SPAWNing Structural Priming Predictions from a Cognitively Motivated Parser

Abstract

Structural priming is a widely used psycholinguistic paradigm to study human sentence representations. In this work we introduce SPAWN, a cognitively motivated parser that can generate quantitative priming predictions from contemporary theories in syntax which assume a lexicalized grammar. By generating and testing priming predictions from competing theoretical accounts, we can infer which assumptions from syntactic theory are useful for characterizing the representations humans build when processing sentences. As a case study, we use SPAWN to generate priming predictions from two theories (Whiz-Deletion and Participial-Phase) which make different assumptions about the structure of English relative clauses. By modulating the reanalysis mechanism that the parser uses and strength of the parser's prior knowledge, we generated nine sets of predictions from each of the two theories. Then, we tested these predictions using a novel web-based comprehension-to-production priming paradigm. We found that while the some of the predictions from the Participial-Phase theory aligned with human behavior, none of the predictions from the the Whiz-Deletion theory did, thus suggesting that the Participial-Phase theory might better characterize human relative clause representations.

Figures (6)

• Figure 1: How is SPAWN different from other models?
• Figure 2: Steps involved in processing each word. Process is repeated till all words are assigned a category.
• Figure 3: Predicted probability of RRC parse from the posterior distribution of the Bayesian logistic regression model. Error bars reflect 95% credible intervals.
• Figure 4: Empirical probability of RRC parse from the posterior distribution of the Bayesian logistic regression model. Error bars reflect 95% credible intervals.
• Figure 5: Syntax tree for "The cat being examined by the doctor...". The words in red are unvoiced in the Whiz-Deletion account. The tree for "The cat examined by the doctor ..." is nearly identical but without the ProgP. In Participial-Phase VoiceP is the sister of cat; in Whiz-Deletion vP is the sister of was.