We're pleased to welcome Dr. Volya Kapatsinski, Professor & Director of Graduate Studies Linguistics, University of Oregon, as our next psychology colloquium speaker. Professor Kapatsinski will present his lecture - "How can experiments on language learning contribute to learning theory?" - this Friday, November 22nd at 2-3pm in Xavier 201. All PLU community members are welcome to attend. 

Learning theory aims to discover the mechanisms by which beliefs and behavior, across domains, are changed by experience. Research on language acquisition has been somewhat disconnected from learning theory since the mid-1950s when Noam Chomsky wrote a scathing review of B. F. Skinner’s book Verbal Behavior, which aimed to apply learning theory to language. However, since the mid-1980s, there has been increasing optimism about the applicability of domain-general learning mechanisms to language acquisition. This has resulted in numerous demonstrations that such mechanisms are sufficient to learn something about language that was claimed to require specialized linguistic knowledge and/or specialized linguistic mechanisms of belief updating. Most recently, we can witness the impressive success of large language models using the same mechanisms that can be used to produce sequences of actions in other domains. However, relatively little work has compared the performance of alternative, minimally different learning mechanisms on language learning data, or identified situations in which they make qualitatively different predictions. I will present some examples of such work from my lab, and argue that such work is essential if we are to have language data speak to learning theory, and to find out why successful language learning models are successful.