Objects and Substances in Vision, Language, and Development
MetadataShow full item record
In this dissertation, I explore how linguistic representations (e.g., lexical and sentential meanings) interface and interact with the innately-specified core cognitive representations. In particular, I focus the interactions between lexical items that refer to objects versus substances (i.e., count-nouns, like ‘cow’, versus mass-nouns, like ‘beef’) and the universally shared non-linguistics quantity representations of approximate number and surface area. I find that while non-linguistic cognition differentiates between number and surface area quantification in terms of representational content (i.e., independent Weber fractions), these two quantification systems share a common representational format (i.e., Gaussian tuning curves on ratio scales). The interface between these systems and linguistic representations respects both this difference and this commonality: count- and mass-nouns tap into the distinction in content, with count-nouns mapping to number quantification, and mass-nouns primarily to area quantification, while the lexical meaning of ‘more’ interfaces with the common format, allowing children to immediately learn that ‘more’ refers to count- and mass-noun quantification. I demonstrate this through six experiments with both adults and children by utilizing the methods of formal semantics, psychophysics, eye-tracking, and cross-sectional developmental psychology. This work contributes to the broader issue of the relationship between language and thought, and attempts to form a bridge between the traditionally disparate fields of vision, language, and development.