A Sequential-Longitudinal Study On the Origins of Theory of Mind Abilities in Children with ASDs

Background: Major questions in the field of Theory of Mind concerns its founding factors. Researchers abundantly focused on declarative pointing, mirror self-recognition and symbolic play. However, there is still no agreement among researchers about the mentalistic nature of these abilities (see for example Perner e Ruffman, 2005; Lillard, 1993), and the longitudinal studies which demonstrate clearly the relationship existing among precursors and later mentalistic abilities are still rare, especially in children with Autism Spectrum Disorders (ASDs), who are known to be impoverished in Theory of Mind despite quite intact metarepresentational skills (Baron-Cohen, 1989).

Objectives: This study aimed to investigate the sequential relationship between declarative pointing, mirror-self recognition and symbolic play before 24 months and a later mentalistic abilities, speech about internal states, that emerge between the 26th and 30 month, in children with ASDs in comparison to developmental age-matched controls.

Methods: A sequential-longitudinal study was conducted considering two groups of ASDs children. The first group (n=17; 15M), mean developmental age at T1=16 months (SD=3.95), mean chronological age=44 months (SD=9.85), was assessed longitudinally from 16 to 24 months, every two months. Participants in this group were tested for declarative pointing, mirror self-recognition and symbolic play. The second group (n=12; 9M), mean developmental age at T1=22 months (SD=1.345), mean chronological age=57 months (SD=5.794), was tested longitudinally from 22 to 30 months, every two months. Participants in this group were tested for speech about internal states. The children of the two groups which resulted homogenous at the age of their overlapping (22-24 months) in the abilities considered were collapsed into an unique longitudinal sample and the sequential correlations between early and later abilities were calculated. The same design was applied to two groups of developmental age-matched Typically Developing controls (TD).

Results: While in the majority of TD children mirror self-recognition was present at 17 months, followed by declarative pointing at 18 months and by symbolic play at 20 months, in children with ASDs all these abilities were lacking except mirror self-recognition, which was very strong at 18 months of developmental age. Speech about internal states was quite lower in both groups at 30 months. However, while in TD there was a correlation respectively between speech about internal states at 30 months and declarative pointing (r=0,825; p<0.05) and mirror self-recognition (r=0,917; p=0,014), there were not correlations between early and later Theory of Mind abilities in ASDs.

Conclusions: These results seems to support the hypothesis of a non mentalistic interpretation of the precursors of Theory of Mind (Perner e Ruffman, 2005), which claims that children might learn to predict the behaviour of the other people on the basis of rules learned by previous experiences and therefore they might be successful in some Theory of Mind tasks, like the ones evaluating speech about mental states, without actually engage in a properly mentalising process.