Reanalysis and semantic persistence in native and non-native garden-path recovery

Abstract

We report the results from an eye-movement monitoring study investigating how native and non-native speakers of English process temporarily ambiguous sentences such as While the gentleman was eating the burgers were still being reheated in the microwave, in which an initially plausible direct-object analysis is first ruled out by a syntactic disambiguation (were) and also later on by semantic information (being reheated). Both participant groups showed garden-path effects at the syntactic disambiguation, with native speakers showing significantly stronger effects of ambiguity than non-native speakers in later eye-movement measures but equally strong effects in first-pass reading times. Ambiguity effects at the semantic disambiguation and in participants’ end-of-trial responses revealed that for both participant groups, the incorrect direct-object analysis was frequently maintained beyond the syntactic disambiguation. The non-native group showed weaker reanalysis effects at the syntactic disambiguation and was more likely to misinterpret the experimental sentences than the native group. Our results suggest that native language (L1) and non-native language (L2) parsing are similar with regard to sensitivity to syntactic and semantic error signals, but different with regard to processes of reanalysis.

Keywords:

• Non-native language processing
• Garden-path recovery
• Reanalysis
• Semantic persistence

We thank Oliver Boxell, Sabrina Gerth, and Clare Patterson for help with the stimulus creation, and Kiel Christianson and two anonymous reviewers for their constructive comments and helpful discussion.This research was supported by an Alexander-von-Humboldt professorship to Harald Clahsen, which is gratefully acknowledged.