Introduction
People experience emotions in relation to numerous areas of their lives. In evolutionary terms, emotions are adaptive devices. For example, negative emotions (NEs) such as anxiety can stimulate our bodies to expend extra energy necessary to combat or escape from dangers. Positive emotions (PEs) such as enjoyment also play an important role by sustaining actions which contribute to our flourishing, such as satisfying work. However, emotions can also work to our disadvantage, in the same way in which physical pain, which is a generally helpful mechanism alerting us to malfunctions of our body parts, is sometimes debilitating. In the context of language education, a prime example is foreign language anxiety, which often interferes with successful L2 learning. Fortunately, we have some capacity to regulate our emotions.
Of the existing definitions of emotion regulation (ER), the one evoked the most frequently in relation to learning stipulates that ER takes the form of āattempts to influence which emotions one has, when one has them, and how one experiences or expresses these emotionsā (Gross, 2015, pp. 4ā5). In many cases, ER may be largely intuitive and automatic (Braunstein, Gross & Ochsner, 2017; Gyurak, Gross & Etkin, 2011), but often we consciously manage the emotions we experience, as when a student experiencing test anxiety before an important exam uses relaxation techniques to relieve the tension. The ER in which we engage voluntarily and of which we are mostly aware can be called strategic ER; hence, we may speak of emotion-regulation strategies (ERSs). They can be applied to oneās own emotions (self-regulation, intrinsic regulation), as well as to somebody elseās (co-regulation, extrinsic regulation; see Bielak & Mystkowska-Wiertelak, 2020b for a study of teachersā co-regulation of language learnersā emotions), but in this contribution we primarily focus on the former. Thus, any reference to ER concerns emotion self-regulation, unless stated otherwise.
The body of research on the regulation of emotions in different contexts is growing. Examples of areas in which such research has been at least moderately popular include health (e.g., Chung, Kweon, Kang, Hong & Hong, 2018), sport (e.g., Martinent, Ledos, Ferrand, Campo & Nicolas, 2015) and business (e.g., Fang He, SirĆ©n, Singh, Solomon & von Krogh, 2018). The field of education has witnessed only a limited research effort concerning ERSs (e.g., Ben-Eliyahu & Linnenbrink-Garcia, 2013; Harley, Jarrell & Lajoie, 2019; Webster & Hadwin, 2015). Likewise, despite the long-standing interest of language learning (LL) researchers in foreign language anxiety and the more recent surge of interest in the role of other emotions in LL (e.g., Dewaele & MacIntyre, 2014, 2016; Gkonou, Daubney & Dewaele, 2017), there has not been much research on ER in this process. This has been so despite the fact that the importance of learner self-regulation, which includes ER, has been recognized as very important for learning in general (e.g., Boekaerts, Pintrich & Zeidner, 2000; Winne & Hadwin, 2008; Zimmerman, 2008) and LL in particular (e.g., Griffiths, 2013; Oxford, 2011, 2017; Tseng, Dƶrnyei & Schmitt, 2006). The aim of this narrative review, covering the last 25 years (1995ā2019), is therefore to survey the existing research on language learnersā strategic regulation of emotions and frame it within research on ERSs in general learning done mostly in educational psychology. By doing so, we hope to stimulate more research endeavors concerning ERSs in LL, and ultimately raise learnersā and teachersā awareness and skills in this area.
In what follows, we (a) define emotions and briefly review the current views and trends in the research on their role in general and LL; (b) present one widely used model of ERSs and review the research on learnersā ERSs in the context of general learning; (c) review the existing research into language learnersā use of ERSs and (d) propose some general pedagogical implications and research directions for the LL strategy field. In our review, we covered a representative sample of studies concerning ERSs in general education, for which we initially searched in Scopus. We also used the lists of references included in these papers to identify further publications. Concerning the studies on ERSs in LL which investigated more than merely their frequency of use as one category, we have tried to identify all the existing ones using Scopus and also Google searches. Here we review all the ones released by respectable publishers.
Emotions and learning
As with many other complex and multifaceted phenomena, one problem with emotions is that they are very difficult to define (Gross, 2015; Izard, 1993). Numerous definitions from different theoretical perspectives have been formulated (Jarrel & Lajoie, 2017). However, most researchers would probably agree that emotions are highly complex and relatively short whole-body reactions to situations viewed as relevant to the achievement of oneās goals (Gross, 2015; Izard, 1993; Reeve, 2015). The complexity of emotions concerns the way they are triggered and manifested, which may happen at the levels of neurobiology, physiology, expression, cognition and behavior (Gross, 2015; Izard, 1993; Reeve, 2015; Shuman & Scherer, 2015).
Emotions may be activated in many ways (Izard, 1993), but in the context of learning and education, the most often discussed, and probably the most important medium of emotion activation, is some sort of cognitive appraisal (Frijda, 1988; Lazarus & Folkman, 1984). It is also a prominent part of Grossās (1998, 2015) model of emotions, which is based on several previous approaches to emotions. In this model, an emotion is triggered by a situation if one pays attention to some aspects of it and appraises them as somehow relevant to oneās well-being or the achievement of oneās goals. Then, an emotional response is triggered, and it may in turn have some bearing on the situation that has given rise to the emotion, as well as on the attentional and appraisal processes (Harley, Pekrun, Taxer & Gross, 2019). This reveals the cyclic nature of emotion generation.
Certain dimensions of the highly complex experience of emotions may be especially important when considering emotions which accompany learning. One is valence (Pekrun, 2006), which may be either positive or negative; hence the distinction between positive (pleasant) and negative (unpleasant) emotions (An, Ji, Marks & Zhang, 2017). The broaden-and-build theory of PEs (Fredrickson, 2001, 2013) stipulates that the functions and influence of the two types of emotions are different. NEs momentarily limit our experience to facilitate a very narrow focus of attention on a pressing matter such as fleeing a deadly danger. PEs, by contrast, broaden our experience and thus facilitate creative play and exploration, which in the long run builds emotional and cognitive resources in a spiraling manner. The theory thus predicts that in terms of long-term learning, PEs are much more adaptive than NEs. The latter may in fact hinder learning by turning attention away from educationally important experience (e.g., in the classroom), although they may in the short term fuel useful behaviors such as intense study at the last minute before a test.
The range of emotions learners experience in the three major academic achievement situations (Harley et al., 2019), that is, classroom learning, self-study and tests, is very wide. These emotions have been termed academic (Pekrun, Goetz, Titz & Perry, 2002) or, more recently, achievement emotions and defined as emotions directly related to achievement activities, such as classroom learning tasks, and achievement outcomes, such as the knowledge acquired in a learning activity (Pekrun, 2006). Defined in this way, achievement emotions include joy and enjoyment, hope, relief, anxiety, pride, sadness, shame, anger, frustration and boredom.
The concern with achievement/academ...