User:John Pate
I am a student at the Ohio State University pursuing a B.A. in Linguistics with minors in Cognitive Science and Philosophy. Concurrently with the B.A., I am pursuing an M.A. in Linguistics through an accelerated double-degree program.
Specifically, I am interested in certain strong statements about linguistic theories that have been made on the basis of an argument called the "Poverty of the Stimulus." This argument holds that the linguistic signal does not contain very much information about the structures it specifies, which would presumably mean that infants, who very apparently succeed in learning those structures, must come "pre-programmed" with strong expectations about what language is like. However, since the representations used in linguistic theories typically abstract away from the actual linguistic signal, it is far from clear just how poverty-stricken the stimulus might be. To assess the information content of the linguistic signal, I have begun looking at the information conveyed by variation in the long-term statistical distribution of linguistic objects, and also by variation in the acoustic signal. The next step, of course, will be to see if children are, in fact, sensitive to the sorts of variation employed by models.