You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

Poster C7, Thursday, November 9, 10:00 – 11:15 am, Harborview and Loch Raven Ballrooms

Conceptual number agreement processing and coreference establishing in Brazilian Portuguese: An ERP study.

Juliana Andrade Feiden1,2, Srđan Popov2, Roelien Bastiaanse2;1International Doctorate for Experimental Approaches to Language and Brain (IDEALAB), Universities of Groningen (NL), Newcastle (UK), Potsdam (DE), Trento (IT), Macquarie University (AU), 2Center for Language and Cognition Groningen (CLCG), University of Groningen, Groningen, The Netherlands

Number agreement depends mainly on two kinds of information: morphosyntactic/grammatical and semantic/conceptual (Bock, Eberhard, & Cutting, 2001; Eberhard, Cutting, & Bock, 2005). For most nouns, morphosyntactic and conceptual numbers coincide (Schweppe, 2013; e.g., the girlSG – one girl and the girlsPL – more than one girl). Conversely, for collective nouns, syntactic and conceptual number do not match (e.g., the gangSG – a group of criminals). When nouns with conceptual number are involved in coreference establishing, the pronoun agrees with the noun’s conceptual number, thus creating an agreement mismatch in grammatical number (e.g., The bandSG played last night. TheyPL were talented). Our study investigates the role of conceptual number in coreference establishing by using Even Related Potentials, since they are differentially sensitive to syntactic and semantic information. We also aim to verify the processes and ERP components that are elicited by noun-pronoun number violations while manipulating the number type (grammatical vs. conceptual). Our hypothesis is that conceptual number agreement relies more on conceptual information when compared to grammatical number processing (only grammatical information). This reliance on conceptual information is expected to be reflected in an N400 effect, since coference establishing is happening at the semantic/conceptual level. Therefore, we predict that sentences containing a conceptual number violation (collective noun – singular + personal pronoun – singular) will elicit a N400 effect, while sentences that present a grammatical number violation will elicit a P600 effect (collective noun – singular + personal pronoun – plural). We tested 27 native speakers of Brazilian Portuguese on a passive reading task while recording EEG. The experimental materials comprised 240 experimental sentence pairs split into 2 lists, with an additional 80 fillers per list. In the experimental sentence pairs, the antecedents in half of the introductory sentences were collective nouns (conceptual number condition), and in the other half the antecedents were plural nouns with grammatical number (grammatical number condition). The number violation was created between a singular pronoun of the second sentence, and a collective or a plural noun in the first sentence. Forty collective nouns were used, of which twenty masculine and twenty feminine. In the collective number condition, the N400 was elicited for conceptual number violation, but on the verb following the pronoun. In the grammatical number condition, the Nref effect was observed on the pronoun. Our results show that conceptual number and grammatical number are processed differently. The Nref effect for the grammatical number violation indicates that the parser is in search of a proper antecedent. Furthermore, since there is a grammatical number mismatch, this process seems to take place at the syntactic level. In the case of conceptual number agreement, the violation elicits the N400 on the element following the pronoun – the verb. Such result indcates that conceptual number is accessed and integrated at the conceptual/semantic level. Curisously, the parser seems to need the additional information from the singular verb following the pronoun in order to establish the intended coreference with the collective noun.

Topic Area: Grammar: Syntax

Back to Poster Schedule