Individual Differences in Spatial Orientation Modulate Perspective Taking in Listeners

Abstract

Previous research suggests that individuals exhibit consistent tendencies towards taking their own (an egocentric) or their partner's (an othercentric) spatial perspective. In addition, several factors such as spatial orientation ability, inhibitory control, and social preferences, have been found to mediate these perspective taking tendencies. However, these factors have not been studied together in the context of a single task. The present study explores these individual differences together in spatial perspective taking, using a task of simulated interaction in which listeners can choose to interpret an ambiguous spatial utterance egocentrically or othercentrically. We use a datadriven approach of latent profile analysis to classify participants into subgroups based on their spatial perspective taking tendencies. Our results show that stable subgroups of participants can be identified who differ in their perspective taking tendencies. This behaviour also correlates with a measure of listeners' spatial orientation ability, but not their inhibitory control or social preferences. Our results can be interpreted within a framework that views spatial perspective taking as an embodied cognitive process of a mental reorientation of the self relative to the environment, providing insight on the nature of the mechanisms underlying this operation.

Perspective Taking Reflects Beliefs About Partner Sophistication: Modern Computer Partners Versus Basic Computer and Human Partners

“…This would be consistent with the idea of an update in technological expectations from the time of their study to today. We also note that in a separate study utilizing the same design with a computer partner but employing a data‐driven method of classifying participants rather than predefined manual thresholds, we observed a similar split of three responder groups across participants (40/30/30% egocentric/mixed/othercentric; Loy & Demberg 2023). The threshold for mixed responders estimated by that model (37.5–75%) is also similar to Duran et al.…”