Las construcciones expresivas desde una perspectiva cognitiva, multimodal e interlingüísticala emoción de alivio

Abstract

This doctoral thesis investigates the pivotal role of interjections and related elements in human communication, challenging their previous dismissal as non-linguistic phenomena in traditional grammar. Contemporary approaches, like Cognitive Linguistics and Construction Grammar, offer an integrative framework linking various linguistic disciplines and multimodality. It conduces to introduce the term expressive constructions to encompass these elements and to explore their connection to emotions. The study emphasizes the need for further research to refine and enhance our understanding of the encoding of verbal and nonverbal expressions, particularly concerning the relief emotion, contributing to a deeper comprehension of this intricate aspect of human communication. The dissertation aims to substantiate the role of expressive constructions in language and expand the research on the multimodal expression of relief emotion, encompassing both verbal and nonverbal modalities. The study develops a methodology grounded in multimodal corpora to achieve these objectives. It establishes two ad hoc multimodal corpora comprising comics and videos, enabling experimentation on the multimodal nature of relief's expressive constructions. The initial empirical studies use comics as a resource to investigate the visual and oral representation of relief's expressive constructions across various languages, including Spanish, French, English, and Japanese. Moreover, the study examines the multimodal translation of these constructions from English and French into Spanish and explores their specific relevance in communication for characters with aphasia. Furthermore, the empirical studies utilize the NewsScape multimodal corpus to extract videos for analyzing the multimodal characteristics of the English expression thank God associated with relief emotion. This analysis employs tools like Praat for phonetic-phonological analysis, Rapid Annotator for systematic and manual annotation of facial and bodily expressions, and OpenPose for automated detection of co-speech gestures in the thank God construction. The research proposes a methodology for the multimodal study of constructions through these innovative computational linguistics tools to investigate and consider their linguistic properties and gestural production. The research findings demonstrate that expressive constructions are fundamental to human communication. The analysis of the multimodal corpora extracts several prototypical expressive constructions for conveying the emotion of relief. Verbal communication exhibits distinct phonetic-phonological and iconic characteristics at both vowel and consonant levels. Similarly, nonverbal communication features specific bodily gestures, such as placing a hand on the chest or forehead, which have metaphorical origins. Noteworthy is the use of morphemes or pictograms symbolizing the exhalation of air through the mouth. Facial expressions include open eyes focused on the interlocutor, raised eyebrows, and parted lips, presumably facilitating air release. While variations exist among languages, the representation of these expressions displays a high degree of similarity. Expressive constructions play a crucial role in expressing and communicating emotions and attitudes. This study reveals distinct characteristics of the relief emotion commonly shared across the investigated languages, highlighting multimodal and cognitive aspects beyond cultural boundaries. The advancements in machine learning offer promising opportunities for comprehensive data analysis and methodologies surpassing linear statistical models, as the relationship between verbal and gestural components appears nonlinear. In conclusion, this research emphasizes the significance of expressive constructions in human communication and provides insights into the intercultural nature of the relief emotion.