This course will provide a hands-on introduction to Generalized Additive Modeling (GAMs). We will introduce methods to identify the best model given the data, and demonstrate how to visualize non-linear effects and non-linear interactions. In addition, we will address several potential problems, including, but not limited to, dealing with the common problem (for time series data) of encountering autocorrelation in the residuals of a model.

Over the last two and a half decades, computational linguistics has been revolutionized as a result of three closely related developments: increases in computing power, the advent of large linguistic datasets, and a paradigm shift toward probabilistic modeling.

This course emphasizes the commonalities among words, idioms and more abstract syntactic patterns in that all are learned pairings of form and function, at varying levels of complexity and abstraction. This emphasis allows us to draw many parallels between language and other cognitive processes such as categorization, parallels that in turn raise the issue of whether language may emerge from a combination of general cognitive abilities, without requiring a unique language faculty.

Memory is an integral part of language comprehension. A firm grasp of language processing therefore calls for an understanding of general theories of memory and the data they are built upon, including data that has historically motivated the fractionation of memory along various lines (time, modality, task, etc). These proposed fractionations have, in turn, affected how linguists conceive and discuss memory retrieval during language processing.

This course explores research on the significance of language variation in shaping thought.  The first unit of the course provides introduces historical and conceptual perspectives on the relation of language and reality that continue to shape our understanding of language variation and surveys early work in anthropology (Boas, Sapir, Whorf) and psychology (Brown, Lenneberg, Carroll) linking language variation to thought.

The nature of a coherent discourse is that the utterances within it do not appear together arbitrarily but, rather, relate to each other in meaningful ways.  The establishment of intersentential coherence relations is hence fundamental to language use and language understanding. This course will introduce students to ongoing research in experimental pragmatics that analyzes the reasoning and inferences that are brought to bear in the establishment of intersentential coherence relations.

This course will cover the basics of adult sentence processing. Beginning with a discussion of foundational assumptions, the course will then take up phrase structure parsing, gap-filling, islands, and ellipsis. It will end with a discussion of the role of Question Under Discussion (Roberts, 2012) and implicit QUD in processing.

The central question for ellipsis resolution is how the phonologically silent material at the ellipsis site is recovered from the antecedent. The answer to this question bears upon some fundamental issues of linguistic inquiry, including whether postulating “silent” syntactic representations is necessary, and what is the division of labor between syntactic and discourse constraints. Against the background of an extensive theoretical discussion on ellipsis, this course examines the existing experimental findings on ellipsis processing. We will primarily focus on three questions.

For over 60 years, speech perception researchers have investigated listeners' ability to resolve the linguistic information encoded in input acoustic signals. More recent investigations have also explored listeners' resolution of the indexical information that is carried in the same signal. Both lines of study (and their important intersection) point toward the remarkable adaptability of perception. Listeners not only achieve perceptual stability across many types of variation, they also use that variation as important information about what a speaker is saying and about who said it.

This advanced course will focus on how speakers encode and navigate linguistic representations in memory. Linguists are impressed by the rich grammatical details that natural languages follow. There is now abundant evidence that speakers and comprehenders show fine-grained control over these details during moment-by-moment speaking and understanding, but how do they do this?