Doctorial Dissertation Research: Sources of Processing Asymmetry in Alternative Word Order Sentences

Human language often allows different ways of saying the same meaning by changing the order of words in a sentence. However, these different word orders are not always understood with equal ease, even when the meaning remains the same. This doctoral dissertation project investigates how people process sentences with flexible word order and examines why some orders are harder to understand than others. It tests two sources of difficulty in understanding flexible word order sentences. One source involves the mental effort needed to connect a phrase that has been moved from its original position (known as filler–gap dependency formation). The other involves a mismatch between what the reader expects to come next and what actually appears (known as prediction mismatch). In addition to training a graduate student, benefits to society include providing innovative training opportunities in research methods including statistical analysis, which supports workforce development for artificial intelligence and other data science industries. To compare how the filler–gap dependency formation and the prediction mismatch affect sentence processing, this doctoral dissertation project uses a combination of online and offline experiments – including self-paced reading and sentence completion tasks – with various types of sentences. The findings deepen our understanding of how people comprehend language, using rigorous experimental methods and high-powered samples, to conduct the research. This project also makes publicly available code used for designing experimental tasks and running data analyses to improve infrastructure for future sentence-processing research. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.