The curious incident of the photo that was accused of being false: issues of domain specificity in development, autism, and brain imaging.

We resume an exchange of ideas with Uta Frith that started before the turn of the century. The curious incident responsible for this exchange was the finding that children with autism fail tests of false belief, while they pass Zaitchik's (1990) photograph task (Leekam & Perner, 1991). This finding led to the conclusion that children with autism have a domain-specific impairment in Theory of Mind (mental representations), because the photograph task and the false-belief task are structurally equivalent except for the nonmental character of photographs. In this paper we argue that the false-belief task and the false-photograph task are not structurally equivalent and are not empirically associated. Instead a truly structurally equivalent task is the false-sign task. Performance on this task is strongly associated with the false-belief task. A version of this task, the misleading-signal task, also poses severe problems for children with autism (Bowler, Briskman, Gurvidi, & Fornells-Ambrojo, 2005). These new findings therefore challenge the earlier interpretation of a domain-specific difficulty in inferring mental states and suggest that children with autism also have difficulty understanding misleading nonmental objects. Brain imaging data using false-belief, "false"-photo, and false-sign scenarios provide further supporting evidence for our conclusions.