|
|
EXTRACTING REFERENTIAL INFORMATION DURING ON-LINE PROCESSING
Elsi Kaiser
University of Southern California
Friday, February 3, 3:30 PM MJH Rm 126
Sponsored by the Stanford Humanities Center/Mellon Foundation
Graduate Research ProgramDifferent languages encode information structure -- in particular, whether a certain referent is given or new information -- with different devices, e.g. articles, word order and intonation. This raises the question of how these different types of encoding are interpreted during real-time language comprehension, and what the consequences are for subsequent processing. In the first part of the talk, I focus on Finnish word order variation. I present experimental results showing that (a) referential information encoded in word order variation is processed rapidly and can even be used to make predictions about upcoming referents, and (b) different word orders create different expectations about what will be mentioned next in the larger discourse and also have an important impact on subsequent anaphor resolution. In the second part of the talk, I discuss collaborative work with researchers at the University of Potsdam investigating whether an arguably more subtle type of information-structural marking, intonation, also causes anticipatory effects. Results from two eyetracking experiments on German reveal early sensitivity to both word order and intonation. Strikingly, the magnitudes of the effects are very similar in the two experiments. As a whole, these results suggest that different word orders have crucially different impacts on processing, and that the human sentence processor makes quick and efficient use of the information encoded not only in word order variation but also in different intonational contours. This seems to point towards a very efficient interaction between different aspects of language. 
|
