Resumen de Prediction error and implicit learning in L1 and L2 syntactic priming

Carrie N. Jackson, Holger Hopp

• Aims and Objectives:

  This study examines the syntactic priming of adverb-first word order among 27 first language (L1) English, 33 L1 German, and 32 L1 German-second language (L2) English speakers to investigate the relationship between short-term priming and longer-term learning across different speaker groups.

  Design/Methodology:

  Participants completed a syntactic priming task in either English or German, in which they heard sentences containing fronted (i.e. adverb-first) or non-fronted adverbial phrases and described pictures containing similar adverbial phrases. Immediately before and after the priming task, participants completed picture description tasks containing similar sentence types to measure their baseline production of fronted adverbials, and whether subsequent production of fronted adverbials was modulated by the priming task.

  Data Analysis:

  We used mixed-effects logistic regression to compare participants’ production of fronted sentences during the priming task according to prime type (fronted vs. non-fronted) and their production of fronted sentences before versus after the priming task, including the additional factors of adverbial phrase (temporal phrase vs. locative phrase) and speaker group (L1 English vs. L1 German vs. L2 English).

  Findings/Conclusion:

  Participants exhibited greater short-term priming during the priming task for temporal versus locative phrases, with the L2 English speakers exhibiting the greatest short-term priming. There was a significant increase in the production of fronted sentences from baseline to post-test, with the L1 English speakers exhibiting the greatest gains.

  Originality:

  This study uses systematic between- and within-language comparisons to examine: (a) whether the extent of syntactic priming depends on the frequency of a construction in a speaker’s L1 or L2 (i.e. prediction error); and (b) whether prediction error and short-term priming lead to longer-term learning.

  Significance:

  Our findings show that the magnitude of prediction error influences both short-term adaptation and longer-term priming. However, the ability to harness prediction error for longer-term learning versus short-term adaptation may vary between L1 and L2 speakers.