Spanish Validation of A New and Advanced Test of Theory of Mind: Kaland's Stories From Everyday Life

Background: One of the most salient characteristics of Autism Spectrum Disorders (ASD) is their failing to attribute mental states to themselves and others. There are some useful instruments to assess Theory of Mind (ToM) (Abell, 2000; Baron-Cohen, 1997) but often subjects with high functioning ASD (HF-ASD) find easy to pass second-order tasks. Among the most recent instruments developed to evaluate “advanced” ToM tasks there is the Kaland’s Stories from Everyday Life (SEL) test (Kaland, 2002). This battery includes the ToM elements in a naturalistic story context. Children and adolescents with Asperger Syndrome performed worse than their matched controls to impute mental states in this test and needed more prompt questions and required more time to complete the tasks (Kaland, 2002, 2007).

Objectives: To validate a brief version of the SEL in a Spanish sample of HF-ASD children and adolescents and to compare with a group of healthy comparison controls (HC).

Methods: Participants were 27 HF-ASD participants (mean age=12.23±2.8) and HC (mean age=13.75±2.89) males. All patients fulfilled ASD criteria on DSM-IV. Exclusion criteria included an IQ<70, an active psychopathological condition and severe neurological or medical disorders.The SEL were translated into Spanish, it were back-translated into English and received the approval of the author. The research team systematized the scoring procedures and wrote a manual with the administration and scoring instructions. The evaluator read the 13 different types’ stories and formulate control questions to check the verbal comprehension and two central questions to test participants’ ability to infer physical (PI) and mental states (MI) from the story context. Prompt questions (PQ) were given if an incomplete answer was provided. The reaction time was recorded before (RT) and after all prompt questions (RT_PQ). There was a second evaluator who was blind to the status of the subjects. Measures of verbal IQ and severity of ASD symptoms were registered.

Results: Internal consistency and inter-rater reliability were moderate to high (.60-.90). After covariation for verbal IQ, only the correlation between SCQ and MI emerged (r= -.30, p= .05), indicating that children with high levels of ASD symptoms kept on showing difficulties to infer the mental states in spite of a good verbal competence. PI and MI scores correctly classified 64% (p= .015) to 67% (p= .021) of participants in patients versus control group. When HF-ASD and HC compared controlling for verbal IQ, significant differences emerged in total PI (t=-3.1, p=.004) and MI (t=-2.56, p=.014) raw scores. This result indicated that the higher difficulties of HF-ASD in inferring physical and mental states were not only explicated by their verbal competence but by the clinical diagnostic.

Conclusions: In our Spanish sample, the SEL instrument showed adequate psychometric characteristics and includes an administration manual. Differences between groups were obtained in PI and MI. Children with more severe ASD symptoms kept on showing more difficulties to infer the mental states in spite of a good verbal competence. These preliminary findings suggest that the SEL is useful to evaluate ToM in HF-ASD children and adolescents.