Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Neural correlates of semantic and pragmatic NPI Processing in English
Poster Session D, Saturday, September 13, 5:00 - 6:30 pm, Field House
This poster is part of the Sandbox Series.
Grace deMeurisse1,2, Edith Kaan1; 1University of Florida, 2Bard College
Negative polarity items (NPIs) are items such as “ever” or “any” in English which may only be grammatical in negative or non-positive contexts. For example, contrast sentences (1a) and (1b) with sentence (1c) below: (1a) No beginner pilots have ever flown that route; (1b) Only beginner pilots have ever flown that route; (1c) *Most beginner pilots have ever flown that route. In sentence (1c) above, the NPI “ever” is ungrammatical because the quantifier “most” is positive. In contrast, in sentence (1a), the NPI “ever” is grammatical given the presence of the negative operator “no”. In sentence (1b), the truth value appears “true” on the surface (beginner pilots have flown that route), but negation is introduced via negative implicature, namely “beginner pilots and no others have ever…” Therein, negation in (1b) can be considered “pragmatic” or provided by more implicit or pragmatic computation (Giannakidou, 2006). Few studies have investigated the processing of NPIs in English using electrophysiological methods such as ERPs, and only one has compared the effect of different licensors (“no”, “only”, “few”, etc.) on NPI processing. Xiang et al. (2016) compared constructions with an “explicit licensor”, i.e., “no/few/only” (collapsed over individual licensors) to an implicit licensor, i.e., “She was surprised that third graders had ever seen one before” (Xiang et al., 2016, p. 76) where negation is introduced via negative implicature, i.e., “she thought no third graders had ever…”. Xiang et al. observed a reduced N400 effect at “ever” for both the explicit and implicit conditions compared to ungrammatical sentences; NPIs under the implicit negation condition elicited a P600 effect compared to the “explicit” negation conditions. The present study seeks to investigate the processing of NPIs in English in native English speakers (n=21; data collection ongoing) by directly comparing two different licensing conditions: a semantic condition (e.g., 1a above) vs. a pragmatic condition, where negation is introduced via negative implicature (e.g., 1b). In this sense, this study follows Xiang et al. (2016), investigating the neural differences between traditional NPI licensing under overt negation (1a) vs. “rescuing” i.e., “pragmatic licensing” by means of negative implicature (1b) (Giannakidou, 2006). Unlike Xiang et al. (2016), we directly compare the effect of two individual licensors, where (1b) should still be considered pragmatic or “implicit” in nature. We include an ungrammatical baseline condition (1c). Our predictions were twofold. First, we predicted that participants would exhibit larger N400 or P600 amplitudes in response to NPIs under the “only” condition than the “no” condition, reflecting more difficulties with licensing via negative implicature. Second, we predicted larger N400 or P600 amplitudes on ungrammatical conditions (e.g., 1c) than the two grammatical conditions collapsed over each other (e.g., 1a, 1b). Data collection is ongoing.
Topic Areas: Syntax and Combinatorial Semantics, Meaning: Lexical Semantics