An empirical test of the minimalism theory of theory of mind development

Introduction. The Minimalist account of theory of mind (ToM) development contends that ToM develops through exposure to repeated behaviours and, later, mental state utterances. Patterns in these actions and utterances are recognised through statistical learning – an extremely early developing ability – which helps to build an understanding of how people act. This understanding allows children to predict others’ actions, serving as the foundation upon which understanding of mental states is later scaffolded. Typically, the understanding of mental states emerges between three- to four-years-old, when children can pass explicit false belief tasks. Method. The two experiments in this thesis were designed to test the minimalist account of ToM. Experiment 1 tested participants aged between 30- and 54-months-old three times, each two months apart, using the tasks from a validated ToM scale. Between the first and second testing session, they were given a set of storybooks that did or did not contain mental state language and did or did not contain object searching. Parents were asked to read these storybooks to their children at least four times a week. Experiment 2 was a longitudinal study that included two cohorts, starting at 9- and 21-months-old respectively. Each participant was tested four times, each three months apart, on a range of age-appropriate social understanding tasks (i.e., early predictors of ToM ability). After testing, they took home a hat for the child to wear which housed a camera, allowing us to record what language and behaviours the child was exposed to during a normal day. This footage was later coded for both language and behaviours, and analysed alongside their performance in the social understanding tasks. Hypotheses. For Experiment 1, I hypothesised that children who were given storybooks containing mental state words and object search would see more improvement on the ToM tasks over the course of the experiment compared to those who received storybooks without those features. For Experiment 2, I hypothesised that exposure to higher levels of mental state language and behavioural repetition at home would result in better social understanding task performance, mediated by the child’s statistical learning ability. Results. Experiment 1 revealed no difference in task performance between experimental conditions, and this held when mental state words and object search were analysed separately. In Experiment 2, for just the older cohort, earlier exposure to repetitions, mental state language, and superior statistical learning abilities predicted better performance on some of the social understanding tasks. Conclusions. While the results of Experiment 1 did not support my hypotheses, this could have been due to the intervention being insufficiently powerful. This claim is strengthened by the fact that, in Experiment 2, there was evidence to support the role of repetitions, mental state language, and statistical learning in ToM development