A comparison of online and offline measures of good-enough processing in garden-path sentences

ABSTRACT

In two self-paced reading and one ERP experiments, this study tested the good-enough processing account, which states that readers sometimes misinterpret sentences like While the man hunted the deer ran into the woods because they fail to fully revise the syntactic structure [Christianson, K., Hollingworth, A., Halliwell, J. F., & Ferreira, F. (2001). Thematic roles assigned along the garden path linger. Cognitive Psychology, 42, 368–407. doi:10.1006/cogp.2001.0752]. Such an account predicts more evidence of reanalysis at the disambiguation on correctly- than incorrectly-answered trials. Experiment 1, which asked Did the man hunt the deer? and Experiment 2, which asked Did the sentence explicitly say that the man hunted the deer? showed no difference in reading time between trials with correct and incorrect responses. Experiment 3 found the amplitude of P600 was unrelated to comprehension accuracy. These results converged to suggest that failure to reanalyse ambiguous sentences is not the primary reason for misinterpretation. Three norming studies revealed instead response accuracy was influenced by likelihood of events described in the sentences and questions.

KEYWORDS:

• Garden-path sentences
• good-enough processing
• reanalysis
• P600
• plausibility

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. A similar problem with response-contingent reading times would have arisen in Experiments 1 and 2 if ANOVAs had been used to analyze them, but the mixed-effect models that were used to analyze reading times can handle situations with missing data.

Additional information

Funding

This work was funded by the Dissertation Completion Fellowship and the Network for Neuro-Cultures Graduate Training Fellowship from the Graduate College, University of Illinois at Urbana-Champaign to Zhiying Qian.