Spanish Ministry of Education and Science (BRAINGLOT, CSD2007­00012/CONSOLIDER­INGENIO 2010), Ministry ofScience and Innovation (FFI2010­20472), Ministry of Economy and Competitiveness (FFI2012­31360), Juan de la CiervaFellowship to Zawiszewski (JCI­2010­07692) and a Ramón y Cajal Fellowship to Erdocia (RYC­2010­06520).

When person meets number:An ERP study on feature processing

Adam Zawiszewski, Kepa Erdocia, Mikel Santesteban & Itziar LakaUniversity of the Basque Country, Vitoria-Gasteiz, Spain

adam.zawiszewski@ehu.es

Materials and procedureVerb agreement has received significantattention both in linguistics [1] [2] and inpsycholinguistics [3]: (1) Results showthat agreement violations usually elicit alate parietal positivity (P600),sometimes preceded by negativecomponents (LAN or N400). (2) Butlinguistic analysis claims that verbagreement is composed of distinct phi­features, such as person and number;(3) there is scarce experimentalevidence that phi­features like personand number are processed distinctly [4][5]. The aim of this study is to exploreagreement feature­processing in detail;in order to do that we investigatedprocessing of agreement features inBasque, both in subject and in objectagreement configurations.

References[1] Baker, M. (2008). The syntax of agreement and concord. Cambridge: Cambridge University Press.[2] Rezac, M. (2011). Phi­features and the modular architecture of language. Berlin: Springer.[3] Molinaro, N., Barber, H., Carreiras. (2011). Grammatical processing in reading: ERP findings and future

directions. Cortex 47, 908­930.[4] Nevins, A., Dillon, B., Malhotra, S., Phillips, C. (2007). The role of feature­number and feature­type in

processing Hindi verb agreement violations. Brain Research, 1164, 81­9.[5] Mancini, M., Molinaro, N., Rizzi, L. and Carreiras, M. (2011). A person is not a number: Discourse involvement

in subject–verb agreement computation. Brain Research, 1410, 64 ­ 764.

Acknowledgments

Introduction

Results

Discussion & Conclusions• Participants faster when detecting the

ungrammaticality of violations involvingperson than the control sentences

• More errors judging the ungrammaticality ofnumber violations than violations involvingperson

• Violations involving the person feature (withor without number) elicited larger P600 thanviolations that did not (number violations).

SUBJECT­VERB AGREEMENTBASELINE 1. Zuk mutila bakarrik utzi duzu kalean.you­S boy­O alone left have­2SG in­the­street'You left the boy alone in the street"PERSON 2. Zuk mutila bakarrik utzi dut kalean.

you­S boy­O alone left have­1SG in­the­streetNUMBER 3. Zuk mutila bakarrik utzi duzue kalean.you­S boy­O alone left have­2PL in­the­streetPERSON & 4. Zuk mutila bakarrik utzi dugu kalean.NUMBER you­S boy­O alone left have­1PL in­the­street

• 22 native speakers of Basque• word­by­word display for 350 ms (ISI = 250 ms).• grammaticality judgment task• 400 sentences randomly distributed over 4 blocks

In sum:• Phi­features such as person

and number are processeddistinctly during morphologicalcomputation. Person is moresalient than number (as [4])

• Different repair processesoccur on the encounter ofdifferent feature violations