Reanalysis and lingering misinterpretation of linguistic dependencies in native and non-native sentence comprehension

Abstract

Research on temporarily ambiguous "garden path" sentences (e.g., After Mary dressed the baby laughed) has shown that initially assigned misinterpretations linger after reanalysis of the temporarily ambiguous phrase in both native (L1) and non-native (L2) readers. L2 speakers have particular difficulty with reanalysis, but the source of this L1/L2 difference is debated.Furthermore, how lingering misinterpretation may influence other aspects of language processing has not been systematically examined. We report three offline and two online experiments investigating reanalysis and misinterpretation of filler-gap dependences (e.g., Elisa noticed the truck which the policeman watched the car from). Our results showed that L1 and L2 speakers are prone to lingering misinterpretation during dependency resolution. L1/L2differences were observed such that L2 speakers had increased difficulty reanalysing some filler-gap dependencies, however this was dependent on how the dependency was disambiguated. These results are compatible with the "good enough" approach to language processing, and suggest that L1/L2 differences are more likely when reanalysis is particularly difficult.

Interference and filler-gap dependency formation in native and non-native language comprehension.

“…It is also possible however that previous studies have overestimated L1/L2 differences owing to small sample sizes. Although we acknowledge that the absence of L1/L2 differences in the present study may be attributable to the lack of statistical power, our sample size is larger than the previous L2 studies that have observed L1/L2 differences in filler-gap dependency formation (e.g., Fujita & Cunnings, 2020; Marinis et al, 2005) and interference in reflexive resolution (Felser et al, 2009; Felser & Cunnings, 2012). Given that a small sample size potentially leads to overestimates of effect sizes (Vasishth et al, 2018), future research will need to test the replicability of the present study and previous ones.…”

“…There are several possible accounts of why the present study did not observe L1/L2 differences while some previous studies did. One is that L1/L2 differences in filler-gap dependencies arise only in certain circumstances, such as in sentences with filler-gap dependencies spanning several clauses or with increased processing cost due to revision (Fujita & Cunnings, 2020; Marinis et al, 2005). Another is that L2 speakers are susceptible to discourse saliency as suggested by Felser and Cunnings (2012).…”

“…Regarding the processing of filler-gap sentences, L2 speakers recover the information of the filler upon encountering a gap (Felser et al, 2012; Williams et al, 2001) but sometimes show non-nativelike patterns (Fujita & Cunnings, 2020; Kim et al, 2015; Marinis et al, 2005). Marinis et al (2005) argued that L2 speakers do not process sentences with filler-gap dependencies spanning several clauses in the same manner as L1 speakers (Gibson & Warren, 2004; but see Pliatsikas & Marinis, 2013, who reported that L2 speakers with extended naturalistic exposure to English, their L2, show a nativelike pattern).…”

“…Marinis et al (2005) argued that L2 speakers do not process sentences with filler-gap dependencies spanning several clauses in the same manner as L1 speakers (Gibson & Warren, 2004; but see Pliatsikas & Marinis, 2013, who reported that L2 speakers with extended naturalistic exposure to English, their L2, show a nativelike pattern). Fujita and Cunnings (2020) showed that, unlike L1 speakers, L2 speakers may fail to revise temporarily ambiguous filler-gap sentences when revision cost is high. Kim et al (2015) found that L2 speakers whose L1 is a wh-in-situ language, where a wh-element does not overtly move to a higher position (Ross, 1967), violate syntactic constraints on when a dependency may be formed during sentence processing.…”

Share the code, not just the data: A case study of the reproducibility of articles published in the Journal of Memory and Language under the open data policy

“…This is because the failure may be driven by many factors that are external to the research question under investigation: differences in the population and/or language studied, the natural variability in the dependent variable, lab settings, equipment, and protocols can come together to lead to very different outcomes. Indeed, it is possible that when it comes to studying subtle and highly variable aspects of human (IR)REPRODUCIBILITY IN PSYCHOLOGY AND PSYCHOLINGUISTICS (IR)REPRODUCIBILITY IN PSYCHOLOGY AND PSYCHOLINGUISTICS (2020), Brandt et al (2020), Brewer et al (2021), Bristol and Rossano (2020), Brothers and Kuperberg (2021), Brysbaert (2019), Bürki et al (2020), Chan et al (2020), Chetail (2020), Collins et al (2020), Corps and Rabagliati (2020), Dıéez-Álamo et al (2020), Falandays et al (2020), Fellman et al (2020), Floccia et al (2020), Fox et al (2020), Fujita and Cunnings (2020), Gagné et al (2020), Garnham et al (2020), Günther, Nguyen, et al (2020), Günther, Petilli, et al (2020), Hesse and Benz (2020), Hollis (2020), Humphreys et al (2020), Hwang and Shin (2019), Isarida et al (2020), Jäger et al (2020), Johns et al (2020), Kaula and Henson (2020), Lauro et al (2020), Lelonkiewicz et al (2020), Li et al (2020), Liang et al (2021), McKinley and Benjamin (2020),…”