My work centers around processing at the sentence level. How do human parsers use competing sources of information to make sense of structure? How can we use diverse languages and language experiences to know what pieces of processing architecture are universal and what parts are shaped by experience?