Social attention and social-emotional modulation of attention in Angelman syndrome: an eye-tracking study

Abstract

Individuals with Angelman syndrome (AS) present with severe intellectual disability alongside a social phenotype characterised by social communication difficulties and an increased drive for social engagement. As the social phenotype in this condition is poorly understood, we examined patterns of social attention and social modulation of attention in AS. Twenty-four individuals with AS and twenty-one young children with similar mental age were shown videos featuring unfamiliar actors who performed simple actions across two conditions: a playful condition, in which the actor showed positive facial emotions, and a neutral condition, in which the actor showed a neutral facial expression. During the passive observation of the videos, participants' proportion of time spent watching the two areas of interest (faces and actions) was examined using eye-tracking technology. We found that the playful condition elicited increased proportion of fixations duration to the actor's face compared to the neutral condition similarly across groups. Additionally, the proportion of fixations duration to the action area was similar across groups in the two conditions. However, children with AS looked towards the actor's face for a shorter duration compared to the comparison group across conditions. This pattern of similarities and differences provides novel insight on the complex social phenotype of children with AS.

Figure 1

Results from the generalised linear mixed effect models (*= p < 0.05). Panel (A) shows the Total Fixations Duration (TF) per trial in CG and ASG estimated through a negative binomial mixed effect model. Panel (B) represents the proportion of fixations duration (P-FD) to the areas of interest (face, actions) in CG and ASG. Panel (C) represents the P-FD to the areas of interest across conditions (playful and neutral) in CG and ASG. Both results in panel B and C were obtained from a beta mixed effect regression model. Bars represent the 95% confidence intervals estimated by the regression models.

Figure 2

Results from the mixed effect models obtained to test for the association between mental age and TF—total fixations duration—(A) and P-FD—proportion of fixation duration—(B) and between chronological age and TF (C) and P-FD (D). TF is expressed in milliseconds and is considered per trial. Chronological and mental age were standardized. The dashed lines represents the 95% confidence intervals estimated by the regression models.

Figure 3

Video-recorded condition. The red oval line represents an interest area for face, the red rectangle line represents an interest area for action.

Figure 4

Examples of video-recorded playful (A,B) and neutral (C,D) conditions. Heat maps represent the time spent (ms) watching the video. Colors, from green to red, represent the increasing time spent watching the videos. In figures (A–C) a participant belonging to CG spent time watching both faces and action areas. In figure C the same participant is interested in objects and the screen, avoiding interest areas.