Syntactic Priming is Sensitive to Both Constituent Structure and Argument Structure

Syntactic priming is a phenomenon in which the processing of a sentence can speed up the comprehension or production of another sentence with similar or identical syntax. Syntactic priming draws on abstract syntactic representation, which integrates the information of the predicate’s argument structure in order to support the Argument Structure Priming Hypothesis (Konradt, 2020). This contradicts the widely used theory of syntactic priming, Constituent Structure Priming, which posits that priming only concerns linear constituent order (Bock & Loebell, 1990). Findings of Bock and Loebell (1990) show that English speakers are equally more likely to describe a picture with a passive sentence after being primed with a passive (e.g. The construction worker was hit by the bulldozer) or a locative sentence (e.g. The construction worker was digging by the bulldozer) than an active sentence (e.g. The construction worker drove the bulldozer). However, it should be noted that the target pictures used by Bock and Loebell presented transitive events, and the intended patient argument was always human while the agent argument was inanimate. According to the Argument Prominence Hierarchy theory (APH) (Titov, 2012), the relative interpretative position of the arguments, which is determined by a number of characteristics including animacy, determines the linear order of arguments in a sentence. Specifically, a human argument is more interpretively prominent than a non-human/inanimate argument, and an animate argument is more prominent than an inanimate argument. APH predicates that the more interpretively prominent argument precedes the less prominent argument. Accordingly, the research design in Bock and Loebell (1990) may lead to unequal prominence in the arguments of the target sentence, which potentially biases the use of passives and offers a natural license for the passive independent of the priming sentence. Thus, Bock and Loebell’s findings could have been influenced by an unequal animacy distribution in the targets rather than by surface constituent structure repetition as they claim. The present study is therefore a replication study of Bock and Loebell (1990) aiming to test whether syntactic priming is sensitive to argument structure or whether it is only triggered by surface constituent structure. This study (N = 100 English native speakers) refines the experimental design using adjusted stimuli - balancing the position of agents and patients in equal animacy targets and unequal animacy targets. Findings show that more passive targets are produced in passive primes than in locative primes. At the same time, there seemed to be an increase in full passives in the locative conditions compared to the active primes in EA and UA together due to the same constituent structure between locatives and passives. Thus, this study suggests that syntactic priming is sensitive to both constituent structure and argument structure, supporting the Argument Structure Priming Hypothesis.