The Effects of Others' Speech during Activity Monitoring on Attention Patterns in Toddlers with and without ASD

Background: Previous research on activity monitoring eye-tracking tasks suggests that toddlers with Autism Spectrum Disorder (ASD) attend less to people and their activities than developmentally delayed and typically developing toddlers (Shic, Bradshaw, Klin, Scassellati, & Chawarska, 2011). Toddlers with ASD divert their attention more towards non-social stimuli (Shic et al., 2014). However, it is unclear whether the presence or absence of others’ speech influences attention allocations in toddlers with ASD.  

Objectives: To examine the impact of others’ speech during activity monitoring in toddlers with and without ASD.   

Methods: Thirty-two toddlers with ASD (M_age = 30 months, SD = 7 months) and 42 toddlers without ASD (M_age = 27 months, SD= 9 months) viewed 16 20s video clips of two female adults interacting and engaging in a shared activity. Participants without ASD consisted of toddlers with typical development and those with developmental delays not classified as ASD. The video clips varied by the actress pair, type of activity, background, gaze behavior of the actresses, and presence of distractors. Participants’ gaze patterns were examined with eye-tracking method. Analyses were conducted to investigate  between-group differences in the percentage of time participants spent looking at the scene (%Scene), the actresses’ heads (%Head), bodies (%Body), both actresses’ heads and bodies (%People), shared activity (%Activity), and background (%Background) during times when actresses were speaking vs. when actresses were silent.  

Results: Compared to non-ASD toddlers, the ASD group demonstrated significantly less looking towards people (p < .01) and activity (p < .001), and significantly more looking at the background (p < .001) throughout the experiment. For both diagnostic groups, %People increased with the presence of speech (p < .05). There was a trend towards decreased looking at activities when speech was present (p =.08). Non-ASD toddlers looked more at heads than bodies when speech was present (p < .05). ASD participants looked more at people when actresses were speaking than when they were not speaking (p < .05), but %Head and %Body within the ASD group did not differ. Univariate ANOVA analysis did not yield any significant interaction between speech and diagnostic groups for any outcome measure (p > .05).  

Conclusions: These results support previous findings that toddlers with ASD demonstrate decreased attention towards people and activities and increased attention towards the background as compared to their peers without ASD. The current analyses further highlight that the presence of speech may play a role in directing toddlers’ attention to social components of the scene. Non-ASD toddlers attended significantly more to heads, whereas toddlers with ASD attended to heads and bodies equally, even in the presence of speech. This may suggest that the speech is not salient enough to direct attention to heads for toddlers with ASD, which is generally the focus of typically developing and developmentally delayed toddlers. Future research should explore the effects of non-speech sounds or movement on the monitoring of social activities.