During my poster session (Friday, 12:10-2:00) you can contact me through this comment function or by e-mail. My e-mail address is daisyji91@gmail.com. I would greatly appreciate it if you kindly give me some feedback on this research! Previous research shows language specific developmental trajectories for different relative clause (RC) types, e.g., subject RCs (SRC) before object RCs (ORC), or vice versa. Respective findings were interpreted from various theoretical perspectives (Chen & Shirai, 2014, for an overview). However, for Mandarin Chinese mixed results were obtained, with some studies showing SRC and others ORC primacy. Interestingly, there is a similar controversy in the literature on adult Mandarin RC processing, with some studies reporting an SRC, others an ORC processing advantage. As processing has been identified as an important determinant to shape the course of RC acquisition (O’Grady 2011), the current study investigates the impact of one of the strongest factors in Mandarin adult RC comprehension on RC acquisition: animacy configuration. Wu et al. (2012) reported that the animacy values of RC head nouns and embedded nouns jointly modulate processing difficulty of SRCs and ORCs in adults. Here, we test whether Mandarin children (2.5 to 6.6 years of age) are sensitive to inherent RC structure properties, animacy configuration patterns, and the interaction of both factors in a cross-sectional language production experiment. Participants (Ps) were 52 children from Jiangsu Province in Mainland China, who formed 4 age groups (age 3, N=14; age 4; N=13; age 5; N=10, age 6; N=15). P’s task was to verbally distinguish between two referents as a response to a question asked by a native speaker experimenter (E). Before each question, E introduced the referents (line drawings) as head nouns (HN) of two SRCs (e.g., This is the bear, that ate the apple/pear), or two ORCs (e.g., This is the apple, that the bear/wolf ate). E then asked Which bear/apple is this? pointing to the bear/apple (Fig. 2). There were 32 critical items, 50% targeting SRC, 50% targeting ORC. Across conditions, half of the materials used the preferred animacy configuration (SRC: embedded N=inanimate, HN=animate; ORC: embedded N=animate, HN=inanimate), the other half used the dispreferred pattern (both N animate in both RC types). 32 fillers were added; items were randomized. Ps either responded with an SRC or ORC. Responses were coded for accuracy and error type. A GLMM was used to evaluate response accuracy (yes/no) as a function of group (4 levels, dev-coded), RC type (2 levels, dev-coded), and animacy configuration (2 levels, dev-coded), and their interaction. Both 3- and 4-year-olds produced significantly fewer correct responses than the Grand Mean; both 5- and 6-year-olds produced significantly more correct responses than the Grand Mean. Overall, SRCs were produced correctly more often than ORCs; overall, preferred animacy configurations led to more correct responses. There was an interaction between RC type and animacy configuration: with preferred animacy configurations, SRCs were produced correctly more often than ORCs; no differences between RC types were found with the dispreferred animacy configuration (Fig. 3). The most frequent error type was omission of the RC-internal verb, resulting in what is structurally a genitive construction in Mandarin – like an RC, a head-final structure. This error type was significantly more frequent in younger children than in older children. Response accuracy results highlight that animacy configurations matter in Mandarin RC acquisition, implicating that RC structure building and the mapping of referents to RC structures are two distinct aspects in RC acquisition. If animacy is controlled for, SRC primacy can consistently be observed. Together with the results from the error type analysis, this finding supports the idea that children acquire new structures based on those they have already mastered. However, the template Mandarin children seem to use is not the simple main clause, but the genitive structure: for Mandarin children, genitive structures may exemplify modification of a head “to the right” (head-final structure). SRCs are structurally more similar to genitives than ORCs (see examples). References Chen, J. & Shirai, Y. (2014): The acquisition of relative clause in spontaneous child speech in Mandarin Chinese. In: Journal of Child Language 41(1), 1 – 29. O’Grady, W. (2011): Relative clauses: Processing and acquisition. In: Kidd, Evan (ed.): The acquisition of relative clauses: Processing, typology and function. Amsterdam: John Benjamins Press, 13 – 38. Wu, F., Kaiser, E. & Andersen, E. (2012): Animacy effects in Chinese relative clause processing. In: Language and Cognitive Processes 27(10), 318 – 329.