The Role of Higher-order Theory of Mind in U-shaped Performance Curve in True Belief Task

• In Event: Theory of mind development and children’s representation of beliefs

Abstract

Recent studies that administered the true belief condition of the classical change-of-location task to a broad age range of children yielded surprising findings of a U-shaped performance curve in this seemingly trivial task. The true belief condition asks children to ascribe a veridical, a true belief to an agent in order to predict or explain her action (analog to the false belief condition by Wimmer & Perner, 1983). Children before age 4 perform competently in the true belief condition. Children who begin to solve the classical false belief condition at the age of 4, however, fail the true belief condition. They predict – systematically wrong – that the agent will search for the object in the previous instead of the actual location. Only from around age 8 to 10, children succeed again (Oktay-Gür & Rakoczy, 2017). New evidence suggests that this pattern of results reflects pragmatic confusions. Children between 4 and 10 years are pragmatically confused by the trivial test question about a subjective perspective without any possibility for divergent perspectives (Rakoczy & Oktay-Gür, 2020). However, still unsolved is the question how older children at the end of the U-shaped curve overcome these confusions and thus succeed in the task again. We hypothesized that children in an intermediate state between 4 and 10 years cannot yet solve these pragmatic confusions because of their restricted higher-order Theory of Mind and fragile pragmatic language understanding. Only when developing an understanding of higher-order forms of Theory of Mind and a more solid grasp of linguistic pragmatics, children become able to resolve the pragmatic confusions in the true belief task. Since little is known about the developmental trajectories of higher-order Theory of Mind beyond second order (as “He thinks that she believes that he knows…”), the present study aimed first, to measure children’s understanding for such mental state ascriptions from second to fifth order and second, to predict their true belief performance at the bottom and end of the U-shaped curve with this ability in higher-order Theory of Mind and their pragmatic language understanding. 81 children between 6 and 10 years were tested in a preregistered, moderated online-study. The data were analyzed using a logistic regression model in which the true belief performance was predicted by higher-order Theory of Mind and pragmatic language abilities controlled for age and language abilities. The model comparison with a model containing the control variables only (X2(5)=25.64, p<.001) shows that children’s abilities for direct higher-order mental state ascriptions (B= 0.79, p<.001, OR=2.21) but not their pragmatic language understanding (all ps>.10), predicts their success in the trivial true belief task. Children seem to be able to solve the pragmatic confusions when reaching a higher-order understanding of Theory of Mind. A follow-up study is planned to first, replicate this relationship between higher-order Theory of Mind and the true belief performance and second, to extent the set of predictors testing for the ability to understand and produce higher-order mental state ascriptions. Data collection of the second study will be finished by April, 2021.

Authors

• Lydia Schidelko, Georg-August-Universität Göttingen (GAU)
  Presenting Author

• Marina Proft, Georg-August-Universität Göttingen
  Non-Presenting Author

• Hannes Rakoczy, University of Göttingen
  Non-Presenting Author