230177 Models of First and Second Language Acquisition (S) (SoSe 2026)

Language acquisition – the process of learning a language through interactions within a communicative setting – is surprisingly difficult to describe theoretically. The scientific study of language acquisition lies at the intersection between psychology, cognitive science and linguistics. It has also long been of interest in the field of artificial intelligence and computational linguistics. In this seminar, we will focus on the question how first and second language acquisition can be modelled computationally. Empirical studies with human infants and older language learners are difficult, time-consuming and expensive, and may even be impossible to carry out with respect to some specific research questions. Computational models offer a tool for systematic analysis and simulation of different aspects and requirements of the learning process and developmental issues (learner models), the interactional aspects (environmental models) and features of the available input and the resulting acquired knowledge (input and output data models).

A basic knowledge of (computational) linguistics is assumed. We will, however, introduce and discuss relevant basic concepts like speech vs text – i.e. continuous vs symbolic representations of language; first vs second language acquisition; learnability; or verbal vs computational models.

In this seminar, different theoretical frameworks or computational approaches will be discussed based on selected papers which will be prepared and presented by the participants and discussed within the seminar. A list of potential readings will be provided at the beginning of the seminar. This seminar should provide an opportunity of lively interactions and discussion. Active participation and contributions from the participants are expected and highly encouraged.