2018 Friday Poster 6519

Friday, November 2, 2018 | Poster Session I, Metcalf Small | 3pm

Default animacy configuration is not always preferred: the effect of animacy on the acquisition of passives
E. Lau

Studies (Mak et al., 2002; Kidd et al., 2007) have demonstrated a critical role of animacy in child and adult processing of long distance dependencies, such as relative clauses. The difficulty associated with the supposedly less preferred constructions can disappear when the predicate matches the preferred animacy expectations. E.g., Kidd et al. (2007) found that object relative clauses with an animate agent (subject) and an inanimate patient (object) improves children’s understanding. In light of these studies on the effect of animacy, this paper examines whether the observed facilitative effect of animacy would also be found in the acquisition of passives.

Similar to the errors children committed in processing long distance dependencies, children often misinterpret passives as actives – they mistake the subject of the passive as the agent of the transitive event, and the oblique NP as the patient. This appears to stem from their semantic and thematic expectations for different grammatical positions. On this view, children prefer (i) the subject to be animate, and (ii) the animate entity to be the agent. Therefore, for sentences like (1), children mistake the boy as the agent of the transitive event block as it occupies the subject position and is an animate entity.

In the study, we crossed animacy with grammatical role, as shown in Table 1. Here, we focus on the contrastive animacy patterns (3-4), which presents the preferred/default animacy configuration and the reverse. If using a preferred animacy configuration in a construction will ameliorate processing difficulty, AI passives should be found easier to process by children. In contrast, as the animacy values of the IA configuration runs contradictory to children’s default expectations, IA passives should be found difficult.

63 Mandarin-speaking children were tested with a picture selection task (Table 2 & Fig.1). A mixed effects logistic regression model was used to analyze rate of comprehension accuracy, with Animacy (AI vs. IA) and Passive Type (Long vs. Short) and their interaction as fixed effects, Age as a co-variate, and random intercepts for subjects (Table 3). Likelihood Ratio Test confirmed that the model with Animacy (X2(1)=19.24, p<.001) and Age (X2(1)=17.24, p<.001) were significant predictors of children’s accuracy, whereas the effects of Passive (X2(1)=2.14, p=.34) and the interaction between the two factors (X2(1)= 2.01, p=.15) were insignificant.

Results (Fig.2) show that children were more likely (1.64 times) to misinterpret AI passives than their IA counterparts in both long and short passives. This suggests that children’s processing may not be primarily guided by the semantic and thematic expectations, but rather by the preference for subject to be animate. This might be related to an idea originally proposed for the predominant subject preference in relative clause processing: the subject prominence effect (O’Grady, 2011). It is hypothesized that, with the animate entity being the subject, the subject becomes more prominent and allows more accessible reference in the discourse. Processing thus becomes easier as the animacy of the subject makes it less effort to be referred to.

References

Mak, W. M., Vonk, W., & Schriefers, H. (2002). The influence of animacy on relative clause processing. Journal of Memory and Language, 47(1), 50-68.

Kidd, E., Brandt, S., Lieven, E., & Tomasello, M. (2007). Object relatives made easy: a crosslinguistic comparison of the constraints influencing young children’s processing of relative clauses. Language and Cognitive Processes, 22(6), 860-897.

O’Grady, W. (2011). Relative clauses: Processing and acquisition. In E. Kidd (Ed.), The acquisition of relative clauses: Processing, typology and function (pp. 13-38). Amsterdam: Benjamins.