The primitives of the indexical space

This is a follow up on a previous experiment pre-registered here: https://osf.io/r684g. The main goal of this project is to test whether context-dependent expressions of different classes (locatives and personal pronouns) share a common semantic basis: Can we analyse both personal and spatial indexicals in terms of an analogous set of features? In the previous experiment (henceforth, Experiment 1), we tested whether, when learners are trained on one partition for pronouns, they generalize it to locatives. The preliminary results of Experiment 1 suggest that this is indeed the case: After being trained on a participant partition of the person space, learners are more likely to infer the same partition for the locative space. In Experiment 2 we aim at testing whether this bias is bidirectional: Do learners also generalize a learned locative partition to a person partition?