Abstract
Autistic Spectrum Disorder (ASD) affects about 1 in 200 individuals and is associated with deficits in social communication and reciprocity. These abilities are highly dependent upon the ability to understand others' intentions and desires by ‘reading’ facial expressions and eye gaze direction. Instruments developed to diagnose ASD find that the ability to follow the direction of other's eye gaze, is impaired in ASD and discriminates between these children and those with other disorders. However, experiments have found that children with ASD reduce their reaction times to a two-choice stimulus, by taking advantage of a precue in the form of an eye gaze stimulus. These studies may be problematic as they employed static images of eyes that were removed from the face, and which could be treated as symbolic stimuli. In this study, 10 children with ASD and an age-matched control group were presented with a realistic computer generated face (composite of 100 real faces) before the presentation of an imperative stimulus (an arrow). Participants were required to press a right key when a central arrow pointed right and a left key when it pointed left. A 10 degree eye-movement within the face seen for 1000ms provided consistently valid precue information. The imperative stimuli would appear 500–3000ms (random) after the face disappeared. In a second block of trials, the eye shift occurred concurrently with a smile or a frown. The smile meant that the cue was valid but a frown indicated an invalid cue. This design feature was explained carefully to the children. This experimental paradigm provides an objective and quantifiable measure of whether children with or without ASD can integrate facial expression and eye-gaze to derive intentional information.