Belief–desire reasoning in the explanation of behavior: Do actions speak louder than words?

A Second Look at Automatic Theory of Mind

“…Some evidence supports automaticity: after reading about an agent unintentionally approaching an object (e.g., approaching a drawer with perfume in it, while trying to find a hair dryer), participants wrongly endorsed mental state-based explanations of the agent's behavior (e.g., because she wanted her perfume; Wertz & German, 2007), suggesting that participants may have automatically computed the agent's mental states based on the agent's behavior. However, because this methodology requires participants to consider the agent's mental states explicitly during the response phase, it is possible that ToM is triggered by the explicit task and not computed automatically (Back & Apperly, 2010).…”

Theory of Mind in the Wild: Toward Tackling the Challenges of Everyday Mental State Reasoning

“…To examine whether preschool aged children (3- to 5-year-olds) would be sensitive to the approach action cue when competing mental state information is present, the explanation tasks from [57] were made more appropriate for use with young children by simplifying the language, adding illustrations, reading the stories aloud, and requiring children to provide their own explanation for the character’s search action. This mode of response was used because preschool children are susceptible to a “yes bias” [66], making the explanation endorsement task used with adults susceptible to false positives and therefore unsuitable.…”

“…We began to investigate this by looking at whether adult participants would attribute mental states to a character based on a simple approach action cue when the character already had a stated desire for a different object [57]. Adult participants read scenarios describing a character’s interactions with a target object.…”

“…Therefore, it was unclear how preschoolers would handle more complex situations with multiple possible mental state interpretations. Accordingly, Experiment 1 tested preschool children using modified versions of the vignettes from the previous investigation [57] to examine whether young children would generate distracter object responses based on the approach action cue even in the presence of competing mental state information about the target object.…”

“…Therefore, we predicted that an explicit statement that the character was searching for the target object would highlight the mental state content about the target object and make it less likely that participants would include distracter object mental state content in their responses. This desire statement was always present in the previous investigation [57], so to investigate its effect we created two versions of the explanation tasks in the current experiments: one with the desire statement present and the other with the desire statement absent. We examined the effect of the desire statement on the explanation production task used with preschoolers (Experiment 1) and adults (Experiment 2), as well as the explanation endorsement task used in the previous investigation (Experiment 3).…”