NP Order • Interaction of age {3 yo, 5 yo} and NP order (Est. = 2.64, SE = .70, z = 3.76, p < 0.001) • Higher accuracy in agent-initial than patient- initial for 5 yo but not for 3 yo • Interaction of age {5 yo, 7 yo} and NP order (Est. = 3.57, SE = 1.07, z = 3.34, p < .001) • Larger effect of NP order in 7 yo than in 5 yo Subject-agent congruence • No main effect of subject-agent congruence nor by-age group interaction Participants: 20 Tagalog speakers per age group 2x2x4 design: NP order x Subject-agent congruence x Age group IVs: NP order (AI: agent-initial, PI: patient- initial), subject-agent congruence (SC: subject-congruent, SI: subject-incongruent), and age group (3-year-olds, 5-year-olds, 7- year-olds, adult group) DV: response accuracy Materials: • 32 audio-recorded semantically reversible transitive sentences similar to (a–d) and 32 fillers • Picture pair: target and distractor • Latin square design: 4 lists (8 experimental items and 8 fillers in each) • Pseudo-randomization Task: Sentence-picture matching Analysis: Mixed-effects logistic regression Thematic role assignment in the L1 acquisition of Tagalog Rowena Garcia 1,* , Irina Sekerina 2 , Jeruen E. Dery 3 , Jens Roeser 4 , & Barbara Höhle 1 1 University of Potsdam, 2 CUNY–College of Staten Island, 3 ZAS Berlin, 4 Nottingham Trent University *Contact: rgarcia@uni-potsdam.de (a) agent-initial, subject-congruent (AI-SC) H<um>ihila ang baka ng baboy. <AV> pull SUBJ cow GEN pig ǮThe cow is pulling a pig.ǯ (b) patient-initial, subject-congruent (PI-SC) H<um>ihila ng baboy ang baka. <AV> pull GEN pig SUBJ cow ǮThe cow is pulling a pig.ǯ (c) agent-initial, subject-incongruent (AI-SI) H<in>ihila ng baboy ang baka <PV> pull GEN pig SUBJ cow The/A pig is pulling the cow. (d) patient-initial, subject-incongruent (PI-SI) H<in>ihila ang baka ng baboy <PV> pull SUBJ cow GEN pig The/A pig is pulling the cow. Target and distractor pictures for the experimental item—ǮThe cow is pulling a pig.ǯ Mean accuracy and standard error for NP order within age group Significance codes: **p < 0.01 . ***p < 0.001 Do Tagalog-speaking children use the noun phrase order in a sentence or the syntactic structure for thematic role assignment? View 1: Children interpret the first noun phrase as the agent because the agent tends to precede the patient (cf. Primus, 1999; 2006). View 2: Children interpret the subject as the agent due to a linking rule that maps the agent to the subject, and the patient to the object (Pinker, 1989; 1994). Children’s use of a word order strategy • Cross-linguistically attested (Chan et al., 2009; Slobin & Bever, 1982) • In previous studies, the first noun phrase was confounded with the subject role. • Either NP order or syntactic structure could be used to assign thematic roles. Tagalog • Verb-initial with flexible ordering of post-verbal arguments (Kroeger, 1993) • Verb morphology indicates the thematic role of the subject, i.e., ang-phrase. • Infix –um– denotes that the subject is the agent (a, b; subject-congruent). • Infix –in– denotes that the subject is the patient (c, d; subject-incongruent). • Permits manipulation of NP order independent of subjecthood • If children use NP order, they should interpret agent-initial sentences (a, c) with higher accuracy than patient-initial sentences (b, d). • However, if they use syntactic structure, they should have higher accuracy in subject-congruent sentences (a, b) than in subject-incongruent sentences (c, d). • Higher scores in agent-initial sentences than in patient-initial sentences from the age of 5, consistent with a first NP as agent strategy (cf. Primus 1999; 2006) • 3 yo: no reliance on NP order • 7 yo: stronger reliance on NP order compared to 5 yo • Adults: reliance on NP order similar to Ferreiraǯs ȋ00Ȍ findings • No tendency to interpret the subject as the agent, in contrast to Pinkerǯs reverse linking hypothesis (1989; 1994) • NP order is used in the acquisition of thematic role assignment. References: Chan, A., Lieven, E., & Tomasello, M. ȋ009Ȍ. Childrenǯs understanding of the agent-patient relations in the transitive construction: Cross-linguistic comparisons between Cantonese, German, and English. Cognitive Linguistics, 20(2), 267–300. Ferreira, F. (2003). The misinterpretation of noncanonical sentences. Cognitive Psychology, 47, 164–203. Kroeger, P. (1993). Phrase Structure and Grammatical Relations in Tagalog. Stanford: CSLI Publications. Pinker, S. (1989). Learnability and Cognition: The Acquisition of Argument Structure. Cambridge, MA: MIT Press. Pinker, S. (1994). How could a child use syntax to learn verb semantics? Lingua, 92, 377–410. Primus, B. (1999). Cases and Thematic Roles: Ergative, Accusative and Active. Tübingen: Niemeyer. Primus, B. (2006). Hierarchy mismatches and the dimensions of role semantics. In I. Bornkessel, M. Schlesewsky, B. Comrie, & A. Friederici (Eds.), Semantic Role Universals and Argument Linking: Theoretical, Typological and Psycholinguistic Perspectives (pp. 53–88). Berlin: Walter de Gruyter. Slobin, D. I. & Bever, T. G. (1982). Children use canonical sentence schemas: A crosslinguistic study of word order and inflections. Cognition, 12, 229–65. Predictions Method Research Question Background Mean accuracy and standard error for subject-agent congruence within age group Architectures and Mechanisms for Language Processing, 3–5 September 2015, Valletta, Malta Results Discussion