Scaling of Theory-of-Mind Tasks Henry M. Wellman and David Liu Two studies address the sequence of understandings evident in preschoolers’ developing theory of mind. The first, preliminary study provides a meta-analysis of research comparing different types of mental state under- standings (e.g., desires vs. beliefs, ignorance vs. false belief). The second, primary study tests a theory-of-mind scale for preschoolers. In this study 75 children (aged 2 years, 11 months to 6 years, 6 months) were tested on 7 tasks tapping different aspects of understanding persons’ mental states. Responses formed a consistent de- velopmental progression, where for most children if they passed a later item they passed all earlier items as well, as confirmed by Guttman and Rasch measurement model analyses. Children’s understanding of persons’ mental states Ftheir theory of mindFis a crucial cognitive development and has been intensely studied in the last 15 years (e.g., see Flavell & Miller, 1998). At times, theory of mind is discussed as a single cog- nitive process or achievement (especially in some areas of inquiry, such as primate cognition or re- search on autism). Relatedly, much theory-of-mind research has focused on a single task paradigm ex- amining children’s understanding of false belief. However, many researchers believe that developing a theory of mind includes understanding multiple concepts acquired in an extended series of devel- opmental accomplishments (for a recent review, see Wellman, 2002). Consequently, investigations of young children’s understandings of intentions, emotions, desires, knowledge, and other states have become prevalent. However, little research empiri- cally establishes developmental progressions in children’s various understandings. Support for one progression comes from studies showing that chil- dren’s understanding of desires seems to precede their understanding of beliefs (e.g., Bartsch & Well- man, 1995; Flavell, Flavell, Green, & Moses, 1990; Gopnik & Slaughter, 1991; Wellman & Woolley, 1990). But other progressions are empirically unclear or contentious (e.g., Mitchell, 1996; Perner, 1995). More serious still, very little research has attempted to investigate comprehensively an extended series of theory-of-mind developments. We assume that, for normally developing chil- dren, certain insights about the mind develop in a predictable sequence. We hypothesize that these in- sights index an underlying developmental progres- sion that could be captured in a theory-of-mind scale. We provide two types of empirical support for this hypothesis. First, we report a preliminary meta- analysis of studies that have compared different types of mental state understandings (e.g., desires vs. beliefs or ignorance vs. false belief). A meta-analysis seems useful to integrate and clarify scattered in- dividual findings that are at times contradictory. Primarily, however, we report a study testing a the- ory-of-mind scale for preschool childrenFa set of methodologically comparable tasks that focus on differing conceptual constructs that may devel- opmentally appear in sequence. As background, our focus concerns preschool developments, a developmental period when there are many changes in mental sate understanding. We do not include second-order false-belief tasks (which regularly are acquired in the early school years, consistently after a first-order understanding of false belief; Perner & Wimmer, 1985), nor do we include tasks representing more mature (Wellman & Hick- ling, 1994) or advanced theory-of-mind under- standings (Happe, 1994) thought to be acquired later in development and that focus largely on metaphor, irony, double deceptions, and complex narratives. Instead, we focus on younger children and consider tasks designed to assess children’s understanding of desires, emotions, knowledge, and beliefs. These tasks are different in focusing on different states (e.g., r 2004 by the Society for Research in Child Development, Inc. All rights reserved. 0009-3920/2004/7502-0020 Henry M. Wellman and David Liu, Center for Human Growth and Development, University of Michigan. Funding for this research was provided by National Institutes of Health Grant HD-22149 to Henry Wellman and by a National Science Foundation graduate fellowship to David Liu. We thank the children, parents, and the staff of the University of Michigan Children’s Centers, Gretchen’s House Six, and St. Joseph Mercy Hospital Child Care for their participation. We thank Shannon Duffany and Angela Kovalak for their help with data collection, and Eric Camburn for his help with Rasch model analyses. Correspondence concerning this article should be addressed to Henry Wellman, Center for Human Growth and Development, University of Michigan, 300 N. Ingalls, 10th Floor, Ann Arbor, MI, 48109-0406. Electronic mail may be sent to hmw@umich.edu. Child Development, March/April 2004, Volume 75, Number 2, Pages 523–541