Proceedings of the Annual Meeting of the Cognitive Science Society

Spatial demonstratives exist in all languages, but currently there is much debate regarding the parameters that affect their use both between and within languages. In this work, we explore ‘perspective taking' as a means of accounting for variation in demonstrative use both between and within languages. Analysing primary and secondary data, we test the effects of egocentric distance and addressee position on demonstrative production in speakers of two languages with two purportedly different demonstrative systems: English and Japanese. Based on individual differences between speakers, we propose a framework unifying different theoretical accounts of demonstrative systems in which demonstratives require a spatial reference frame to be chosen prior to the application of a range of routines to select the appropriate term in a given context.