Di Nardo, Julia C. (2010) Building Event Meanings from Linguistic and Visual Representations: Evidence from Eye Movements. PhD thesis, Concordia University.
- Accepted Version
This paper reports three studies that were conducted to explore how the meaning of events is constructed through the use of spoken language comprehension and dynamic scene information. These studies served as an extension of prior work conducted by the present author with the overarching aim of modifying aspects of the methodology to modulate the salience of the linguistic context. Participants eye movements were recorded as they watched short movie clips of everyday events and listened to sentences related to those events. Specifically, we were interested in measuring how quickly the target object in each scene (the grammatical referent of the main verb in each sentence) would be fixated after the verb was uttered, and in some version of the movies, after the agent initiated movement toward the target object (synchronized with the verb onset, known as the disambiguating point). The first experiment replicated a previous study (Experiment 2 in Di Nardo, 2005) with the aim of increasing the visual angle of the scenes, thus reducing attentional shifts without corresponding eye movements. Experiment 2 investigated whether the absence of the linguistic context would alter the pattern of eye movements across the scene. Experiment 3 tested the hypothesis that introducing a more semantically salient initial clause to the sentences would lead to faster eye movements to the target object. Results showed that increasing the visual angle did not serve to reduce saccade onset times (SOTs), although the presence of spoken language did shorten SOTs even in the context of agent motion toward the target object. Finally, introducing semantically restrictive initial clauses produced a mixed pattern of results; SOTs were reduced in the absence of agent motion when the verb was not semantically restrictive, although SOTs were longer when the agent moved toward the target object and the verb was not semantically restrictive. Results are discussed within the framework of the Coordinated Interplay Account (Crocker, Knoeferle, & Mayberry, 2009).
|Divisions:||Concordia University > Faculty of Arts and Science > Psychology|
|Item Type:||Thesis (PhD)|
|Authors:||Di Nardo, Julia C.|
|Degree Name:||Ph. D.|
|Date:||3 December 2010|
|Thesis Supervisor(s):||de Almeida, Roberto G.|
|Deposited By:||JULIA DI NARDO|
|Deposited On:||13 Jun 2011 15:03|
|Last Modified:||13 Jun 2011 15:03|
Repository Staff Only: item control page