A face recognition system that simulates perception impairments of autistic children

We use a face recognition algorithm to model differences in perception between autistic and non autistic children. With our model it is possible to reproduce several phenomena of autism by assuming that autistic children lack the ability to abstract from horizontal invariants. In particular, we can explain why autistic children are able to better recognize faces from parts of the face while the overall recognition of faces is worse than in non-autistic children. We would like to consider whether ASD may be the result of a version of a sophisticated perceptual system that makes less explicit use of invariants in the real world environment than the typically-developing brain. Some of these invariants may be hard-coded into the system rather than learned. The key point of our system is not the face recognition but the model which can mimic the autistic brain. In the discussion we extend the model by suggesting a general reduced ability to abstract from many different types of invariants and relate these as explanations to typical behavioral issues. In this way we hope to give a complementary insight into autism and ASD.