Attentional Biases towards Nonsocial Objects Vary as a Function of An Observer's Autistic-Like Traits

Background: Current research conducted on visual attention in children with Autism Spectrum Disorder (ASD) has shown these children to exhibit selective attention to nonsocial stimuli in their environment (Sasson et al., 2010). The authors suggest this selective attention to nonsocial stimuli may have a negative impact on the development of social abilities, social impairments being highly indicative of ASD.  In support of Sasson et al.’s conclusions, past research conducted by Klin et al. 2002 found autistic social impairments to correlate significantly with fixations on objects. Longer fixation times on objects were associated with greater social impairments in individuals diagnosed with ASD. The findings of these studies suggest further research is needed to investigate the relationship between selective attention to objects and social impairments as well as the disengagement from socially relevant stimuli such as human bodies.  In the current study we utilized a dot-probe paradigm to examine whether attentional biases exist when typical observers viewed objects and human bodies. We examined the relationship of these biases to social abilities as measured by the Autism-Spectrum Quotient (AQ) (Baron- Cohen et al., 2001).

Objectives: To determine whether the magnitude of autistic traits is associated with attentional biases to objects and whether this bias, if it exists, is a function of engagement towards nonsocial objects or a disengagement from human bodies.

Methods: Male and female observers (n=14, mean age 20.1), completed a modified dot-probe task to assess attentional biases across scenes containing human bodies and objects. After a fixation is presented, a body of the same gender as the observer was presented simultaneously with an object (coke bottle) one above the other (location counterbalanced). After 500ms, the body and object disappeared and an arrow appeared (arrow direction and location counterbalanced) in the previous location of one of the stimuli. Participants reported arrow direction with a key press. Reaction times in correct trials were analyzed to determine whether observers showed an attentional bias towards human bodies or coke bottles. Observers also completed a control baseline condition in which they observed two bodies or two bottles presented simultaneously. Attentional bias scores computed from these trials will determine whether observers are engaging or disengaging from the targeted stimulus.  Autistic-like traits were measured using the AQ.

Results: All subjects scored below the level of classification for autistic traits  (AQ>32) on the AQ (M=19.9, SD= 5.0). Correlations revealed a significant relationship between AQ scores and an attentional bias towards objects r(12)=.495, p=.036.  Increased autistic traits in typical observers were associated with increased visual attentional biases towards coke bottles. 

Conclusions: The strong relationship observed between autistic traits and selective attention to nonsocial objects support the hypothesis that a failure to attend to socially relevant stimuli may underlie social impairments related to autism. Future studies with observers with ASD are needed to further examine the extent of this relationship and its contribution to the ASD phenotype.