Gestures As Facilitators for Word Learning in Children with ASD: The Role of Social Intent and Attentional Cues

Background:  

As Typically Developing (TD) children begin to follow joint attention, they become avid word learners (Baldwin, 1992). These developments are influenced by an emerging appreciation of gestures as cues to referential intent of speakers (Booth et al, 2008). Children with Autism Spectrum Disorders (ASD) show deficits in both joint attention and word-learning abilities.  Research suggests that ASD children may err in learning labels because they focus on objects they are attending to during labeling rather than objects of speaker intent (Hennon, 2002). However, it is not clear which cues to speakers’ intent children with ASD are capable of using or whether their word-learning strategies are similar to TD children’s.

Objectives:  

We investigated the roles of social referencing and attentional cues as facilitators for word learning in children with ASD, and sought to reveal qualitative differences in word-learning strategies and eye gaze patterns of ASD children relative to those of TD children.

Methods:  

Twenty-five children with ASD and matched controls participated in a within-subjects design word-learning task. For each trial, the experimenter paired an object-directed label with one of four levels of gestural support (gaze, point, touch, or manipulation). Children’s eye-gaze patterns during training and word-mapping accuracy were recorded.  

Results:  

Children were stratified into younger (2- 6) and older (6-10) groups. Older ASD children and matched controls performed near ceiling on the comprehension test. The younger ASD group performed worse than matches in all conditions and particularly poorly in the gaze condition (10% correct) relative to other conditions (57% correct).  For ASD children in the gaze condition, those who word-mapped accurately were more likely to look at the object during labeling  (64%) than those who did not (39%).  In the other conditions, accurate word-mapping trials did not differ in looks to the object of intent (59%) from incorrect trials (66%).  These data indicate that ASD children may understand the referential intent of the speaker and use this information to learn new words when a manual gesture is used.  However, when labeling is paired with gaze alone, ASD children rely heavily on their own attentional focus to learn new labels.  Further analyses will compare the eye gaze patterns of ASD and control group participants.

Conclusions:  

Young children with ASD showed impaired abilities to utilize eye gaze as a signal of referential intent during object labeling.  Interestingly, older children were able to learn the object labels using only gaze, indicating that maturity or interventions may help children with ASD develop word-learning strategies.  ASD children who learned object labels appeared to do so on the basis of some understanding of gestures as cues to referential intent, except perhaps in the gaze condition, in which attention to the target at the time of labeling was a particularly strong predictor of successful learning. Results suggest that younger children with ASD may be unable to effectively use eye gaze as a cue to referential intent in the service of learning new words, and thus resort to less effective strategies when faced with this insufficient cue alone.