Do 10-month-old infants understand others’ false beliefs?

Abstract

As adults, we know that others’ mental states, such as beliefs, guide their behavior and that these mental states can deviate from reality. Researchers have examined whether young children possess adult-like theory of mind by focusing on their understanding about others’ false beliefs. The present research revealed that 10-month-old infants seemed to interpret a person’s choice of toys based on her true or false beliefs about which toys were present. These results indicate that like adults, even preverbal infants act as if they can consider others’ mental states when making inferences about others’ actions.

Introduction

We interpret others’ behavior by referring to their mental states such as goals, dispositions (e.g., preferences), perceptions, and beliefs (e.g., Carey and Spelke, 1994, Leslie, 1994; Premack and Premack, 1995, Tomasello et al., 2005, Wellman, 1991). A crucial aspect of such “theory-of-mind” understanding is to realize that mental states are representations, not copies, of reality, and can be false (e.g., Gopnik and Wellman, 1994, Wellman and Bartsch, 1988). Researchers have questioned whether young children possess adult-like theory-of-mind. Children younger than 4 years of age fail verbal or “standard” tasks examining their false-belief understanding (e.g., Wellman et al., 2001, Wimmer and Perner, 1983). However, they may fail because these tasks require abilities in addition to knowledge about others’ minds, for example, language and inhibitory-control skills (e.g., Birch and Bloom, 2003, Bloom and German, 2000, Caron, 2009, Leslie et al., 2005, Moses, 2001, Southgate et al., 2007). Recently, children in their second year of life have been found to hold false-belief understanding, using nonverbal tasks (e.g., Buttelmann et al., 2009, Onishi and Baillargeon, 2005, Scott and Baillargeon, 2009, Song and Baillargeon, 2008, Song et al., 2008, Southgate et al., 2010, Surian et al., 2007).

In their ground-breaking study, Onishi and Baillargeon (2005) used a looking-time procedure commonly employed to explore infants’ knowledge (e.g., Baillargeon, 1986, Gergely et al., 1995, Spelke et al., 1992, Woodward, 1998) and demonstrated that 15-month-olds appear to understand others’ beliefs. They found that infants act as if they expect an agent to search for a toy according to where she believes it is located (i.e., location-A where the agent had hidden the toy), as opposed to the toy’s actual location (i.e., location-B, to where the toy had moved during her absence), and respond with prolonged looking when their expectation is violated. Alternative interpretations for these results have been put forward, for example, that infants follow a behavioral rule that an agent looks for an object where she last saw it and respond to violations of this rule (Perner and Ruffman, 2005, Ruffman and Perner, 2005). The debate on whether infants’ responses should be explained in behavioral terms or as evidence for their abilities to attribute to agents mental states such as beliefs and false beliefs has inspired researchers to design belief-inducing situations other than the above change-of-location task (for a review, see Baillargeon, Scott, & He, 2010). For example, Scott and Baillargeon (2009) found that 18-month-olds act as if they attribute to others false beliefs about the identity of an object. In the present research, we examined infants’ responses in yet another type of situations that involve an agent’s beliefs and false beliefs.

Beliefs refer to “the actor’s knowledge, convictions, suppositions, ideas, and opinions” (Wellman & Bartsch, 1988, p. 240). Beliefs can be true or false; justified true beliefs are knowledge, while false beliefs are beliefs incongruent with reality. Importantly, beliefs are derived from perceptual experience, e.g., seeing or hearing about the environment (Wellman & Bartsch, 1988).¹ The connections between beliefs, including false beliefs, and perceptions are implemented in all versions of false-belief tasks to date: an agent who does not witness a critical event holds a false belief. Therefore, seeing leads to knowing, and not seeing leads to not knowing and hence knowing wrong or outdated information. To explore understanding about beliefs, the present research thus built upon prior findings that infants are sensitive to others’ perceptions when interpreting their intentional actions.

We know that from 3 to 5 months, infants who watch an agent repeatedly act towards object-A but not object-B respond as if they take the agent’s actions as evidence for a preference for object-A over object-B (e.g., Luo, 2011, Luo and Baillargeon, 2005, Woodward, 1998). A preference is defined as a dispositional state that helps explain why an agent (i.e., an entity that can detect its environment and exert control over its actions, e.g., Leslie, 1995, Luo et al., 2009) chooses a particular goal-object over another. Infants seem to expect the agent to continue acting towards object-A even when the two objects’ positions are reversed and respond with prolonged looking when the agent acts toward object-B (two-object condition). In contrast, when object-B is absent (one-object condition), infants’ responses are consistent with their having recognized that the agent’s same actions towards object-A do not suggest a preference. They may still interpret the agent’s actions towards object-A as goal-directed. However, this goal-attribution is insufficient for them to infer a preference. They therefore do not respond with increased attention when the agent later acts towards the newly-introduced object-B (e.g., Luo, 2011, Luo and Baillargeon, 2005). Interestingly, when object-B is present but hidden from the agent (hidden-object condition), infants respond similarly as in the one-object condition (e.g., Luo and Baillargeon, 2007, Luo and Johnson, 2009, Sodian et al., 2007). Although the infants can see both objects, they seem to construe the agent’s actions based on how she views the situation: she can only see object-A, and therefore this is essentially a one-object condition to her.

Therefore, infants seem to consider the agent’s perceptions of which toys are present (both object-A and object-B or object-A only) when attributing preferences to her. When the agent cannot see all objects in a setting, infants act as if they acknowledge that her perception is incomplete. This is similar to the situation in which agents hold false beliefs because they have not seen all events that happen in the setting. For example, in Onishi and Baillargeon (2005), 15-month-old infants seem to understand that the agent did not see the relocation of the toy and hence falsely believes that the toy is still in its original location. They therefore act as if they recognize that when the agent wants to get the toy, her false belief guides her to the wrong location.

Would infants also act as if they consider the agent’s beliefs of which objects were present in their preference-attributions? Specifically, would infants attribute to the agent a preference when she acted towards object-A if she believed, truly (she saw and hence she knew) or falsely (she did not see and hence had outdated information), that both object-A and object-B were present in the setting? Would infants not attribute to her such a preference if she believed, truly or falsely, that only object-A was present? The present research examined these questions in two experiments. Experiment 1 used the situation in which the agent falsely believed that two objects were present whereas Experiment 2 used the situation in which the agent falsely believed that only one object was present. Given that evidence of perception-based preference-attributions is well documented in the first year, 10-month-old infants were tested. If positive results were obtained, this would extend false-belief understanding to infants younger than previously demonstrated. Importantly, it would shed light on the debate about whether early understanding of agents’ actions should be cast in mentalistic or behavioral terms. We will return to this issue in the Section 4.