Abstract
Abstract: :
Purpose: Previous studies have suggested that i) neuro-typically developing (NTD) children, compared to children with autism (CA), show greater BOLD responses of fusiform "face areas" by faces, and ii) CA prefer looking at faces with peripheral (noncentral) gaze to central gaze. The present study aimed to investigate the accuracy and speed with which NTD children and CA can identify changes in faces with either central or noncentral eye gaze. Methods: Sixteen children with mild to high functioning autism (aged 7-12 years), were matched with 32 NTD children on nonverbal IQ and age, and tested on a change detection task designed in Macromedia Flash 5.0. Each of 36 trials initially presented two dynamic objects (shapes, human or train faces), for 125ms. This presentation was followed by a transient blanking of the screen for 250ms, and then a probe version of the initial presentation of duration determined by a 2AFC. Eye gaze position of the human and train faces was counterbalanced (centre, noncentre). Participants were then required to identify which of the two stimuli had changed. Results: Children with autism were slower and less accurate at detecting changes in all three stimulus groups. CA were better at detecting changes in shapes (M = 342ms), whereas NTD children revealed a change detection advantage for faces (M = 225ms). All children were able to detect changes in faces significantly faster when eyes were looking noncentrally (MNTD = 216ms, MCA = 275ms), compared to centrally (MNTD = 252ms, MCA = 481ms). Conclusions: It appears that CA are better at detecting changes in shapes than faces (human, train), but are slower overall. Interestingly, all children were better at detecting change peripherally than centrally. The question remains as to whether this performance facilitation is due to salience or lack of it in the faces.
Keywords: attention • face perception • eye movements