In a text, several objects may linguistically be localized relative to a protagonist, the orientation of which remains unchanged.
Such a static description yields a distinctive description perspective. On the other hand, the same layout can be described with the protagoist continually reorienting, while only objects are described
which he is currently facing. Such a dynamic description does not yield an invariant description perspective. The special character of the description perspective in egocentric mental
models was experimentally confirmed. Two accounts are discussed to explain it. The functional account considers a difference
in salience of egocentric directions before and after a first imagined reorientation. The representational account suggests
that only an invariant orientation allows to maintain a mental image in addition to the mental model.
This research was supported by the Deutsche Forschungsgemeinschaft (DFG) in the project “Modelling Inferences in Mental Models”
(Wy 20/2-2) within the priority program on spatial cognition ( “Raumkognition” ).