Sentence Processing and Reanalysis in Non-Native Speakers

Sentence processing and reanalysis in non-native speakers (2017-2019)
Hiroki Fujita
Language Learning Dissertation Grant

This project, funded by a Language Learning Dissertation Grant, examined the reanalysis processes in non-native (L2) English speakers that may occur when reading temporarily ambiguous sentences. For example, a sentence like “After Mary dressed the baby laughed happily” is temporarily ambiguous, such that readers may initially interpret it to mean “Mary dressed the baby” while in fact the correct interpretation is actually “Mary dressed herself”. As a result, reanalysis is required at “laughed”, which disambiguates towards “the baby” being the subject of the main clause (‘the baby laughed’) rather than the direct object of the subordinate clause (‘After Mary dressed…’). Recently, studies on native (L1) sentence processing have started to examine effects of reanalysis on language comprehension (e.g., Christianson et al., 2001; Slattery et al., 2013). These studies have revealed that L1 speakers often persist with the initially-assigned interpretation (“Mary dressed the baby”), and that persistence of this misinterpretation results from a failure to erase the memory trace of the initially assigned interpretation from memory, rather than an ability to conduct syntactic reanalysis and understand the structure of the sentence.

Regarding L2ers, recent studies have shown that L2 readers may be more likely to misinterpret such sentences than L1 speakers (Jacob & Felser, 2016). Potentially, this results from L2 speakers failing to construct the correct structure of the sentence, which may be inline with the Shallow Structure Hypothesis of L2 processing (Clahsen & Felser, 2006). Alternatively, it is possible that increased reanalysis difficulty in L2 readers is due to interference from the initially assigned interpretation in memory, as in L1 speakers (Cunnings, 2017; Slattery et al., 2013). To tease apart these accounts, this project conducted four studies. Two studies used ‘offline’ comprehension question measures, while two studies used ‘online’ eye-tracking during reading to examine sentence processing. The results showed that consistent with previous studies, L2 speakers persist with misinterpretation more greatly than L1 speakers, but that the source of L2 reanalysis difficulty is a failure to erase the memory trace of the initially assigned interpretation from memory, rather than a failure to conduct syntactic reanalysis.

Related Publications
Fujita, H. and Cunnings, I. (2021). Reanalysis processes in non-native sentence comprehension. Bilingualism: Language and Cognition, 24, 628-641.

Fujita, H. and Cunnings, I. (2021). Lingering misinterpretation in native and nonnative sentence processing: Evidence from structural priming. Applied Psycholinguistics, 42, 475-504.

Fujita, H. and Cunnings, I. (2020). Reanalysis and lingering misinterpretation of linguistic dependencies in native and non-native sentence comprehension. Journal of Memory and Language, 115, 104154.