Theories of Distributional Models, Embodied Cognition and Hybrid Approaches on Acquisition of Abstract Concepts in Human Cognition

Abstract

Understanding human comprehension of the origin and purpose of lexical words beyond their mere sensory features (e.g. how they look when they are written or how they sound when they are pronounced), also known as the “symbol grounding problem” remains one of the most profound challenges in cognitive sciences. It is particularly challenging for abstract concepts because they lack tangible references in the surrounding world for intuitive and universal understanding of their true meaning. In cognitive science, there are two major theories: the distributional model and embodied cognition model. However, the distributional model struggles with how words become meaningful, and the embodied approach is limited in explaining highly abstract ideas. Since both major approaches have critical gaps in their explanations of abstract concept comprehension, an alternative approach is needed to bridge these gaps and provide a more comprehensive understanding of human concept acquisition. This review examines how the distributional model, the embodied approach, and the hybrid model explain abstract concept acquisition, by referencing relevant empirical evidence. The review also discusses the strengths and limitations of both models, explaining why it is important to adopt an alternative approach.