Similarities and differences in visual and spatial perspective-taking processes.

Processes for perspective-taking can be differentiated on whether or not they require us to mentally rotate ourselves into the position of the other person (Michelon & Zacks, 2006). Until now, only two perspective-taking tasks have been differentiated in this way, showing that judging whether something is to someone's left or right does require mental rotation, but judging if someone can see something or not does not. These tasks differ firstly on whether the content of the perspective is visual or spatial and secondly on whether the type of the judgement is early-developing (level-1 type) or later-developing (level-2 type). Across two experiments, we tested which of these factors was likely to be most important by using four different perspective-taking tasks which crossed orthogonally the content of judgement (visual vs. spatial) and the type of judgement (level-1 type vs. level-2 type). We found that the level-2 type judgements, of how something looks to someone else and whether it is to their left or right, required egocentric mental rotation. On the other hand, level-1 type judgements, of whether something was in front of or behind someone and of whether someone could see something or not, did not involve mental rotation. We suggest from this that the initial processing strategies employed for perspective-taking are largely independent of whether judgements are visual or spatial in nature. Furthermore, early developing abilities have features that make mental rotation unnecessary.