Embodied cognition theories hold that cognitive processes are grounded in bodily states. Embodied processes in autism spectrum disorders (ASD) have classically been investigated in studies on imitation. Several observations suggested that unlike typical individuals who are able of copying the model's actions from the model's position, individuals with ASD tend to reenact the model's actions from their own egocentric perspective. Here, we performed two behavioral experiments to directly test the ability of ASD individuals to adopt another person's point of view. In Experiment 1, participants had to explicitly judge the left/right location of a target object in a scene from their own or the actor's point of view (visual perspective taking task). In Experiment 2, participants had to perform left/right judgments on front-facing or back-facing human body images (own body transformation task). Both tasks can be solved by mentally simulating one's own body motion to imagine oneself transforming into the position of another person (embodied simulation strategy), or by resorting to visual/spatial processes, such as mental object rotation (nonembodied strategy). Results of both experiments showed that individual with ASD solved the tasks mainly relying on a nonembodied strategy, whereas typical controls adopted an embodied strategy. Moreover, in the visual perspective taking task ASD participants had more difficulties than controls in inhibiting other-perspective when directed to keep one's own point of view. These findings suggested that, in social cognitive tasks, individuals with ASD do not resort to embodied simulation and have difficulties in cognitive control over self- and other-perspective.

Asperger syndrome (AS) is a neurodevelopmental condition within the Autism Spectrum Disorders (ASD) characterized by specific difficulties in social interaction, communication and behavioural control. In recent years, it has been suggested that ASD is related to a dysfunction of action simulation processes, but studies employing imitation or action observation tasks provided mixed results. Here, we addressed action simulation processes in adolescents with AS by means of a motor imagery task, the classical hand laterality task (to decide whether a rotated hand image is left or right); mental rotation of letters was also evaluated. As a specific marker of action simulation in hand rotation, we assessed the so-called biomechanical effect, that is the advantage for judging hand pictures showing physically comfortable versus physically awkward positions. We found the biomechanical effect in typically-developing participants but not in participants with AS. Overall performance on both hand laterality and letter rotation tasks, instead, did not differ in the two groups. These findings demonstrated a specific alteration of motor imagery skills in AS. We suggest that impaired mental simulation and imitation of goal-less movements in ASD could be related to shared cognitive mechanisms.

We studied a family in which the same 10 Mb inverted duplication of 2p25.3 -p25.1 segregates in two children and their father, all showing a trisomy phenotype. As FISH analysis demonstrated that the duplication was inverted, we suspected that a contiguous terminal deletion was also present, according to the classical inv dup del type of rearrangements. Although FISH with 2p and 2q subtelomeric probes gave normal results, 100 kb resolution array-C\GH (aCGH) showed that, beside the duplication, a 273 kb deletion was also present. The presence of a single-copy region between the deleted and duplicated regions was further suspected through high-resolution aCGH analysis (B20 kb), although only one informative spot having a normal log ratio was detected. The precise structure of the rearrangement was re-defined by real-time PCR and breakpoint cloning, demonstrating the presence of a 2680 bp single-copy sequence between deleted and duplicated regions and the involvement of a simple repeat with the potential for forming a non-B DNA structure. The rearrangement was not mediated by segmental duplications or short inverted repeats, and the double-strand break might have been repaired by nonhomologous end joining or microhomology-mediated intrastrand repair. These data highlight the fact that concomitant deletions associated with inverted duplications are very likely to be more frequent than classical cytogenetic methods alone have been able to demonstrate. The phenotypic effects of the trisomy and of the terminal 2p deletion are discussed.

These findings showed that PECS intervention (Phases I-IV) can improve social-communicative skills in children with autism. This improvement is especially evident in standardized measures of adaptive behaviour and measures derived from the observation of children in an unstructured setting.

These findings demonstrated that PECS training can promote long-term enhancement of specific socio-communicative skills in children with autism.