Computational Psycholinguistics

Graduate / Professional

When processing language, the human mind can perform remarkable feats. For instance, we can acquire a language from a small amount of data (thousands of times less data than current systems in artificial intelligence), and we can infer what another person means even when that person’s intended message goes beyond the literal meaning of their words. This course explores how computational modeling can help us characterize our incredible capacity for language learning and processing. We focus on three modeling traditions—symbolic algorithms, Bayesian models, and neural networks—and their application to a range of psycholinguistic phenomena, including parsing, pragmatics, speech perception, word learning, and language acquisition. We also discuss how artificial intelligence can inform theories of human language processing and vice versa. TTh 9am-10:15am