This project represents a first step towards thinking about how iterated learning experiments deal with the problem of open-ended meaning spaces. Current studies tend to use meanings that are discrete, finite, pre-specified, and low-dimensional (e.g. Kirby, Cornish, & Smith, 2008). However, many of these properties are conspicuously absent in natural language, which is typified by open-ended meaning structure. While recent work has begun to explore the effects of continuous spaces (e.g. Perfors & Navarro, 2011), we present an experiment that explores the cumulative cultural evolution of category structure in a meaning space that is not just continuous, but contains no reoccurring items (every meaning is novel), is not specified by the experimenter a priori, and is high-dimensional.
展开▼
