2. From Language Structure to Language Processing
An Introduction to Relational Network Theory - History, Principles, and Descriptive Applications - Adolfo Martín García
Adolfo Martín García [+ ]
Institute of Cognitive and Translational Neuroscience (INCYT, Argentina) and the National Scientific and Technical Research Council (CONICET, Argentina)
Description
This chapter introduces the terminology, notational conventions, and diagrammatic representations used in Relational Network Theory. Relational networks are presented as dynamic systems of information processing. They are based on interconnected nodes of varying strength which operate bidirectionally and in parallel. The nodes allow for activation patterns to spread through several linguistic strata, each subsuming either elements of similar composition or tactic relations. Speech production and comprehension phenomena are illustrated, including anomalous and idiomatic examples. The chapter also introduces the notions of connection strengthening and threshold adjustment, which can account for language learning throughout the lifespan. Finally, network-based accounts are provided for complicated performance phenomena such as context-driven lexeme selection, slang expressions, and creative variants of well known fixed phrases.