British Journal of Psychology (2015) © 2015 The British Psychological Society www.wileyonlinelibrary.com Limits on efficient human mindreading: Convergence across Chinese adults and Semai children Bo Wang, Nur Shafiqah Abdul Hadi and Jason Low* Victoria University of Wellington, New Zealand We tested Apperly and Butterfill’s (2009, Psychological Review, 116, 753) theory that humans have two mindreading systems whereby the efficient-system guiding anticipatory glances displays signature limits that do not apply to the flexible system guiding verbal predictions. Experiments 1 and 2 tested urban Mainland-Chinese adults (n = 64) and Experiment 3 tested Semai children living in the rainforests of Peninsular Malaysia (3- to 4-year-olds, n = 60). Participants – across different ages, groups and methods – anticipated others’ false-beliefs about object-location but not object-identity. Conver- gence in signature limits signalled that the early-developing efficient system involved minimal theory-of-mind. Chinese adults and older Semai children showed flexibility in their direct predictions. The flexible mindreading system in ascribing others’ beliefs as such was task-sensitive and implicated maturational and cultural contributions. The development of our everyday mindreading or ‘theory of mind’ – typically investigated through false-belief tasks – involves the awareness that people’s actions are motivated by beliefs that are constrained by particular representations of the world. The standard task involves making action predictions about a boy (Maxi) who holds a false-belief as a result of some unexpected change in the location of his chocolate (Wimmer & Perner, 1983). Many 3-year-olds answer that Maxi will look for his chocolate where it actually is whilst many 4- to 5-year-olds judge that Maxi will look for the chocolate in the original location. The basic developmental trajectory in predicting the effect of false-beliefs on others’ actions is robust and uniform in populations around the world (Wellman, Cross, & Watson, 2001). The assertion that understanding of belief represents a paradigm case of conceptual change is qualified by evidence of early sensitivity in indirect tasks. Three-year-olds show correct anticipatory looking to the location where Maxi thought the object was despite incorrectly answering that Maxi would look for the object in the new location (Clements & Perner, 1994; Low, 2010). Recent studies employing violation-of-expectation and helping tasks suggest that infants as young as 13–18 months of age already represent others’ false-beliefs about an object’s location, content, and property (Baillargeon, Scott, & He, 2010; Buttelmann, Over, Carpenter, & Tomasello, 2014). There may even be an early adaptation to track facts relevant to others’ false-beliefs. Wang, Low, Jing, and Qinghua (2012) found that individual Mainland-Chinese kindergar- teners’ gaze anticipations showed coherence over a variety of object-location tasks in which false-beliefs were induced via different kinds of behaviour. Barrett et al. (2013) *Correspondence should be addressed to Jason Low, School of Psychology, Victoria University of Wellington, PO Box 600, Wellington 6140, New Zealand (email: Jason.Low@vuw.ac.nz). DOI:10.1111/bjop.12121 1