Project B6 investigates the types of common ground updates that can be triggered by syntactic structures, focusing on non-canonical inversion structures such as English locative inversion. The hypothesis is that these structures deliver a scenic update, specifying perceptual perspective, which needs to be captured as a perspective feature in an adequate model of CG. As the occurrence of these structures seems to depend upon text type, the project will explore CG management in asymmetric discourse types. The approach is to employ both corpus data and experimentation to build an account informed by grammatical theory.