The goal of these experiments is to collect data from self-paced reading studies related to questions of online pragmatic processing. While a lot of research into syntactic and semantic processing has used self-paced reading studies, much less has happened in experimental pragmatics. Here, we are interested in whether participants' pragmatic expectations about felicitous (good, informative, relevant) descriptions of pictures affects the SPR data. We pursue the general hypothesis that pragmatic responders might expect a felicitous pragmatic description and therefore show SPR-lags when they encounter a word that is pragmatically (in the context; not syntactically or semantically) surprising.
We are interested in whether manipulating certain experimental design choices has the effects on SPR-times that a probabilistic constraint-based approach to pragmatic processing makes. In particular, we would hypothesize that expectations of pragmatic felicity are a function of several factors, including:
- the co-presence of other (more felicitous) descriptions in the same experiment
- the co-task, e.g., the question that participants have to answer about the descriptions (truth, adequacy, did a certain word occur, ...)
- the overal rate with which true and/or felicitous sentences occur in the experiment
- regression analysis for data from second pilot
- check spill-over reagions
- check temporal developement (trial effects)
- check subjective patterns
- consider using a "stops making sense task" (Schwarz & Tiemann 2017, "Presupposition Projection in Online Processing", JoS, Section 3.3)
- manipulate proportion of felicitous & infelicitous examples