Abstract
In 4 experiments we examined whether reasoning about spatial scenes acquired through narratives, can be influenced by people’s physical movement in space during encoding. Participants read a narrative that described the geometry of a store and objects located at canonical and diagonal axes.
In Experiment 1, participants encoded locations by walking towards every object described in the narrative and then returned to the center of the room. After an instruction to visualize all objects, they physically rotated 90o to the right of their initial facing orientation and carried out a series of trials that entailed pointing to objects from imagined perspectives. In follow-up experiments we
gradually reduced the extent of physical movement during learning. In Experiment 2 participants rotated their body to the direction of the described objects instead of walking to them, while in Experiment 3 they only rotated their heads to these directions. In Experiment 4, we also eliminated
the instructions to visualize the objects. Results from the first three experiments revealed a performance benefit for responding from the perspective that participants physically occupied at testing, suggesting that they had updated their representation during the physical rotation to that
orientation. However, results from Experiment 4 showed that many participants did not exhibit the benefit for the physical orientation. Interestingly, only participants who claimed that they had linked the described environment with the immediate space of the laboratory exhibited such a benefit. These findings indicate that (1) an influence from one’s physical orientation on reasoning about described environments is present if the remote environments are linked with participants’ sensorimotor framework that typically maintains immediate spatial information (2) visualization instructions are sufficient to produce such a link.