Determining the Optimal Environmental Information for Training Computational Models of Lexical Semantics and Lexical Organization

Experiential theories of cognition propose that the external environment shapes cognitive processing, shifting emphasis from internal mechanisms to the learning of environmental structure. Computational modelling, particularly distributional models of lexical semantics (e.g., Landauer & Dumais, 1997) and models of lexical organization (e.g., Johns, 2021a), exemplifies this, highlights the influence of language experience on cognitive representations. While these models have been successful, comparatively less attention has been paid to the training materials used to train these models. Recent research has explored the role of social/communicatively oriented training materials on models of lexical semantics and organization (Johns, 2021a, 2021b, 2023, 2024), introducing discourse- and user-centred text training materials. However, determining the optimal training materials for these two model types remains an open question. This article addresses this problem by using experiential optimization (Johns, Jones, & Mewhort, 2019), which selects the materials that maximize model performance. This study will use experiential optimization to compare user-based and discourse-based corpora in optimizing models of lexical organization and semantics, offering insight into pathways towards integrating cognitive models in these areas. Experiential theories of cognition assert that cognitive processing is shaped by the external environment, primarily through the learning of environmental structure. Computational models demonstrate this influence, emphasizing the impact of language experience on the forming of cognitive representations. Recent research in this area has explored the effect of social/communicative training materials on these models, yet the optimal training material type for each system remains uncertain. This study utilizes a machine learning framework to compare the impact of different corpora on computational models of lexical organization and lexical semantics, providing important clues about how to integrate these model types into a larger framework.