Daniel Freudenthal

I’m a psychologist with an interest in the simulation of cognitive processes, in particular language acquisition. The main focus of my work is on MOSAIC, a computational model that is used to investigate how children’s early multi-word speech is shaped by the statistical properties of the language they hear. Simulations with MOSAIC have revolved around children’s increasing ability to correctly inflect verbs. Across a range of languages, MOSAIC produces the same types of errors at rates comparable to those in children because it is sensitive to the distributional statistics of the input to which it is exposed in a manner that places great weight on the last items in the speech stream (i.e. has an utterance-final bias, or strong recency effect).