The question of knowledge induction, i.e. how children are able to learn so much about, say, what words mean without any explicit instruction, is one that has vexed philosophers, linguistics, and psychologists alike. Indeed, inferring the vast amount of knowledge that children learn almost effortlessly from an apparently “impoverished stimulus” seems paradoxical. The Latent Semantic Analysis model (Landauer & Dumais, 1997) is a theory for how meaning representations might be learned from encountering large samples of language without explicit directions as to how it is structured. To do this, LSA makes two assumptions about how the meaning of linguistic expressions is present in the distributional patterns of simple expressions (e.g words) within more complex expressions (e.g. sentences and paragraphs) viewed across many samples of language…. An Introduction to Latent Semantic Analysis