51³Ô¹Ï

Topic

L1 & L2 extended metaphor comprehension: behavioural insights into the effects of language nativeness and metaphor novelty

School of Languages, Linguistics and Cultures

Date & location

• Thursday, July 10, 2025
• 10:00 A.M.
• Virtual Defence

Examining Committee

Supervisory Committee

• Dr. John Archibald, School of Languages, Linguistics and Cultures, 51³Ô¹Ï (Co-Supervisor)
• Dr. James Tanaka, Department of Psychology, UVic (Co-Supervisor)

External Examiner

• Dr. Iain Higgins, Department of English, UVic

Chair of Oral Examination

• Dr. Karen Dunmall, School of Environmental Studies, UVic

Abstract

This thesis investigates how native (L1) and non-native (L2) English speakers comprehend metaphorical language, with a specific focus on nominal extended metaphors. While prior research has extensively examined single nominal metaphors, often of the “X is a Y” type, comparatively little is known about the processing of extended metaphoric expressions, particularly among bilinguals. This study addresses that gap by exploring if and to what extent metaphor conventionality and figurative context modulate comprehension in L1 and L2 speakers.

The primary experiment employed a semantic judgment task (SJT), in which participants were presented with short context paragraphs followed by target phrases. They were asked to judge whether the phrases were meaningful or meaningless, while both reaction times and accuracy rates were recorded. The experiment was conducted using PsychoPy v2024.2.4 and included both single and extended metaphor conditions, with target utterances falling into four categories: novel metaphors, conventional metaphors, literal sentences, and anomalous sentences. The results revealed several important findings. First, there was no evidence that extended metaphor contexts induced a shift in the processing mode for conventional metaphors. That is, the presence of context did not lead participants to engage in alignment-based (comparison-driven) processing. However, metaphorical context did have a clear facilitatory effect on the processing of both novel and conventional metaphors, though differences emerged between groups. Native speakers processed both types more quickly and accurately when metaphor-related contextual support was provided, while non-native speakers showed a similar trend, but the effect for conventional metaphors was not statistically significant. Notably, the study found that both L1 and highly proficient L2 speakers benefited from metaphor conventionality in the single condition, contrary to earlier findings which suggested that L2 speakers are generally less sensitive to this semantic dimension. This suggests that at high levels of proficiency, L2 speakers may engage with familiar figurative expressions in ways that resemble native processing. At the same time, the data also indicated that L2 participants made less use of extended context, with smaller numerical gains and reduced sensitivity compared to their L1 counterparts.

These findings provide new behavioural evidence that additional vehicle-related elements in extended metaphors reduce processing costs. Moreover, they highlight the importance of language proficiency in shaping metaphor comprehension, suggesting that L2 users — at least those that are highly proficient — can achieve sensitivity to figurative conventions comparable to that of native speakers. This thesis contributes to and builds on our currently available knowledge of how metaphor structure, conventionality, and language status interact in cognitive processes associated with semantic ambiguity resolution in real-time language comprehension.