Individual differences in identifying creative metaphors from video Ads

Abstract

Reliable identification of metaphors from multimodal discourse has attracted scholarly attention in recent years. However, the role of individual differences in identifying creative metaphors from video ads is underexplored from an empirical perspective. This includes the extent to which individual differences influence metaphor identification in multimodal discourse and how the individual differences result in divergent identification. Our study contributes to addressing these issues by investigating how the background of researching metaphors influences identifying creative metaphors from video ads. We compared results of creative metaphor identification from three metaphor analysts and three external annotators who were novice to metaphor research and probed into the underlying reasons for divergent identification through discussions among six annotators. Both groups of annotators applied Creative Metaphor Identification Procedure for Video Advertisements (C-MIPVA) (Pan & Tay, 2021) into the same 20 Chinese video ads through a systematic process of inter-rater reliability examinations. Results from Fleiss’ Kappa and Percentage Agreement provided substantial support for reliable identification, regardless of the metaphor research background. Discussions among annotators revealed that the interplay between the individual differences in life experience and the influences of temporal and dynamic discourse lead to extra identification, different content, and missing cases of metaphors.