Verb aspect and perceptual simulations

Abstract

Two experiments investigate the influence of verb aspect on situation representations. The results demonstrate that comprehenders use verb aspect as a cue to regulate the activation of ongoing simulations of situations over time. Experiment 1 measured word-by-word reading as well as sensibility judgements on sentences in which a target object word had been replaced by a picture. For the past imperfective sentences, participants were faster to process the picture, the two words following the picture, and the sensibility judgements when objects were pictured in use rather than not in use. However, this in-use facilitation was limited to processing of the picture for the past perfect sentences. Experiment 2 served as a control to ensure that the use effect and its interaction with verb aspect were a result of contextual manipulations rather than surface features of the pictures themselves. The results are interpreted within the framework of perceptual simulations during language comprehension.

Keywords:

• Language comprehension
• Representations
• Perceptual simulations
• Grammar

Acknowledgments

This research was supported by a grant from the Erasmus University Trustfonds to Carol J. Madden. Thanks to Evelyn Chiang, John Murty, Stephanie Lai, and Emily Kuch for help with stimulus materials and data collection.

Notes

¹ The perfect aspect (e.g., past perfect: “The boy had built a doghouse”) is sometimes confused with the perfective aspect (e.g., simple past: “The boy built a doghouse”). Both aspects convey the situation as collapsed and completed, but the perfect aspect focuses attention on the resulting state of the situation and was thus preferred given the representational aims of the present study.

² For counterbalanced designs only subject analyses should be performed (Raaijmakers, 2003; Raaijmakers, Schrijnemakers, & Gremmen, 1999).

³ This same reasoning applies to the integration of other relevant event concepts, such as verbs themselves. When a verb (kick) is encountered, we assume that a lexical-level simulation of the action of kicking is initially activated, regardless of whether the verb occurred in the imperfective (was kicking) or perfect (had kicked).

⁴ We designed our materials such that in-use pictures would be compatible, and not-in-use pictures would be incompatible with an ongoing situation, but we did not intend that the in-use and not-in-use pictures would have any temporal qualities. Though temporal stage of a situation was not intended to factor into our materials, it could be that some not-in-use pictures in the current study were incompatible with the end state of a situation because they clearly depicted a beginning state. Upon inspection of the current set of materials, there were only three experimental items in which the not-in-use picture was incompatible with an end point of a situation (unburned candle for “had burned”, full pitcher for “had poured”, and empty salt shaker for “had refilled”). However, the removal of these items did not strengthen the use effect for the perfective sentences.