Do readers misassign thematic roles? Evidence from a trailing boundary-change paradigm

ABSTRACT

We report an eye-tracking experiment with a trailing boundary-change paradigm as people read subject- and object-relative clauses that were either plausible or implausible. We sought to determine whether readers sometime misassign thematic roles to arguments in implausible, noncanonical sentences. In some sentences, argument nouns were reversed after participants had read them. Thus, implausible noncanonical sentences like “The bird that the worm ate yesterday was small” changed to plausible “The worm that the bird ate was small.” If initial processing generates veridical representations, all changes should disrupt rereading, irrespective of plausibility or syntactic structure. Misinterpretation effects should only arise in offline comprehension. If misassignment of thematic roles occurs during initial processing, differences should be apparent in first-pass reading times, and rereading should be differentially affected by the direction of the text change. Results provide evidence that readers sometimes misassign roles during initial processing and sometimes fail to revise misassignments during rereading.

KEYWORDS:

• Thematic-role reversal errors
• noncanonical sentences
• good-enough theory
• eye tracking
• language processing

Acknowledgments

This research was funded by the National Science Foundation (BCS-1628347; Christianson). The authors wish to thank all the members of the EdPsych Psycholinguistics Lab at the Beckman Institute for Advanced Science and Technology for assistance with data collection and cleaning, especially our dedicated and skilled crew of undergraduate students. Finally, we wish to thank anonymous reviewers and Michael Meng for invaluable comments and critiques during the peer review process.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 If accurate, the post-interpretive account would also raise serious concerns about reading comprehension assessment and testing in educational settings, where post-hoc, metalinguistic questions are routinely posed to students and test-takers. These concerns might well be justified, so we feel this is an important aspect of the current theoretical discussion that has received little attention or consideration in the psycholinguistic literature.

Additional information

Funding

This work was supported by National Science Foundation (United States): [Grant Number BCS-1628347].