Seeing the world through the eyes of an avatar? Comparing perspective taking and referential coding

Previous studies have shown that users spontaneously take the position of a virtual avatar and solve spatial tasks from the avatar's perspective. The common impression is that users develop a spatial representation that allows them to "see" the world through the eyes of the avatar-that is, from its virtual perspective. In the present paper, this perspective taking assumption is compared with a referential coding assumption that allows the user to act on the basis of changed reference points. Using a spatial compatibility task, Experiment 1 demonstrated that visual perspective of the avatar was not the determining factor for taking the avatar's spatial position, but that its hand position (as the reference point) was decisive for the spatial coding of objects. Experiment 2 showed, however, that if the participant's hand position was not corresponding with the avatar's hand positions, the spatial referencing by the avatar's hands expired, thereby demonstrating the limits of referential coding. Still, the present findings indicated that referential coding may be at the base when taking the avatar's perspective. Accordingly, any study in perspective taking needs to consider and evaluate possible mechanisms of referential coding. (PsycINFO Database Record (c) 2020 APA, all rights reserved).