Seminar für Sprachwissenschaft

In my work I focus on the functional and distributional aspects of speech signals. I explore how context and the distribution of uncertainty across contexts affect articulations. Some of the questions I address are: How does learning from speech samples shaped by distributed experience affect articulations across the lifespan? Do lifelong learning and language structure change the way we speak as we get older? How do speakers co-ordinate their expectations and discriminate signals from noise as differences in their experience grow? I am interested in the structure of time interval distributions in speech signals and the way these temporal regularities shape learning, behavior and signal/noise discrimination across the lifespan and across contexts.

I hold degrees in computer science (BSc) and computational linguistics (MA). My background is in human-computer interaction and information design of natural user interfaces for learning environments. I am interested in a variety of topics related to thinking and talking, self-organization in humans and other animals and the evolution of cultural convention in general, and language in particular.

Teaching

SS 2018, 2019
Problem-Solving for Linguists

This course is about language science and introduces topics from a wide variety of disciplines relevant to computational modeling of language as a method of scientific inquiry. The first four sessions provide an overview of ideas underlying language theory, their impact on linguistic terminology, the design of language resources and the operationalization of linguistic abstractions, addressing the strengths and weaknesses of methods and abstractions commonly used in modeling language. In addition to this, we will review ideas and technological developments that have influenced the tradition of computational modeling of aspects of human cognitive performance and discuss them from the perspective of philosophy of science and human cognition. In the main part, we discuss readings on a variety of topics closely connected to language, covering behavior, aspects of decision making, measurement and the fundamental assumptions of quantitative methods. The objective of the course is to give a broader perspective on language, culture and thought, similarities and differences in the evolution of human and animal societies and the particulars of the human brain relevant to language, efficient performance, and organization. The idea behind all of this is that insights from a wide variety of disciplines can help understand and frame language questions in more interesting and productive ways.

SS 2020, 2021, 2022
Information Structure of Conversational Speech

In this class we will talk about the information structure of speech in time and what it can tell us about the way humans learn, process information and contribute to changes in the way we speak individually and collectively - as speaker communities. We will read and discuss literature on the following topics: ways to quantify and describe information structure structural differences in vocal communication of humans and other animals, learning to speak and how speaker experience affects speech production across adulthood, structural differences in vocal communication between humans and other humans, the function of variation and error in human communication and speaker alignment, speech in context and speech contrast distributions, structural differences between speech and text and quantitative methods, computational and theoretical models of human and artificial communication

Publications

Linke, M., & Ramscar, M. (2021). Finding Structure in Silence: The Role of Pauses in Aligning Speaker Expectations. arXiv preprint arXiv:2112.08126.

Linke, M., & Ramscar, M. (2020). How the Probabilistic Structure of Grammatical Context Shapes Speech. Entropy, 22(1), 90. https://doi.org/10.3390/e22010090

Baayen, R. H., and Linke, M. (accepted). An introduction to the generalized additive model. In Gries, S. Th. and M. Paquot (Eds.) A practical handbook of corpus linguistics. Springer, Berlin

Linke, M., Bröker, F., Ramscar, M., & Baayen, H. (2017). Are baboons learning “orthographic” representations? Probably not. PLoS ONE. https://doi.org/10.1371/journal.pone.0183876

Kammerer, Y., & Bohnacker, M. (2012). Children's web search with Google: the effectiveness of natural language queries. In H. Schelhowe (Ed.), Proceedings of the 11th International Conference on Interaction Design and Children IDC '12 (pp. 184-187). New York, NY: ACM Press. 

Kammerer, Y., & Bohnacker, M. (2012). Unterstützung der Informationssuche von Grundschulkindern im Internet: Empfehlungen zur Gestaltung von Kinder-Suchmaschinen.In H. Gapski & T. Tekster (Eds.), Informationskompetenz von Kindern und Jugendlichen. Schriftenreihe Medienkompetenz des Landes Nordrhein-Westfalen (Bd. 14, pp. 51-65). Düsseldorf, München: kopaed.