- Who Brian Dillon
- When Tuesday / Thursday 10-11:15AM
- Where ILC N470
Course overview: This course is about prediction. Over the last decade or two, the claim that language comprehension relies on predictive processing has gone from a controversial claim to something of a central tenet in psycholinguistics. On this view, language comprehension is facilitated by predictive processing that plays out at multiple different levels of representation, from preactivation of semantic features to detailed preactivation of syntactic structures right down to the prediction of specific linguistic forms. The reason we are such skilled, rapid sentence processors, on this view, is in good part because we are skilled at generating expectations about upcoming linguistic events. In view of this general hypothesis about sentence comprehension, the course has two twin goals.
The first goal is theoretical/empirical. We will attempt to understand our work as psycholinguists in light of this broad hypothesis. We will ask whether (or perhaps, to what degree) fine-grained linguistic representations and constraints guide online predictive processing. We will cover broad issues, and review key findings and controverises. We will do a deep dive on two specific subareas where predictive processing matters: managing expectations at the level of the discourse, and actively predicting syntactic structure. Along the way, we will also engage with recent empirical debates over how reliable key findings in this literature are, and reflect on these issues as they impact the work we will engage with in this class.
The second goal is methodological: we will attempt to get the ERP lab in our department up and running for sentence comprehension experiments. Approximately one-third of in-class time over the semester will be devoted to learning the ERP methodology, and enrolled students will be expected to spend class time (and out of class time!) in the ERP lab working on coding experiments, setting up lab protocols, learning to run participants, and collecting pilot data. The reason for this methodological focus will become evident over the course of the semester: in discussions about the predictability of linguistic features and form, the event-related potentials (ERP) paradigm inevitably takes a leading role, with the beloved N400 in center stage.
Enrolled students will be expected to:
- Attend biweekly class meetings, read assigned readings, and come ready to discuss. The course will be less “lecture” style, and more “discussion” style: you shouldn’t expect too much in the way of formal lectures, but rather, in-depth discussion of our readings.
- Collaborate on the construction of Google doc discussion summaries to summarize highlights of our class discussion, with the goal of identifying areas of interesting/fruitful future research through discussion.
- Give two presentations throughout the semester. One presentation will be of a paper that bridges the students’ individual interests and the theme of the class: this need not come from the planned readings. The other presentation will be a tutorial session on one aspect of running an ERP experiment (e.g. how to code an experiment, how to run a participant, how to visualize data using the software tools in the lab).
- Collaborate with the class participants as a group to design and pilot a class ERP experiment.
- Write a final term paper on a topic of the students’ own choosing, under the broad rubric of prediction, and present the result of their research at the end of the semester. Students will be especially encouraged to find common ground between predictive comprehension and their current research interests outside of this class; come meet me early on to discuss this.
ERP-based studies are marked with *
|Tu 1⁄22||Comprehension through prediction (1)||Kuperberg & Jaeger (2016), Pickering & Gambi (2018)|
|Th 1⁄24||Comprehension through prediction (2)||Kutas et al. (2012), Delong et al (2005)*; Optional: Wicha et al. (2004)*|
|Tu 1⁄29||Primary literature preface: Issues of replicability + reliability||Nieuwland (in press); especially pp. 1-47, Nieuwland et al (2018)*; Optional: Mante’s blogpost on RetractionWatch|
|Th 1⁄31||Lexical predictability , Carolyn’s handout||Staub (2015), Smith & Levy (2013), Staub et al. (2015)||Carolyn presents Staub et al (2015)|
|Tu 2⁄5||Phonological predictability, Katie’s handout||Hall et al (2018)||Katie presents Hall et al. (2018)|
|Th 2⁄7||Phonological predictability||Cohen-Priva (2017)||Max presents Cohen-Priva (2017)|
|Tu 2⁄12||Prediction and discourse management (presuppositions)||Shetreet et al. (2019)*||Alex presents Shetreet et al. (2019)|
|Th 2⁄14||Prediction and discourse management (presuppositions/implicature)||Xiang & Kuperberg (2015)*||Erika presents Xiang & Kuperberg|
|Tu 2⁄19||NO CLASS||-|
|Tu 2⁄21||Prediction and discourse management (discourse relations)||Scholman et al (2017), van Bergen & Bosker (2018)|
|Th 2⁄26||Prediction and syntactic analysis (fillers & gaps)||Atkinson et al. (2018)|
|Th 2⁄28||Prediction and syntactic analysis: dependency length and resumption)||Wagers & Phillips (2014), Chow & Zhou (2018)|
|Tu 3⁄5||Prediction and syntactic analysis (gaps & resumption)||Chacon (submitted)|
|Tu 3⁄7||NO CLASS (CANCELLED)|
|Tu 3⁄12||SPRING BREAK||-|
|Th 3⁄14||SPRING BREAK||-|
|Tu 3⁄18||Prediction and syntactic analysis (argument structure)||Omaki et al. (2015), Staub (2007)|
|Th 3⁄21||Event-related potentials: What are they good for?||Van Petten & Luka (2012)*|
|Tu 3⁄26||Uncertainty and prediction in phonological processing||Gwilliams et al (2018)||Bethany presents Gwilliams et al. 2018|
|Th 3⁄28||NO CLASS (CUNY!)|
|Th 4⁄4 9AM - 11:15||P300 & Oddball paradigm: Lab Session #1, Erika & Brian||Luck, 2014; Chapter 1,Net Station Acquisition, Technical Manual (2003) Chapter 1||Erika presents Luck, Chap. 1|
|Th 4⁄11 9AM - 11:15||P300 & Oddball paradigm: Lab Session #2, Max & Carolyn||Luck, 2014; Chapter 2,Net Station Acquisition, Technical Manual (2003) Chapter 1||Max & Carolyn present Luck, Chap. 2|
|Th 4⁄18 9AM - 11:15||P300 & Oddball paradigm: Lab Session #3, Bethany & Katie||Luck, 2014; Chapter 3,Net Station Acquisition, Technical Manual (2003) Chapter 1||Bethany & Katie present Chap. 3|
|Th 4⁄25||Baselines + Filters||Tanner et al. (2015)*; Tanner et al. (2016)*|
|Tu 4⁄30||Student presentations|
|Th 5⁄2||Student presentations|