Computational Modeling
Ping Li, Xiaowei Zhao · 2017
Computational modeling has played significant roles both for psycholinguistic theorizing and as a research tool. Computational models offer particular advantages in dealing with complex interactions between variables that are often confounded in natural language situations. This chapter provides an overview of two approaches of computational modeling in psycholinguistics: the probabilistic approach and the connectionist approach. It discusses the assumptions and rationales behind each approach, along with methodological challenges related to both. In particular, the chapter discusses how modeling is conducted by illustrating these approaches with examples, and with their applications in psycholinguistic studies by focusing on co-occurrence based semantic representation and lexical development in children and adults. Depending on the simulation goals and tasks, the apparatus used for computational linguistic modeling could be as simple as one personal computer equipped with any type of programming language. The chapter discusses surveys some basic algorithms and discusses practical considerations related to their implementation.