16651
Sticky Attention and Word Learning in Children with ASD

Thursday, May 15, 2014
Atrium Ballroom (Marriott Marquis Atlanta)
C. E. Venker1 and S. Ellis-Weismer2, (1)Waisman Center, University of Wisconsin-Madison, Madison, WI, (2)University of Wisconsin-Madison, Middleton, WI
Background:  

Children with ASD demonstrate atypical patterns of attention that may impact language development. One of the most striking and earliest emerging differences seen in ASD is difficulty disengaging visual attention from a salient stimulus—a phenomenon known as sticky attention (Landry & Bryson, 2004). Sticky visual attention may negatively affect children’s abilities to switch attention among multiple candidate word referents, which is a requirement for successfully extracting co-occurrence statistics from ambiguous contexts (i.e., cross-situational word learning; Smith & Yu, 2008).  

Objectives:  

The primary objective of this study was to determine whether stickier attention (i.e., longer latency to disengage) was associated with poorer performance in a cross-situational word learning task. A secondary objective was determining the specificity of this relationship. It was hypothesized that sticky attention, but not overall shifting speed, would relate to cross-situational word learning in the children with ASD but not in the typically developing (TD) children.

Methods:  

Participants were 20 children with ASD (4 – 7 years) with heterogeneous language and cognitive abilities, and 27 TD children (2 – 7 years) matched on receptive vocabulary. Children in the ASD group had a community diagnosis of ASD and diagnoses were confirmed through administration of the ADOS. Children participated in a cross-situational word-learning task presented on an eye tracker that required attention to the co-occurrences between words and objects across individually ambiguous trials. In the test phase, children were asked about one of the novel words they had learned (e.g., Where’s the coro?). Accuracy was measured by the proportion of looks to the correct image during the test phase. Children also participated in a non-social eye-gaze task that measured latency to shift and disengage attention. In the disengage condition, children saw two competing stimuli. In the shifting condition, stimuli were not competing. 

Results:  

Mean latency to shift was 0.52 seconds in the TD group and 0.81 seconds in the ASD group. Mean latency to disengage was 2.27 seconds in the TD group and 1.67 seconds in the ASD group. Contrary to previous findings, the ASD and TD groups did not differ in mean latency for disengagement or shifting of attention, p > .95; latencies were longer for disengagement than for shifting.   

Consistent with predictions, sticky attention (i.e., longer latencies to disengage) was negatively associated with cross-situational word learning, r = -.43, p =.03 (one-tailed). This relationship was non-significant in the TD group, p = .20. Shifting speed was not significantly correlated with cross-situational learning in either group, ps > .60, pointing to a specific role of visual disengagement.

Conclusions:  

Sticky attention was related to cross-situational word learning in children with ASD. This finding is important because it demonstrates the negative impact of sticky attention—an early-emerging and atypical feature of ASD—on a specific type of word learning. Differences in disengagement of visual attention may help to explain delays in vocabulary learning and language processing. Future work will focus on the impact of sticky attention on moment-to-moment patterns of eye gaze.