Domain-Specific Impairment in Flexible Value Learning in Young Children with ASD

Background: Limited selective attention toward social objects such as faces constitute one of the core symptoms of autism. Evidence from both human and primate studies suggests that the process of selecting objects for processing is guided by the appraisal of object values. Here we tested a hypothesis that impairments in selective attention in ASD may reflect a disruption of a system supporting learning about object values in social domain.

Objectives: To compare the value learning in social (faces) and nonsocial (fractals) domains on selective attention in young children with ASD and typically developing (TD) controls.

Methods: Participants included children with ASD (n=26, age M=44.0±17.3) and TD controls (n=28, age = 38.3±13.6). All children were tested with a novel gaze-contingent task employing instrumental reversal learning. The task design was inspired by a value learning task previously developed and validated in non-human primates (Kim et al., 2013). The task consisted of 4 blocks, each block consisted of training and testing trials (Subplot A). Training: The same two objects, either two fractals or two faces, were presented in all blocks. Within each block, one object was assigned an a priori high-value (HV) and the other one a low-value (LV). If a child looked at a HV object, the object became dynamic, turning into a smiling face (Social) or a colorful fractal animation (Nonsocial). Fixation on the LV object resulted in no changes. Test: The two HV and LV objects from the training trials were presented simultaneously. A saccade toward the HV object activated its dynamic display; a saccade toward the LV object resulted in no change of the display. The proportion of test trials during which a child fixated on the HV object #HV /(#HV + #LV) across all testing blocks served as a dependent variable (%HV).

Results: An ANOVA on %HV revealed no significant effect of diagnosis (F(1, 43)=2.21, p = .144), or condition (F(1, 43)=.59, p=.542 but a significant diagnosis x condition interaction (F(1, 43)=8.88, p=.005). Between-group comparisons showed no significant difference in the Nonsocial condition (p=.577, d=.29) but a significant TD group advantage in the Social condition (p = .007, d = .85).(Subplot B) Within-group contrasts showed that, following training, the bias toward HV objects in the TD group in the Social condition was stronger compared to the Nonsocial condition (­p=.007), but this bias was comparable between conditions in the ASD group (p=.164). A significant correlation between performance in the Social and Nonsocial conditions was found in the ASD group (r(22) =.47, p=.025) but not in the TD group (r(24) = .086, p=.689).(Subplot C).

Conclusions: Children with ASD exhibit domain-specific impairments in value learning which may impact their attention in natural environments. Unlike their typically developing peers, children with ASD’s learning in social and nonsocial domains may rely on the same underlying neural mechanism. These findings have implications for designing novel treatments targeting impaired selective attention to social targets and for studies of neural mechanisms of value learning in autism.