Boundary Conditions for Language in Biological and Artificial Neural Systems

Explore the computational properties of human language and their implications for biological and artificial neural systems in this 54-minute lecture by Andrea E. Martin, Lise Meitner Group Leader at the Max-Planck Institute for Psycholinguistics. Delve into the fundamental puzzle of representing information that is both algebraic and statistical, examining the computational requirements and the role of neural oscillations in language processing. Gain insights from cognitive neuroimaging data and computational simulations as Martin outlines a developing theory that integrates linguistics and psycholinguistics with neural computation. Investigate topics such as hierarchical relationships between linguistic units, the unbounded ability to combine smaller units into larger ones, and the boundary conditions for artificial systems interfacing with human language. Follow the lecture's structure, covering introduction, brain rhythms, experimental design, theoretical models, and key takeaways to understand the complex interplay between language and neural systems.