Background: The skill to attribute mental states to the self and others, or Theory of Mind (ToM), is a problem seen universally amongst children with autism spectrum disorders (ASD) and may also affect the ability to make inferences about characters while reading narrative text. Aim: The aim of this study was to teach four male participants with ASD an explicit cognitive strategy to answer inferential questions and provide feedback regarding their answers to improve their reading comprehension and ToM. Method: A single case study ABC design was used to assess the effect of the intervention. The participants read five short narrative passages each session for 20 sessions, and answered one factual and one inferential question following each passage. Specific feedback was used to respond to the answers of each question in the intervention phase. Pre- and post-intervention levels of reading comprehension and ToM were measured. Results: All four participants improved their reading comprehension in a pre and post-intervention test and three of the four participants improved their ToM understanding, although their ability to answer inferential questions involving ToM did not improve greatly. Limitations: Limitations of the study include not asking enough inferential questions involving ToM each session. During post-hoc analysis of the results it became clear that the participants could have benefited from more practice of using the strategy. Individual reading ability should have been assessed before the intervention began to determine the appropriate reading level at the beginning of the intervention. Conclusions: The explicit cognitive strategy had limited effect on participant’s ability to answer inferential questions involving ToM; however the participants improved their scores on pre and post-intervention tests of reading comprehension and ToM. These results suggest that making inferences in text may involve ToM and that the problems seen in the reading comprehension profiles of children with ASD may be attributed to their limited ToM skills.

Portrayals of artificially intelligent robots are becoming increasingly prevalent in children’s culture. This affects how children perceive robots, which have been found to affect the way children in school understand subjects like technology and programming. Since teachers need to know what influences their pupils' understanding of these subjects, we need to know how children’s preconceptions about robots affect the way they attribute mental states to them. We still know relatively little about how children do this. Based on the above, a qualitative approach was deemed fit. This study aimed to (1) investigate the reasoning and preconceptions underlying children’s mental state attribution to robots, and (2) explore the effectiveness of moral questions as a way to do this. 16 children aged 5- and 6 years old were asked to rate the mental states of four different robots while subsequently being asked to explain their answers. Half of the children were interviewed alone and half in small groups. A thematic analysis was conducted to analyze the qualitative data. Children’s mental state attribution was found to be influenced by preconceptions about robots as a group of entities lacking mental states. Children were found to perceive two robots, Atlas, and Nao, differently in various respects. This was argued to be because the children perceived these robots through archetypal frameworks. Moral questions were found successful as a way to spark reflective reasoning about the mental state attribution in the children.