Interpreting the results of explicit and applied theory of mind collectively in autistic children: A solution from Rasch analysis

Abstract

Multidimensional theory of mind assessments should include items assessing both explicit theory of mind (theory of mind knowledge) and applied theory of mind (application of theory of mind knowledge in real-life contexts). However, the two theory of mind scores cannot be interpreted collectively to identify children having mismatched explicit and applied theory of mind due to the lack of links between the two scores. To map the information between two theory of mind scores, data of 221 autistic children aged 3–12 years were extracted. Items of the Theory of Mind Task Battery and Theory of Mind Inventory-2 were used to respectively assess the explicit and applied theory of mind. Rasch analysis and correlation coefficients were used to examine the model fits/dimensionality and relationships. After the removal of misfit items, the remaining 45 items showed satisfactory model fits (infit and outfit mean squares < 1.35). Large correlation ( r = 0.63) was found between the Rasch scores of the two theory of mind constructs. The scatter plots highlighted that each applied theory of mind score could reflect multiple explicit ToM scores. Accordingly, the information from each theory of mind measure is unique and irreplaceable. Moreover, cutoff scores for identifying children with mismatched explicit and applied theory of mind were proposed to determine the priority of interventions. Lay abstract Theory of mind is an ability to infer others’ mental states, which is a foundation for generating appropriate social responses. Theory of mind can be conceptually divided into two related but distinguishable constructs: explicit theory of mind (conceptual knowledge/information about others’ mental states) and applied theory of mind (the ability to use theory of mind skills in real-life contexts). Although these two theory of mind scores can be described by the percentages of children in the early, basic, and advanced developmental stages, the resulting information may not be sufficient to determine the corresponding relationships between these two theory of mind constructs or identify children with mismatched theory of mind abilities (e.g. children who have difficulty in effectively applying their theory of mind knowledge in real-life contexts). To resolve these limitations, methods for simultaneously interpreting the relationships between the two theory of mind scores are proposed. Based on the findings, each applied theory of mind score can reflect multiple scores of explicit theory of mind. In particular, the results do not take measurement error into consideration, which would make them more ambiguous. Therefore, the scores of applied theory of mind should be interpreted carefully, given that children who have the same applied theory of mind score may actually have high or low explicit theory of mind. Regarding the method for joint interpretation, cutoff scores were selected to identify children who have mismatched theory of mind abilities (high explicit theory of mind with low applied theory of mind or low explicit theory of mind with high applied theory of mind) and determine the priority for interventions.