Poster A31, Thursday, August 16, 10:15 am – 12:00 pm, Room 2000AB
Experimental L2 Semantics/Pragmatics of Scalar Implicature: An ERP Study
Euiyon Cho1, Wonil Chung1, Myung-Kwan Park1;1Dongguk University
There have been a few studies investigating the processing of scalar implicature using ERPs. Noveck and Posada (2003) measured ERPs while participants read and judged underinformative sentences such as “Some cats have ears”, which is correct under a semantic interpretation but incorrect under a pragmatic interpretation. At the critical word such as ‘ears’, a reduced N400 was elicited relative to the control condition. Niewland et al. (2010) reported that participants with high pragmatic ability showed a greater N400 for the underinformative relative to the informative critical word. The present study examined Korean English L2ers’ comprehension of scalar implicature of English ‘some’ and ‘no’ to investigate whether they make a distinction between semantic and pragmatic interpretations. Nineteen Korean L2 learners with a high level of English proficiency participated in our experiment. In our experiment we adopted a picture-sentence verification design (modelled after Politzer-Ahles et al. (2012)) to compare the neural responses to scalar quantifiers such as some and no. On each trial the participants were presented with a picture, followed by a sentence that correctly or incorrectly describes it. In the experiment 1 we manipulated picture type, particularly, Some-type picture and constructed the four types of picture, such as (i) some apples in the basket, (ii) no apple in the basket, (iii) all apples in the basket, and (iv) some pineapples in the basket. The quantifier some was used in the experimental condition such as In the picture, being thrown into the basket by the boy was some candies. Likewise, in the experiment 2 we manipulated No-type picture and constructed the four types of picture, such as (i) no apple in the basket, (ii) some apples in the basket, (iii) all apples in the basket, or (iv) no banana in the basket. The quantifier no was used in the sentence such as In the picture, being thrown into the basket by the boy was no candy. ERPs were measured at the critical phrase (e.g., some candies or no candy). EXP 1 showed (i) a significant anterior P600 in No (semantic violation: SV) condition, (ii) a significant P200 in All (pragmatic violation: PV) condition, and (iii) a significant N400 in lexical violation (like ‘some pineapples’ in the picture) condition, relative to Some condition, respectively. EXP 2 showed (i) a significant N400 in Some (SV) condition, (ii) a marginal P700 in All (PV) condition, and (iii) a significant N400 in lexical violation (like ‘no banana’ in the picture) condition, relative to correct No condition, respectively. Our results make the following points. The lexical violation conditions with both ‘some’ and ‘no’ elicited the same response: a significant N400. However, the two quantifiers in question diverge: P600 for ‘some’ and N400 for ‘no’ in semantic violation, and P200 for ‘some’ and a marginal P700 in pragmatic violation. These findings suggest that semantic/pragmatic aspects of meaning are processed using different mechanisms than lexical aspects of meaning, and that different quantifiers may make a distinct contribution in semantics/pragmatics.
Topic Area: Grammar: Syntax