The event will be held at Hirao Memorial Seminar House of Konan University.  Please note that the venue, Hirao Memorial Seminar House, is OFF CAMPUS, a few kilometers away from Konan University’s main campus.  See the link below for directions to the venue.

Douglas Roland "Prediction as a consequence of integration"

It is widely agreed upon that comprehenders actively make predictions about upcoming linguistic material. However, there is much less agreement about the nature of the predictions, or even what constitutes a prediction on the part of the comprehender. I will consider evidence from reading time experiments examining the comprehension of structures such as relative clauses and instrument phrases to argue for a view where most “predictions” about upcoming material arise as the consequence of decisions made about how to integrate the information that has already been encountered.

Hannah Rohde "Discourse dependencies:  Raising and answering questions in discourse"

When we study what makes a text or a conversation coherent, we can be said to be identifying the implicit or explicit questions that are raised and answered by subsequent sentences in the discourse. Such an approach casts discourse relations, and the connectives that help express them, as local connections that contribute to a larger structure defined by an architecture of questions and subquestions. In that framework, one can ask what cues give rise to particular questions in comprehenders' minds, how long expectations for an answer to such questions persist, and how closely comprehenders' expectations align with speakers' decisions about what questions to address as the discourse proceeds. In this talk, I present collaborative work targeting these open questions. The goal is to see how a mix of psycholinguistic and corpus-based methods can provide a window into the comprehension and production of structured discourse.

To the extent that a discourse is comprised of local and non-local question-answer relationships, one can ask how well comprehenders track those long-distance dependencies. I present a set of reading-time studies testing the role of contrastive topics and the discourse markers "On the one hand"/"On the other hand" in signalling upcoming discourse structure. In passages that set up a contrastive pairing ("Linda and Bob have very different views about...") and mention one component of the pairing ("Linda thinks..."), comprehension is facilitated for content that addresses the unanswered question (What about Bob?) when the dependency distance is local. With the discourse marker "On the other hand", comprehenders appear to anticipate more than just the simple "On the other hand" counterpart; rather, they keep track of embedded constituents and establish non-local dependencies. The results support memory-based models of processing by showing that discourse dependencies, while they are built as fine-grained representations, are not unbounded in real-time processing.

To look beyond lab experiments with constructed passages, I also present a corpus-based approach to assessing the content and cues that guide comprehenders' guesses about upcoming material in naturally occurring discourse. Using a corpus of TED talks, questions were elicited at different points in the talks from crowdsourced non-expert annotators; annotators were asked to indicate whether and where their questions are answered. The corpus contains pre-existing annotations of local discourse relations, namely the explicit and implicit discourse relations that hold between clauses (Penn Discourse Treebank labels). I will describe preliminary results on the rates at which questions are answered and the relationship between question predictability and implicitness.

Both psycholinguistic studies and discourse-annotated corpora have the potential to bring new evidence to bear in our understanding of discourse coherence, and the examples presented in this talk will hopefully showcase this potential and invite new directions.

Hannah Rohde "Informativity in cooperative communication"

In studying meaning in communication, a question arises as to which meanings are favored by interlocutors. While a range of candidate meanings may be possible and even plausible, how do speakers select which meanings to communicate and how do listeners make guesses as to the most probable meaning when trying to recover what a speaker intends or when anticipating what a speaker will say next?

In this talk, I compare two hypotheses for ranking candidate meanings that a speaker might contribute to a discourse. Under one account, listeners' guesses simply reflect the probability that different meanings hold true: Speakers are taken to generate sentences that describe the world they see and listeners come to expect sentences about the typical situations speakers find themselves in. A second account combines this component with a component capturing the likelihood that a speaker, knowing some meaning to be true, would select that meaning as one worth conveying to a listener in an utterance. I present a series of psycholinguistic studies measuring listeners' awareness of speakers' production likelihoods. For example, although bananas are prototypically yellow, speakers rarely mention this yellowness in their utterances. In an eye-tracking study measuring anticipatory looking, listeners who hear a speaker use an ambiguous color adjective are found to anticipate subsequent mention of an object for which that color is less typical in the real world.  Similarly, in a study on comprehenders' guesses of what a speaker will say next, participants are shown to disprefer upcoming material that describes situation-typical outcomes.  Further studies target properties of the speaker and show that the more aware comprehenders are of the speaker as an intentional knowledgeable communicator, the more informative they expect the speaker's contribution to be and the more inferences they draw from the speaker's content selection.  This work raises questions about the ways in which learners (humans, machines) acquire information about the world if their linguistic input favors the description of real-world-atypical content.

The findings highlight the importance of establishing not only which meanings are possible and how they are derived, but also which meanings are probable as likely contributions to coherent discourse, despite -- or perhaps as a result of -- denoting non-typical situations.