Tuesdays & Thursdays, 2-3:20pm in SSPA 3182
Instructor: Lisa Pearl, Department of Cognitive Sciences, SSPB 2243
Email is the best way to reach me outside the classroom.

Announcements:

  • 1/8/08: Welcome to the class webpage!
  • 1/10/08: Because Political Science made a mistake with scheduling, we're being the good guys and moving permanently to SSPA 3182, where we met today. We are nice.

Language is an incredibly complex system of knowledge. Not only are there multiple levels of representation - sounds and words and phrases and meanings - but within a given level, even simple output forms can be derived from multiple interacting pieces of knowledge. Yet as speakers of any given language, we are often blissfully unaware of how much we need to know in order to be able to communicate with language.

Nonetheless, this is precisely the knowledge children must acquire. And their task is not simple. The patterns of knowledge can be difficult to discern from the available input and, to top it off, the data children learn from is often ambiguous and full of exceptions anyway. Yet despite all this, all normally-developing children learn their native language nearly effortlessly, generalizing from noisy input in very specific ways. The degree of proficiency attained by very young children in their native language is almost never achieved by adults who are far more cognitively developed. How is this possible?

In this class, we delve into the process of language acquisition, exploring the way in which infants and very young children unconsciously uncover the rich systematic knowledge of their native language. There are arguments for both general-purpose learning strategies and innate biases that young children bring to the language-learning task. We focus on both experimental methods and computational studies that quantitatively investigate the "how" of language acquisition.

We will be reading selections from three books, as well as articles. These can be found on the readings section.