Emotional intelligence, working memory, and emotional vocabulary in L1 and L2: Interactions and dissociations

Highlights

• Trait emotional intelligence predicts the number of emotion words retrieved.

• Negligible effect of working memory capacity on the retrieval of emotion words.

• Valence effects are less language-dependent and more psychologically-dependent.

• Common psychologically salient words in L1 and L2; predominance of positive words.

• Negative link between working memory capacity and trait emotional intelligence.

Abstract

Cognition and emotion are interrelated concepts. However, very little is known about the relation between working memory capacity (WMC) and emotional intelligence (EI) and their effect on the retrieval and generation of emotional vocabulary. This study aimed to explore correlation patterns between WMC and four factors of trait EI (well-being, self-control, emotionality, sociability). It also sought to examine whether WMC and trait EI are significant determinants of the number and perceived pleasantness of emotion words freely retrieved and produced in a non-emotionally charged context and the extent to which language of retrieval had an effect on this association. The results showed negative correlations between WMC and two factors of trait EI (well-being and emotionality). Trait EI was the sole predictor variable of the perceived pleasantness of the words retrieved, whereas trait EI, language of retrieval, and gender explained a statistically significant amount of the variance in the number of emotion words generated. Qualitative analysis of the emotional vocabulary revealed a slight predominance of positive words and common patterns in the most highly activated words in both first and second languages.

Introduction

Emotion and cognition are two sides of the same coin, influencing each other in multiple ways (Pessoa, 2008). As Swain (2013: 195) pointed out, “Emotions … have a significant impact on what has happened in the past, what is happening now, and what will happen in the future. In fact, emotions are an integral part of cognition”. However, very little is known about the interplay between two widely studied constructs, emotional intelligence (EI) – a concept that connects emotion and cognition (Salovey and Pizarro, 2003) – and working memory (WM) – traditionally seen as a cognitive mechanism par excellence. Moreover, EI and WM are multifaceted constructs, and one of the most challenging questions is whether and how different facets of EI and WM are linked to each other and, by extension, to emotional language processing and production. The current study aimed to expand this line of research by exploring correlation patterns between four factors of trait EI (well-being, self-control, emotionality, sociability) and WM capacity (WMC). In addition, this study sought to examine whether trait EI and WMC are significant determinants of the number and perceived pleasantness of emotion concepts freely retrieved and produced in the absence of an emotionally charged context.

Productive emotional vocabulary was the focus of the study for several reasons. The ability to use emotion concepts constitutes an essential part of human communication and is an indispensable ingredient of sociocultural and pragmatic competence (Dewaele, 2008, Dewaele, 2013, Dewaele, 2018, Ekman, 1999, Pavlenko, 2013). As Ekman (1999: 317) pointed out, “Without being able to name feelings, it is harder to distinguish them, think about them, plan regarding them, etc.”. Therefore, the ability to use a precise and rich vocabulary is an important step towards recognition, acceptance and regulation of emotions. It also promotes the development of effective interpersonal skills, conflict resolution strategies, and negotiation skills (Fisher and Shapiro, 2005). This does not only apply to first language (L1) but also to second language (L2) contexts, as we live in a globalised world where interaction, negotiation, and decision making in an L2 is an unavoidable reality. However, our knowledge regarding productive emotional vocabulary derives mainly from cross-linguistic studies (e.g. Panayiotou, 2006, Pavlenko, 2008b, Semin et al., 2002), and psychological or personality factors underpinning its retrieval and use have received much less attention (Meara, 2002). It is believed that high EI individuals are better equipped to use a wide range of emotion concepts (Barrett, 2017). Some scholars further argued that emotional granularity¹ and WMC go hand in hand (Lee et al., 2017). Therefore, a plausible hypothesis is that EI and WMC would have an additive effect on the ability to retrieve and use emotion words.

There is also an extensive body of literature that investigates how emotions – and emotionality – foster memory capacity (Christianson, 1989, Christianson et al., 1991, Kensinger, 2009, LeDoux, 2003, Zimmerman and Kelley, 2010). Specifically, this work documents that emotional cues and contexts enhance certain aspects of WMC, as well as the subjective vividness and confidence related to retrieval processes. Indeed, emotion prompts and emotionally charged contexts may trigger and induce faster emotional or memory responses. However, being able to retrieve and produce emotion concepts in a neutral (non-cued) condition might require extra cognitive effort. This is particularly true when retrieval processes take place in an L2 because of the cognitive load imposed by L2 processing and the lack of automaticity when L2 users try to access and activate L2 vocabulary that has not been fully consolidated.

Empirical evidence further suggests that L2 vocabulary learning and use require complex underlying mechanisms that depend on individual differences in WMC (Ellis, 1996, Martin and Ellis, 2012; see Gathercole, 2006, for a review) and trait EI (Skourdi and Rahimi, 2014; but see Alavi and Rahimi, 2011). However, it is unclear whether this is a general effect or whether it also applies to specific types of words. Pavlenko (2008a) argued that emotion words should be treated as a unique word class in the mental lexicon because they are represented, encoded, and retrieved differently compared to concrete and abstract terms. Therefore, in order to elucidate the full spectrum of the why and how of emotional expression in general and of the retrieval and production of emotional vocabulary in particular, some minimal consideration of the role of individual differences in both cognitive and affective functions is of paramount importance.

Moreover, the current study qualitatively compared the emotional vocabulary retrieved and produced by English L1, Spanish L1, and Spanish L2 users. Previous research mainly focused on the emotional vocabulary of Spanish L1 (Alba-Juez and Pérez-González, 2019, Bisquerra and Filella, 2018), English L1 (Schrauf and Sanchez, 2004), and bilingual users (Ferré et al., 2010, Ferré et al., 2013). The aim of including Spanish L2 users was to provide data that would contribute to the study of emotion and bilingualism and, eventually, allow for a more comprehensive account of the role of language on memory retrieval processes for emotion words. This analysis focused on the most highly activated words and their valence. The positivity bias theory, also known as the Pollyanna hypothesis (Boucher and Osgood, 1969), posits a higher prevalence of positive words in human communication. It suggests that people's lexicon includes more positive terms, which are used more frequently, appear in the early stages of language development, are learnt faster, and are recalled easier. Therefore, it was hypothesised that participants would mainly retrieve positive words regardless of language of retrieval.