I know you don’t know I know… Children’s use of recursive mental state reasoning for peer coordination

Humans have to coordinate their behavior with others. While we do this with great efficiency as adults this facility is not present from birth and can hide significant psychological challenges: In order to coordinate successfully, we must often reason about what others are likely to do. A problem arises since others will be reasoning similarly about what we will do (which depends on what we think they will do, and so on). What is required is some kind of “meeting of minds” (Schelling, 1960), that is, some form of shared knowledge. We hypothesize that the ability to reason recursively about mental states can facilitate coordination by helping us determine the extent to which we share knowledge with others. In my talk, I will present data showing that this is already within the capacity of 6-year-old children. Participants (N = 104) were presented with acoordination task requiring them to insert a ball into the same of four boxes as a partner in order to access a reward. One saliently marked box always contained a larger reward than the others making it the obvious solution at training. At test, however, children were informed that the largest reward was mistakenly placed into a different box (i.e. not the saliently marked one). In addition, we manipulated what children thought their partner knew about the erroneous reward placement. The result was that in two experimental phases children successfully adjusted their decisions to a partner’s false belief (phase 1) and a partner’s false belief about their own belief (phase 2) about the location of the largest reward. By age six, children can thus use recursive mental state reasoning to aid their coordination efforts.