To what extent does autism reflect a perceptual (central) barrage to sensory (peripheral) experiences during receptive phases of development? This intriguing issue can be explored by means of a fourfold table of people with/without blindness (or other sensory-perceptual damage) and with/without autism. Linda Pring, Professor of Psychology at Goldsmith College, London, precisely obtains this from a well-composed range of experts on autism and/or visually disadvantaged children.

The book brings together theories and findings in a field that maintains a vigorous level of controversy. It challenges some collaterals of the ‘mindblindness’ construct that has been used to describe the qualitatively different development of social cognition in autism (Reference Baron-CohenBaron-Cohen, 1995). If autism and blindness were ‘phenocopies’, not much would be left for the genotype of idiopathic autism. The last chapter, by Helen Tager-Flushberg, Neuroanatomy Professor in Boston, may expediently be read first. She sums up the volume parts, spanning from Peter Hobson's clinicotheoretical essay (her preferred) to Rita Jordan's psychoeducational approach.

Susan Leekam and Shirley Wyver critically review how some people manage to grow up without attributing mental functions to others. They point out that sighted and blind children do not differ in prenatal exposures. Having reviewed Piaget's and Gibson's conflicting theories on specific/modal representations in cognitive development, they elegantly deviate from the theory-of-mind doctrine in favour of the role of social interaction with caregivers as a basis for strengthening the capacity for intersensory coordination and mentalisation. Neuroscience is addressed in Naomi Dale's chapter on a case-control study of blind v. typical infant. There are similarities between the neurodevelopmental regression (setback) experienced by blind and often multi-functionally impaired infants and those who are going to be retrospectively diagnosed with autism. The other chapters, by G. Collis, G. Conti-Ramsden, R. Gibbons, V. Lewis, M. Pérez-Pereira and V. Tadic, are also remarkably interesting. They add complexity to the blindness-autism fourfold table taking into account different levels of vision and social impairments.

In spite of the ‘blindisms’ - echolalia, pronoun-reversal, stereotypical behaviours, poor symbolic play, etc. - by and large the association of autism with congenital blindness remains weak in grown-up individuals. When they coexist, it is an opportunity to look at the development of what Leo Kanner called ‘affective contact’ two generations ago in a way similar to the studies on auditory impairment or socially deprived children (Reference Rutter, Andersen-Wood and BeckettRutter et al, 1999). How can visually impaired infants acquire the gestures of non-verbal communication and eventually picture the same world as others’ in their minds? As sighted people do, many elements have to be connected.