Towards solving the puzzle of early belief-ascription

Pierre Jacob (Institut Jean Nicod)

Much recent evidence based on spontaneous tasks strongly suggests that preverbal human infants can track others' false beliefs. However, twenty years of empirical research have earlier shown that not until they are at least 4 year old can the majority of human children pass elicited-response false-belief tasks. For example, when asked where an agent with a false belief about an object's location will look for it, most 3-year-olds who know where the object is will point to the object's actual location, not to the empty location. The puzzle is: why do 3-year-olds fail elicited-response tasks if preverbal human infants are able to represent another's false belief? I'll try to suggest a solution to the puzzle.

Admission Free. All welcome.