Autism spectrum disorders are a leading cause for childhood disability and affect the child for life. While the first clinical signs are often present in toddlers, clinical diagnosis becomes more reliable after children turn three. The causes are currently unknown although evidence for a genetic underpinning is mounting.

About 50% of autistic children have average to superior intellectual ability with some displaying outstanding mathematical skills or an excellent memory of facts. While gifted in some ways, all autistic children have profound social difficulties; that is difficulties in making sense of social cues such as gestures and facial expressions in any given social context. A key problem is that autistic children avoid eye contact when socially interacting with othersi, whereas looking at the eyes and mouth of another person reveals important clues about their feelings and intentions (e.g. when making an ironic comment).

Our study aims to help autistic children use facial information more effectively by playing a specifically developed computer game. The game asks children to recognise emotions in still photographs of other children’s faces of a similar age. In this game, the photographs are initially covered with tiles and the task is to remove as few tiles as possible in order to make a confident decision about the expressed emotion and to waste as little game credits as possible. Soon the child will learn that removing tiles over the likely eye and mouth area maximises the chances of performing well. The degree of game difficulty is increased by presenting more difficult to interpret facially expressed emotions as the child improves. We will evaluate the effectiveness of this game-like intervention by recording a face-specific brain wave and the child’s eye movements to face stimuli before and after training. Once validated to effectively improve children’s social communication skills, the intervention can be scaled up and made widely available as a computer app. 

Research Area 
Project type 
Project Grant
Year of funding 
Funded by