False Belief & Scalar Implicature

This week, we read two studies: Barner et al. (2010) and Stiller et al. (2011). Barner et al. (2010) explores scalar implicature, suggesting that children struggle with scalar implicature because they fail to understand relevant alternatives for a given scale (e.g. 'all' as an alternative to 'some'). Stiller et al.(2011) shows that while young children struggle with scalar implicature, the underlying inferential mechanisms needed for scalar implicature may be present much earlier than was previously assumed.

The dichotomy between these two studies brought to mind material I recently learned about in a class I am taking on developmental psychology (psych 60). One of the TAs gave a great lecture on Theory of Mind as it pertains to development. Namely, we talked about young children's struggle to reason about false belief. The Sally-Anne task was what researchers first used to show that children had trouble reasoning about false belief until around age 4. The Sally-Anne task presents children with the following story. Sally has a basket. Anne has a box. Sally has a marble that she puts into her basket. Sally goes out for a walk. While Sally is out on her walk, Anne takes the marble out of the basket and puts it into her box. Sally returns from her walk eager to play with her marble. Children are then asked where they think Sally will look for her marble. Until around age 4, children fail to understand that Sally will look for the marble where she falsely believed it to be (i.e. in her basket); instead, they reason about where the marble actually is and answer that Sally will look in the box. However, this task places a lot of attentional, memory-related, and linguistic demands on children that go beyond measuring their ability to reason about false belief.

Southgate, Senju, & Csibra (2007) show that by removing these task demands (especially linguistic ones), we see that children as young as 25 months reason correctly about false belief. Instead of a story given in language and a question as their task, children that participated in this study were shown a person standing over and watching a puppet show. In this puppet show, there is a puppet and two boxes. The child sees the person watch the puppet put an object in one of the boxes. The child sees the person look away while the puppet secretly moves the object out of the box where the person last saw it. When the person looks back, the children see a flashing light that signals that the person is going to reach through one of two windows to get the object out of the box that they believe it is in. Where the child looks is measured. This study found that children as young as 25 months look at the window corresponding to where the person falsely believes the object to be rather than the actual location of the object.

I see interesting similarities in how the Southgate, Senju, & Csibra (2007) study removed task demands on children to show that the underlying ability to reason about false belief is present much earlier that was previously thought and how the Stiller et al. (2011) study showed that the underlying inferential mechanisms needed for scalar implicature is present much earlier than was previously assumed despite the inability to understand scales constructed from conventional linguistic factors. I wonder if a study could be designed to show that the underlying inferential mechanisms needed for scalar implicature are present even earlier by removing more task demands from the tasks in the Stiller et al. (2011) study. This idea of children being able to successfully reason about various phenomena without understanding them in a linguistic capacity raises interesting questions about how children transfer their non-linguistic knowledge of phenomena like false belief and scalar implicature to understand them and reason about them linguistically.