.  Treatment targeting joint attention impacts language, social skills, and other symptoms of autism (Kasari et al., 2008; Kasari & Rotheram-Fuller, 2005).  Language skills at 12 months are related to changes in adaptive behaviors between 12 and 36 months (Gomez et al., 2009), but the relationship between joint attention during infancy and later adaptive behavior has not been evaluated.

Objectives:

1.     Compare development of adaptive skills between infants later diagnosed with autism (ASD), non-autistic siblings of children with autism (HR-non-ASD), and low-risk controls (LR).

2.     Determine if RJA or IJA is associated with change in adaptive behaviors.

Methods:

Twenty-nine LR, 42 HR-non-ASD, and 10 ASD infants were assessed at 12, 18, and 36 months. Adaptive behaviors were assessed through parental interview using the Vineland Adaptive Behavior Scales (VABS; Sparrow et al., 1984). The Early Social Communication Scales (Mundy et al., 2003) were administered at 12 and 18 months, yielding measures of proximal RJA (following points to pictures in a book) and distal RJA (gaze/point following to posters on the wall) and frequencies of low-level IJA (eye contact/gaze alternation) and high-level IJA (pointing/showing). Univariate analyses examined change in VABS scores in relation to joint attention.

Results:

VABS raw scores for daily living skills (DLS; p = .005), social (p=.04), and communication skills (p <.001) changed less from 12 to 36 months among infants later diagnosed with ASD than HR-non-ASD and LR infants.

Neither high- nor low-level IJA predicted change in VABS scores. When gender, verbal and nonverbal mental age, and diagnostic outcome were included in analyses, proximal but not distal RJA at 12 months predicted change in DLS from 12 to 36 months F(1,74)=4.981, p =.029. Distal but not proximal RJA at 18 months predicted change in DLS from 18 to 36 months F(1,70)=6.931, p =.01. ASD diagnosis did not moderate the relationship between RJA and change in DLS. RJA was not related to changes in social, communicative, or motor skills as measured by the VABS.

Conclusions:

RJA predicted change in DLS, over and above effects associated with IQ and autism.  Baseline IQ predicted improvement in DLS (Freeman et al., 1999; Sigman & McGovern, 2005) in individuals with autism.  But DLS appears to be less impaired than other adaptive skills among people with autism (Carter et al., 1998). As RJA skills extended to more distal targets from 12 to 18 months, the distal measure became a better predictor of adaptive skills.  These results suggest that the ability to understand referential acts supports the acquisition of a variety of other skills including language, personal care, and self-help.  This study affirms the importance of targeting RJA in early intervention programs.