Bidirectional Language Effects: Evidenceon the Emotionality of Bilinguals’Autobiographical Memories fromPupillometry, Eye-movements,and Self-reports

• Dieter Thoma ^[1] ; Pernelle Lorette ^[1] ; Agnieszka Ewa Krautz ^[1]
  1. [1] University of Mannheim

     University of Mannheim

     Stadtkreis Mannheim, Alemania

     
• Localización: Recent advances in second language emotion research / coord. por Irini Mavrou, Mercedes Pérez Serrano, Jean-Marc Dewaele, 2022, ISBN 978-84-1125-575-2, págs. 67-94
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • While research on language-dependent emotions in bilingualsis booming, it becomes increasingly important to warrant the generalizability of its findings. So far, most studies have examined emotion in mono-measure designs with neurophysiological or experiential reactions to linguistic stimuli in receptive tasks. Yet, emotions are multifaceted and may be co-constructed through language(s) during verbalization, which calls for multi-measureapproaches and more productive tasks. This chapter therefore reports on a novel methodology using emotional autobiographical memories (AMs) encoded and told in the first (L1) or second language (L2). Specifically, we investigated if the productive re-experience of AMs replicates reduced L2 emotions for sad but not for happy memories previously observed in receptive tasks with successive bilinguals. We designed a language (L1 German vs. L2 English) by AM-valence (positive vs. negative) experiment and triangulated automatic eye-movement (pupil dilation and fixations) and conscious self-report measures of emotion. Participants assessed AM phenomenology further (e.g., vividness, coherence). Linear mixed-effects models with the pupillometry and self-report data (but not fixations) replicated a language-valence interaction. Yet, instead of the expected pattern of weaker negative L2 emotions, remembering and telling a sadAM in L2 led to stronger emotional reactions than telling a happy one,while positive vs. negative valence did not matter in L1. We assume that sad L2 memories are rare, which grants them salience in memory. Overall, the findings suggest that the multi-level measurement of emotions induced by telling language-specific AMs offers promising insights into the complex nature of language-dependent emotions.