1. Introduction
Although Swain (Reference Swain2013) described emotions in the context of language learning as ‘the elephants in the room’ (p. 205) to indicate that they were insufficiently studied, the past decade has seen significant growth in this area, with emotions being now extensively investigated as part of individual differences in language learning research (see Dewaele, Reference Dewaele2019a). Indeed, the growing interest in emotions is reflected in the number of books (e.g. Miyahara, Reference Miyahara2015; Pawlak et al., Reference Pawlak, Kruk and Zawodniak2024; Pawlak et al., Reference Pawlak, Zawodniak and Kruk2020a; Piniel, Reference Piniel2024), edited volumes (e.g. Csizér et al., Reference Csizér, Smid, Zólyomi and Albert2024; Gkonou et al., Reference Gkonou, Daubney and Dewaele2017), as well as journal articles and special issues (e.g. Dewaele & Li, Reference Dewaele and Li2018; Li, Reference Li2024) published in renowned international journals in the field in recent years. In addition, the growing interest has also led to the founding of the Emotion Special Interest Group for the International Association for the Psychology of Language Learning in 2022.
In order to navigate this new ‘renaissance’ (Li, Reference Li2020) of emotion research in Second Language Acquisition (SLA), we will describe important developments in the field in this paper, including the shift of learner emotions as a peripheral variable in SLA research to a variable that enjoys prominence (Prior, Reference Prior2019), as well as the shift from the exclusive focus on negative emotions (e.g. Horwitz et al., Reference Horwitz, Horwitz and Cope1986) to holistic approaches (e.g. Dewaele & MacIntyre, Reference Dewaele and MacIntyre2014) and dynamic ones (Gregersen et al., Reference Gregersen, MacIntyre and Meza2014a), alongside greater methodological variety and rigor. In the following section, we will then propose a research agenda by highlighting four broad areas, in which we see a need for future research to deepen our understanding of the role emotions play in FL learning. In each of these, specific tasks will be suggested, which will galvanize researchers into investigating as yet unexplored territory in this vibrant field of study, from which both researchers and practitioners would benefit.
2. Theoretical and empirical perspectives on learner emotions
For a long time, emotions had been largely ignored in SLA, partly due to their perceived inferiority to cognitive factors (Swain, Reference Swain2013), as they were often seen as irrational (Prior, Reference Prior2019) and of lower importance due to the smaller effect sizes often found for emotion predictors as opposed to cognitive predictors (Botes et al., Reference Botes, Resnik, Greiff and Stempfer2025b). Dewaele and Li (Reference Dewaele and Li2020) refer to this phase in SLA, which lasted approximately until the mid 1980s, as the ‘Emotion Avoidance Phase’. In early studies in the field, the focus was exclusively on the role of negative emotions, with Horwitz et al. (Reference Horwitz, Horwitz and Cope1986) being pioneers in researching foreign language classroom anxiety (FLCA). FLCA refers to “an individual’s tendency to be anxious in […] the situation of language learning”, meaning individuals have “the trait of feeling state anxiety when participating in (or sometimes even thinking about) language learning and/or use” (Horwitz, Reference Horwitz, Gkonou, Daubney and Dewaele2017, p. 33). As mentioned by Horwitz (Reference Horwitz, Gkonou, Daubney and Dewaele2017; see also Horwitz et al., Reference Horwitz, Horwitz and Cope1986), FLCA is a distinct type of anxiety which bears a resemblance to trait anxiety but only arises in specific contexts. Horwitz et al. (Reference Horwitz, Horwitz and Cope1986) developed the 33-item Foreign Language Classroom Anxiety Scale (FLCAS), which has been widely used to measure FLCA. Their work marked the beginning of what MacIntyre (Reference MacIntyre, Gkonou, Daubney and Dewaele2017) calls the 'Specialised Approach' in anxiety research, meaning the uniqueness and specificity of anxiety in the domain of FL learning and/or use had been acknowledged (see, e.g. Gkonou et al., Reference Gkonou, Daubney and Dewaele2017). This ‘Anxiety-Prevailing Phase’ (Dewaele & Li, Reference Dewaele and Li2020), when anxiety was the only emotion that had been extensively investigated in its own right (Li, Wei, & Jiang (Reference Li, Wei and Jiang2024); MacIntyre, Reference MacIntyre, Gkonou, Daubney and Dewaele2017), lasted until the early 2010s.
With the introduction of positive psychology (PP) (Seligman & Csikszentmihalyi, Reference Seligman and Csikszentmihalyi2000) to SLA (MacIntyre & Gregersen, Reference MacIntyre and Gregersen2012; Mercer & MacIntyre, Reference Mercer and MacIntyre2014), researchers realized that it is equally important to focus on positive emotions in the context of language learning. As suggested by the broaden-and-build theory, a foundational theory of PP (Fredrickson, Reference Fredrickson2001), positive emotions aid learners in developing resources and make them more open to new information, thereby fostering their overall flourishing during the process of language learning. According to Resnik and Mercer (Reference Resnik and Mercer2024), a common misinterpretation of PP is that positive psychologists ignore or deny systemic problems or negative experiences. Instead, positive psychology acknowledges that positive emotions fulfil a different role from negative ones. While positive emotions have a short-term thought-action broadening function (e.g. a broadened perspective and a stronger desire to explore in L2) as well as a long-term resource-building function (e.g. better social resources) when learning an FL, negative emotions have a narrowing effect, typically hindering progress in the FL (e.g. by affecting cognitive processes such as information processing), as well as learning outcomes, and overall language development (Dewaele & MacIntyre, Reference Dewaele and MacIntyre2014, Reference Dewaele, MacIntyre, MacIntyre, Gregersen and Mercer2016; Li, Wei, & Jiang, Reference Li, Wei and Jiang2024; MacIntyre & Gregersen, Reference MacIntyre and Gregersen2012). Due to their different functions in FL learning, positive and negative emotions alike have to be considered when examining emotions in language learning.
Consequently, holistic approaches to exploring the role of positive and negative emotions have been adopted in the ‘Positive and Negative Emotions Phase’ (Dewaele & Li, Reference Dewaele and Li2020), which exponentially increased after Dewaele and MacIntyre’s (Reference Dewaele and MacIntyre2014) landmark publication on 1746 FL learners’ positive and negative emotions across the globe. In this study, the authors introduced foreign language enjoyment (FLE), which they later defined as ‘a complex emotion, capturing interacting components of challenge and perceived ability that reflect the human drive for success in the face of a difficult task’ (Dewaele & MacIntyre, Reference Dewaele, MacIntyre, MacIntyre, Gregersen and Mercer2016, p. 216). Their study corroborated MacIntyre and Gregersen’s (Reference MacIntyre and Gregersen2012, p. 193) claim that positive and negative emotions are not necessarily opposite poles on the same dimension, as in their study, FLE was shown to be a separate dimension from FLCA (Dewaele & MacIntyre, Reference Dewaele and MacIntyre2014; see also the meta-analysis by Botes et al., Reference Botes, Dewaele and Greiff2022a). With the development of a new instrument, the 21-item FLE scale, the authors spearheaded the (positive) ‘affective turn’ (Prior, Reference Prior2019) in the field. A principal component analysis revealed two dimensions of FLE, namely social and private FLE (Dewaele & MacIntyre, Reference Dewaele, MacIntyre, MacIntyre, Gregersen and Mercer2016). While the former includes aspects of FLE which are related to the classroom environment and group dynamics among the learner, teacher, and peers, such as forming a tight group and frequently laughing together in class, the latter captures ‘the personal reactions to learning that reflect relatively inner thoughts and feelings’ (Dewaele & MacIntyre, Reference Dewaele, MacIntyre, MacIntyre, Gregersen and Mercer2016, p. 223), such as a learner feeling proud of their accomplishments or finding what they learn in class interesting. This scale formed the basis of the later validated nine-item Short Foreign Language Enjoyment Scale (S-FLES), which revealed a three-factor structure of the higher-order FLE factor: Personal Enjoyment, which refers to whether a learner is, for example, proud of their accomplishments; Social Enjoyment, which refers to whether the peers form a tight group and if they laugh a lot; and Teacher Appreciation, which refers to whether the teacher is perceived as friendly, nice, and supportive (Botes et al., Reference Botes, Dewaele and Greiff2021). The three subdimensions are measured with three items each. An 11-item FLE scale showing a similar three-factor structure was developed for the Chinese context by Li et al. (Reference Li, Jiang and Dewaele2018). In their ground-breaking study, Dewaele and MacIntyre (Reference Dewaele and MacIntyre2014) additionally extracted eight items from the 33-item FLCAS, leading to a Short Form FLCAS (S-FLCAS), which made it easier to integrate FLCA as a variable in more complex studies and, consequently, allowed deeper insights. The eight-item S-FLCAS was later psychometrically validated, and confirmatory and exploratory factor analyses confirmed its unidimensional structure (Botes et al., Reference Botes, van der Westhuizen, Dewaele, MacIntyre and Greiff2022b).
In short, scholarly work since then has provided deep insights into the conceptualisation of FLE and FLCA, the development of sound instruments to measure them (e.g. Botes et al., Reference Botes, Dewaele and Greiff2021; Botes et al., Reference Botes, van der Westhuizen, Dewaele, MacIntyre and Greiff2022b; Li et al., Reference Li, Jiang and Dewaele2018), and a granular understanding of influential learner-internal and -external variables in learners’ experiences of both in in-person FL classes. Our previous studies, for instance, showed that a higher overall number of languages known was positively linked to FLE and negatively to FLCA, and so was a higher perceived relative standing in the group of learners. Older learners and highly educated ones showed similar patterns (Dewaele & MacIntyre, Reference Dewaele and MacIntyre2014). Small gender differences have emerged, too: female learners reported significantly more FLE and FLCA than male learners (Dewaele et al., Reference Dewaele, MacIntyre, Boudreau and Dewaele2016). Learners who perceived themselves as highly proficient in the language tended to experience higher levels of FLE and lower levels of FLCA (Dewaele & MacIntyre, Reference Dewaele and MacIntyre2014; see also Li et al., Reference Li, Dewaele and Jiang2020). The meta-analyses by Botes et al. (Reference Botes, Dewaele and Greiff2020; Reference Botes, Dewaele and Greiff2022a) demonstrated a moderate negative correlation between FLCA and FL performance and a positive one between FLE and FL performance (see also Li, Reference Li2020). The links between FL emotions and trait emotional intelligence (TEI) (e.g. Li, Reference Li2020; Resnik & Dewaele, Reference Resnik and Dewaele2020), willingness to communicate (Dewaele, Reference Dewaele2019b), L2 grit (e.g. Lee, Reference Lee2020; Sudina et al., Reference Sudina, Brown, Datzman, Oki, Song, Cavanaugh, Tiruchelvam and Plonsky2020), and flow (Dewaele & MacIntyre, Reference Dewaele and MacIntyre2024) have been identified, too. More autonomous students were furthermore shown to experience higher levels of FLE at the tertiary level (Resnik & Dewaele, Reference Resnik and Dewaele2020). One key finding emerging from previous studies, which should be communicated to teachers around the world, is that FLE is a more fleeting experience than FLCA, as it is more related to the classroom environment (Li et al., Reference Li, Huang and Li2021) and is influenced more by teacher characteristics than FLCA (e.g. Dewaele & Dewaele, Reference Dewaele and Dewaele2020). The latter, in turn, is more stable and more related to a learners’ personality, including neuroticism and TEI (Dewaele & MacIntyre, Reference Dewaele, MacIntyre, Sato and Loewen2019; Li et al., Reference Li, Huang and Li2021).
More recently, a diversification of the type of emotions studied has been observed, with research into FL boredom being on the rise (e.g. Dewaele et al., Reference Dewaele, Botes and Meftah2023; Kruk, Reference Kruk2016; Li et al., Reference Li, Dewaele and Hu2023; Pawlak et al., Reference Pawlak, Zawodniak and Kruk2020a). Li et al. (Reference Li, Dewaele and Hu2023) define it as ‘a negative emotion with extremely low degree of activation/arousal that arises from ongoing activities […]. These activities are typically over-challenging or under-challenging and/or of little significance, relevance, or meaning [original emphasis] to the learners’ (p. 234). Based on data gathered from 107 Polish students majoring in English, Pawlak, Kruk et al. (Reference Pawlak, Kruk, Zawodniak and Pasikowski2020) validated the 23-item Boredom in Practical English Language Classes Questionnaire – Revised (BPELC-R), a modified version of Kruk and Zawodniak’s (Reference Kruk and Zawodniak2017) Boredom in Practical English Language Classes Questionnaire (BPELC), which is the very first instrument developed to measure FL-specific boredom. Exploratory factor analysis revealed two dimensions of boredom, namely 1) disengagement, monotony, and repetitiveness, as well as 2) lack of satisfaction and challenge. The development of the scale sparked numerous quantitative studies in the field (e.g. Elahi Shirvan et al., Reference Elahi Shirvan, Yazdanmehr, Taherian, Kruk and Pawlak2023; Kruk et al., Reference Kruk, Pawlak, Elahi Shirvan, Taherian and Yazdanmehr2022a), which followed the predominantly case study-based early approaches used up until this point to explore its role in the FL classroom (e.g. Pawlak et al., Reference Pawlak, Kruk and Zawodniak2020b, Reference Pawlak, Kruk and Zawodniak2020c). The BPELC-R was later revalidated (Kruk et al., Reference Kruk, Pawlak, Elahi Shirvan and Soleimanzadeh2023). Further scales to measure the construct were developed as well, such as Li et al.’s (Reference Li, Dewaele and Hu2023) Foreign Language Learning Boredom Scale (FLLBS) for the Chinese context, originally containing 32 items which loaded onto seven factors, which was later shortened to an 11-item short form that revealed a three-factor structure, including FL activity boredom, FL classroom boredom, as well as general learning boredom (Li, Feng, et al., Reference Li, Feng, Zhao and Dewaele2024). Studies identified that FL boredom is positively linked to FLCA and negatively linked to FLE (e.g. Li et al., Reference Li, Dewaele, Pawlak and Kruk2022). Additionally, FL boredom was shown to be a negative predictor of FL learners’ willingness to communicate (Li et al., Reference Li, Dewaele, Pawlak and Kruk2022) and of academic achievement in the FL. Using latent dominance analysis, Dewaele et al. (Reference Dewaele, Botes and Meftah2023) demonstrated that FL boredom was a stronger predictor of academic achievement than FLE among Moroccan EFL learners; yet it had a weaker predictive effect on academic achievement than FLCA. In brief, over the course of the last decade, a solid foundation has been laid regarding what we know about FLCA, FLE, and FL boredom – their measurement, their predictors, and how they in turn predict certain outcome variables. The time has now come to build upon these foundations and explore new research avenues.
One such avenue is related to the spectrum of emotions studied, which has been broadened in recent years, including shame and guilt (Teimouri, Reference Teimouri2018), happiness (Helgesen, Reference Helgesen, MacIntyre, Gregersen and Mercer2016), pride (Khajavy & Lüftenegger, Reference Khajavy and Lüftenegger2024), curiosity (Mahmoodzadeh & Khajavy, Reference Mahmoodzadeh and Khajavy2019), peace of mind (Zhou, Dewaele et al., Reference Zhou, Dewaele, Lochtman and Xi2023), and embarrassment (Khajavy et al., Reference Khajavy, Smid, Mercer and Murillo-Miranda2025), to name just a few emotions that serve(d) as a new impetus. The findings from previous studies in the field have, furthermore, demonstrated the need for a deeper understanding of emotion dynamics more generally, as emotions have been shown to be, for instance, contagious in the FL classroom, with learner and (perceived) teacher emotions often being aligned (Moskowitz & Dewaele, Reference Moskowitz and Dewaele2021).
Additionally, research into FL learner emotions pointed to their context-sensitivity, with differences observed in their relationships in online and on-site classes (e.g. Resnik & Dewaele, Reference Resnik and Dewaele2023) but also in their levels of experience in different cultures (e.g. Li, Reference Li2020; MacIntyre et al., Reference MacIntyre, Dewaele, Macmillan, Li, Al-Hoorie and MacIntyre2019). Linked to contextual differences, the growing need to explore languages other than English (LOTE) has also become apparent (e.g. Dewaele & Saito, Reference Dewaele and Saito2024). Furthermore, despite the recent increase in studies into FL learner emotions and, along with it, greater methodological variety to study them (see, e.g. Dewaele & Pavelescu, Reference Dewaele and Pavelescu2019; Gregersen et al., Reference Gregersen, MacIntyre and Meza2014a; Kruk et al., Reference Kruk, Pawlak, Elahi Shirvan, Taherian and Yazdanmehr2022b), a recent systematic review by Zhang et al. (Reference Zhang, Gao and Liu2024) on research into FLE revealed that most studies – 67.8%, to be precise – used quantitative designs, of which the vast majority were cross-sectional studies. These were followed by mixed method studies (28.8%), with only 3% of the research designs being qualitative in nature, calling for greater methodological variety to gain a deeper understanding of these complex phenomena. For instance, the attempts to close the gap between researchers and practitioners by conducting intervention studies have been comparatively scarce (e.g. Li & Xu, Reference Li and Xu2019). Hence, in the following, we will outline a research agenda in four areas of enquiry, where we feel further research is necessary to improve our theoretical understanding of the role of FL learners’ emotions and to make the knowledge gained practically more useful for practitioners. These areas include 1) the broadening of the type of emotions studied, 2) understanding their dynamics, 3) the need for a diversification of research contexts, and 4) bridging the research-practice gap. For each of these areas, we will provide a suggestion of several concrete tasks which, we believe, would advance our understanding of emotions in FL learning in the future. While the list of these promising research areas, including the tasks linked to them, is by no means exhaustive, they are aimed at giving a much-needed new impetus to research in the field.
3. Research tasks
3.1. Diversifying emotions and the underpinning of emotional theory
3.1.1. Research task 1: Diversifying the emotions studied in SLA
Research into SLA has made significant strides in integrating emotions into the broader nomological network of variables often studied in the field, such as motivational factors (e.g. Saito et al., Reference Saito, Dewaele, Abe and In’nami2018), engagement (e.g. Hamedi et al., Reference Hamedi, Pishghadam and Fadardi2019; Khajavy, Reference Khajavy, Hiver, Al-Hoorie and Mercer2021), academic achievement, and language proficiency (e.g., Botes et al., Reference Botes, Dewaele and Greiff2020, Reference Botes, Dewaele and Greiff2022a). In doing so, affective variables have been studied through diverse lenses in SLA research; however, the scope of emotions included in these research endeavors remain limited. Anxiety and enjoyment have formed the cornerstones of emotion research in the field, undoubtedly due to the influential works of Horwitz (Horwitz et al., Reference Horwitz, Horwitz and Cope1986; Horwitz, Reference Horwitz, Gkonou, Daubney and Dewaele2017) and Dewaele and MacIntyre (Reference Dewaele and MacIntyre2014, Reference Dewaele, MacIntyre, MacIntyre, Gregersen and Mercer2016). Somewhat more recently, the emotion boredom has also received considerable research attention, with research studies incorporating boredom into the nomological network of SLA (see Kruk et al., Reference Kruk, Pawlak, Elahi Shirvan, Taherian and Yazdanmehr2022b; Pawlak et al., Reference Pawlak, Zawodniak and Kruk2020a; Pawlak et al., Reference Pawlak, Kruk and Zawodniak2024). This trio of emotions (anxiety, enjoyment, and boredom) has become ubiquitous in SLA emotion research to the extent that Dewaele et al. (Reference Dewaele, Botes and Meftah2023) likened the three emotions to a ‘three-body problem.’ While these emotions have proven to be highly relevant, the overt focus limits our understanding of the full affective landscape that shapes language learning.
Some limited additional inroads have been made in terms of additional emotions studied in SLA. A conceptual framework and accompanying measure of pride in FL learning has recently been introduced by Khajavy and Lüftenegger (Reference Khajavy and Lüftenegger2024). In addition, psychometric measures of shame and guilt were also introduced by Teimouri (Reference Teimouri2018), while qualitative explorations of love of language learning (Pavelescu & Petrić, Reference Pavelescu and Petrić2018) and theoretical discussions on love of language teaching (Barcelos & Coelho, Reference Barcelos, Coelho, MacIntyre, Gregersen and Mercer2016) can also be found in the literature. However, these explorations of complex emotions beyond the oft-studied anxiety, enjoyment, and boredom remain few and far in-between, with studies often examining emotions in isolation as opposed to parts of a greater theoretical model of diverse emotions.
In comparison, emotion research in psychology offers numerous empirically grounded frameworks that contain diverse ranges of emotions, often categorized as multi-faceted, hierarchical, and context dependent. For example, the basic emotions as categorized by Ekman (Reference Ekman1992) include anger, fear, disgust, sadness, happiness, and surprise. These emotions are theorized to be universal, each with distinct facial expressions and physiological responses.
In contrast, the Core Affect Model (Russell, Reference Russell1980; Russell & Barrett, Reference Russell and Barrett1999) conceptualizes emotions not as discrete categories but also states that they vary along two continuous dimensions: (1) valence (pleasant – unpleasant) and (2) arousal (activating – deactivating). Affective experiences are therefore seen as a combination of valence and arousal, with more basic emotions (e.g. fear as negative valence and high arousal) and more complex emotions such as calmness (positive valence, low arousal) or boredom (negative valence and low arousal) all captured within one model.
The Control-Value Theory (CVT) of Pekrun (Reference Pekrun2006) also offers distinct categories of emotions, viewed through the lens of valence and arousal. The CVT examines achievement emotions of learners as a result of their perceived control over their learning and the subjective value they place on a learning outcome. The theory’s associated measure (Achievement Emotions Questionnaire) captures eight distinct emotions within the context of learning, including positive valence and high arousal emotions (enjoyment, hope, and pride), negative valence and high arousal emotions (anxiety, anger, and shame), and negative valence and low arousal emotions (boredom and hopelessness).
Studies in education psychology have in recent years gravitated towards the CVT and its associated measure, with researchers examining the eight emotions and their complex interactions in conjunction with learning outcomes (e.g. Earl et al., Reference Earl, Bishop, Miller, Davison and Pickerell2024; Ranellucci et al., Reference Ranellucci, Hall and Goetz2015). Studies in SLA examining a broad range of emotions in the context of language learning remain rare. MacIntyre and Vincze (Reference MacIntyre and Vincze2017) examined 10 positive emotions and nine negative emotions as categorized by Frederickson’s (Reference Fredrickson2013) Differential Emotions Scale and individual differences in SLA. The study found consistent strong positive associations between positive emotions and motivation-related variables, with negative emotions providing mixed results and weaker associations. The use of CVT in SLA has also led to some studies examining a range of emotions, such as Shao et al. (Reference Shao, Stockinger, Marsh and Pekrun2023), who found similar patterns to MacIntyre and Vincze (Reference MacIntyre and Vincze2017), where positive emotions were more strongly related to motivation and achievement outcomes than negative emotions.
These studies examining a range of emotions in language learners offer particular promise in SLA research by demonstrating that a richer affective palette can illuminate patterns of motivation, engagement, and performance that remain invisible when the focus is restricted to anxiety, enjoyment, and boredom. To capitalize on this promise, future work must systematically expand the emotional repertoire under investigation, while anchoring that expansion in well-articulated theoretical frameworks such as Basic Emotions Theory, Core Affect, and CVT.
Research task 1: Expand and map the emotional landscape of FL learners using a dimensional framework grounded in real-time data collection.
Most SLA emotion research has centered on a small set of discrete emotions (primarily anxiety, enjoyment, and boredom), leaving much of the emotional terrain of language learning unexplored. To move beyond this narrow focus, researchers are encouraged to adopt a dimensional model of emotion, such as the Core Affect Model (Russell, Reference Russell1980; Russell & Barrett, Reference Russell and Barrett1999), which conceptualizes emotional experience along two key dimensions: valence (pleasant – unpleasant) and arousal (high – low activation). This framework enables the inclusion of a broader and more nuanced range of emotions, including not only basic states like fear or happiness, but also more complex or mixed affective experiences such as calmness, tension, curiosity, or frustration.
Using the Experience Sampling Method (ESM) (Goetz et al., Reference Goetz, Sticca, Pekrun, Murayama and Elliot2016), researchers can collect in-the-moment data from FL learners during regular classroom activities over a sustained period (e.g. 2–3 weeks). Learners can report on various emotional states multiple times per day using brief measures such as single item questions per emotion state. This approach provides a dynamic and fine-grained picture of learners’ emotional trajectories across various tasks and interactional contexts.
Once collected, data can be analyzed using multilevel modeling to capture within- and between-person variability, and clustering techniques (e.g. latent profile analysis or k-means clustering) to identify emotion patterns or profiles that tend to co-occur during certain learning events. These emotion clusters can then be examined in relation to key outcome variables such as learner engagement, willingness to communicate, or perceived task success. This opens up the possibility of identifying emotion constellations (e.g. high arousal-positive valence states like excitement and curiosity) that are particularly conducive to learning or to identify negative constellations (e.g. high arousal-negative valence states such as anger and fear) that may be particularly disruptive for language learning progress.
By incorporating a dimensional framework and emphasizing emotional diversity, this research task offers a way to diversify the range of emotions examined within the context of SLA and moves beyond isolated variables to develop an empirically grounded map of learner affect, with clear implications for classroom design and affect-aware pedagogy.
3.2. Broadening the understanding of emotion dynamics
3.2.1. Research tasks 2 and 3: Examining emotional contagion in the FL classroom
While FL learner and teacher emotions have been extensively investigated separately in the past decade (e.g. Dewaele, Reference Dewaele2019a; Dewaele & MacIntyre, Reference Dewaele and MacIntyre2014; Proietti Ergün & Dewaele, Reference Proietti Ergün and Dewaele2021), research into their interplay in the FL classroom has been comparatively scarce (Moskowitz & Dewaele, Reference Moskowitz and Dewaele2021). This is surprising given that previous research showed that teacher behaviour influences learner emotions, in particular FLE (Dewaele et al., Reference Dewaele, Witney, Saito and Dewaele2017; Dewaele & Dewaele, Reference Dewaele and Dewaele2020; Dewaele & MacIntyre, Reference Dewaele, MacIntyre, Sato and Loewen2019), and learner behaviour, in return, influences teachers’ emotional experiences in class (e.g. Hagenauer et al., Reference Hagenauer, Hascher and Volet2015). Hagenauer et al. (Reference Hagenauer, Hascher and Volet2015), for instance, found that students’ engagement, to which emotions are key, significantly predicted teacher emotions. Additionally, the study showed that a perceived lack of discipline was a significant predictor of teachers’ anger. The strongest predictor of the teachers’ joy and anxiety was the bond formed between teachers and students, which illustrates that the teacher-student relationship in the classroom is not only crucial for students’ achievement (see, e.g. Hattie, Reference Hattie2008) but also for teachers themselves.
Learning in classrooms or even in informal groups is a social experience, with dynamic interactions between students, teachers, and peers. These dynamic interactions take place within the context of an emotional classroom climate (Alonso-Tapia & Nieto, Reference Alonso-Tapia and Nieto2019) shaped by social and cultural norms regarding emotion (Safdar et al., Reference Safdar, Friedlmeier, Matsumoto, Yoo, Kwantes, Kakai and Shigemasu2009; Valiente et al., Reference Valiente, Swanson, DeLay, Fraser and Parker2020), where engagement leads to emotion feedback loops that are often termed emotional contagion. This process captures an individual’s or group’s emotions influencing the emotions and behaviours of another individual or group, either consciously or unconsciously, by triggering emotional and behavioural responses (Schoenewolf, Reference Schoenewolf1990). Previous research has found evidence of emotional contagion in classrooms, with teachers’ emotions influencing students, students’ emotions influencing teachers, and students’ emotions influencing their peers (Mottet & Beebe, Reference Mottet and Beebe2000; Shao et al., Reference Shao, Jiang, Kutuk and Parkinson2025). Indeed, emotional contagion has been linked to positive valence and the feeling of emotional support (Frisby, Reference Frisby2019), teacher satisfaction (Houser & Waldbuesser, Reference Houser and Waldbuesser2016), and student engagement (Frenzel et al., Reference Frenzel, Becker-Kurz, Pekrun, Goetz and Lüdtke2018).
The language classroom may prove fruitful ground for examining emotions in a social context, especially through the lens of emotional contagion. Very few studies have examined emotional contagion within the context of language learning (Kong, Reference Kong2022). By measuring moment-to-moment changes in FLE, Talebzadeh et al. (Reference Talebzadeh, Elahi Shirvan and Khajavy2020) found evidence for emotional contagion of state emotions via mimicry in five dyadic teacher-student interactions. In addition, three further studies examined the relationships between student emotions and perceived teacher emotions. Amiri et al. (Reference Amiri, Meftah and Botes2025) examined the relationship between quantitatively measured trait-level student emotions and perceived teacher emotions as predictors of student attitudes, whereas Amiri (Reference Amiri2024) qualitatively examined the relationship between student and perceived teacher anxiety, enjoyment, and boredom. Furthermore, Moskowitz and Dewaele (Reference Moskowitz and Dewaele2021) also examined perceived teacher emotion, by exploring the link between perceptions of teacher happiness and language learner attitude. Another study on emotional contagion in the FL class by Dewaele and Li (Reference Dewaele and Li2021) focused on the relationship between tertiary-level learners’ perceived teacher enthusiasm, their FLE and boredom in EFL classes, as well as the students’ socio-behavioural engagement in class. Based on Frenzel et al. (Reference Frenzel, Goetz, Lüdtke, Pekrun and Sutton2009) and Keller et al. (Reference Keller, Hoy, Goetz and Frenzel2016), they defined teacher enthusiasm as being characterized by both affective and behavioural dimensions. Statistical analyses of the data from 2,002 Chinese students revealed that perceived teacher enthusiasm is a reliable predictor of learners’ socio-behavioural engagement. FLE and boredom functioned as mediators in the model. Lastly, Sulis and Mercer (Reference Sulis and Mercer2025) investigated the development of and interrelationship between Austrian secondary-level EFL learners’ and their teacher’s engagement, of which emotions are a crucial element (Dewaele & Li, Reference Dewaele and Li2021), in three English lessons in real time based on ESM (Goetz et al., Reference Goetz, Sticca, Pekrun, Murayama and Elliot2016). The ESM data was complemented by recordings of the lessons, classroom observations as well as stimulated recall interviews. The analyses showed that in contrast to behavioural engagement, the students’ and their teacher’s emotional engagement in class were much more synchronized. Furthermore, the teacher’s explanations demonstrated her awareness of the powerful impact of emotional contagion, and they additionally indicated that emotional contagion occurred bi-directionally.
Although all five studies found evidence of emotional contagion between students and their teacher, it should be noted that no trait-level study measured teacher emotions directly but rather relied on students’ perceptions of their teacher’s emotion. Similarly, a recent study by Shao et al. (Reference Shao, Jiang, Kutuk and Parkinson2025) demonstrated emotional transmission between peers in the FL classroom based on self-report of learners’ own emotions and their perception of those experienced by their peers. Future research examining the interplay of student and teacher emotion within the context of the language classroom is therefore needed, which can take the form of a longitudinal trait-based study (Research Task 2) or ESM state-based study (Research Task 3).
Research task 2: Examine emotional contagion in the FL classroom between students and teachers and the effect of such contagion on language learning outcomes.
Researchers can draw on studies situated in educational psychology, with a prime example of such a study being Frenzel et al.’s (Reference Frenzel, Becker-Kurz, Pekrun, Goetz and Lüdtke2018) study exploring longitudinal emotional transmission of enjoyment in the classroom between students and teachers. Through examining n = 1,643 students and n = 69 teachers across a six-month longitudinal multi-wave study, Frenzel et al. (Reference Frenzel, Becker-Kurz, Pekrun, Goetz and Lüdtke2018) were able to demonstrate that enjoyment contagion was related to enthusiasm and engagement. Studies in SLA can also utilize such a longitudinal design and collect data from students and their respective teachers over several data points. Emotions can be measured as trait variables using existing scales in SLA, such as the S-FLES (Botes et al., Reference Botes, Dewaele and Greiff2021) and the Foreign Language Teaching Enjoyment Scale (Proietti Ergün & Dewaele, Reference Proietti Ergün and Dewaele2021). Researchers are encouraged to examine emotional contagion in relation to outcome variables, à la Frenzel et al. (Reference Frenzel, Becker-Kurz, Pekrun, Goetz and Lüdtke2018), including language learning engagement, language learning motivation, and academic achievement.
Researchers are also further encouraged to examine group-level differences in the effects of emotional contagion, as some students may be more susceptible to the effects of emotional contagion (Marx et al., Reference Marx, Frenzel, Fiedler and Reck2024), gender differences may arise (Sonnby-Borgström et al., Reference Sonnby-Borgström, Jönsson and Svensson2008), or teacher behaviours such as the use of humour may play a role in the manifestation of emotions and emotional contagion (Neff & Dewaele, Reference Neff and Dewaele2023). Although such research naturally lends itself to quantitative designs via mixed effects modelling or latent modelling, researchers are also encouraged to utilize mixed-methods designs (see, e.g. Shao et al., Reference Shao, Jiang, Kutuk and Parkinson2025) as more varied and detailed information on emotion contagion may be generated through qualitative data analysis.
Research task 3: Examine the development of and interplay between students’ and teachers’ emotions during FL lessons in real time.
Research into emotional contagion in the FL classroom has predominantly focused on tertiary-level learners’ ratings of their own emotions and how they perceive their teachers based on web-based questionnaires (Dewaele & Li, Reference Dewaele and Li2021; Moskowitz & Dewaele, Reference Moskowitz and Dewaele2021). As previous research showed that emotions fluctuate over time (e.g. Elahi Shirvan & Talebzadeh, Reference Elahi Shirvan and Talebzadeh2017; MacIntyre, Reference MacIntyre2012), future research exploring emotional contagion in real time based on teachers’ and their learners’ ratings of their own emotions is much needed to gain direct insights into the dynamic interplay between both.
Focusing on the development of learner and teacher engagement during several lessons, Sulis and Mercer (Reference Sulis and Mercer2025) were able to show the benefits of ESM to capture such trajectories by collecting self-reported data from participants in real time longitudinally. At regular intervals marked by a signal (Sulis & Mercer, Reference Sulis and Mercer2025 chose seven-minute intervals in their study), the participants were asked to complete a short survey (Goetz et al., Reference Goetz, Sticca, Pekrun, Murayama and Elliot2016). Future research could apply this method to focus on specific emotions, such as FLE (Botes et al., Reference Botes, Dewaele and Greiff2021) and FLTE (Proietti Ergün & Dewaele, Reference Proietti Ergün and Dewaele2021) or happiness (Moskowitz & Dewaele, Reference Moskowitz and Dewaele2021), and could even include variables which have been shown to be related to contagion, such as enthusiasm (Dewaele & Li, Reference Dewaele and Li2021). Similar to Sulis and Mercer (Reference Sulis and Mercer2025), an app, such as M-Path, could be used on tablets, which would allow learners and their teacher to rate their emotions at specific points in time, which consequently allows a quantification of the development of the specific emotion(s) investigated at a particular time during the lesson. This method eventually allows researchers to see to what extent students’ and their teacher’s emotions converge or diverge during the lesson (both between peers as such, as well as between the teacher and peers). In addition to video-audio recordings of the lessons, Sulis and Mercer (Reference Sulis and Mercer2025) furthermore included classroom observations as well as stimulated recall interviews (Gass & Mackey, Reference Gass and Mackey2016), which proved meaningful and crucial for gaining insights into possible reasons underlying the participants’ experiences. Investigating emotions of learners and teachers alike this way would allow a granular understanding of yet unexplored emotional classroom dynamics.
3.2.2. Research task 4: The expression of teaching emotions in the FL classroom
Although emotions experienced by FL teachers have gained increased research attention in recent years (see Proietti Ergün & Dewaele, Reference Proietti Ergün and Dewaele2021), the distinction between felt emotions and expressed emotions is rarely made in the literature. For example, the oft-used Foreign Language Teaching Enjoyment Scale (Proietti Ergün & Dewaele, Reference Proietti Ergün and Dewaele2021) measures teaching enjoyment through items that capture the feeling of enjoyment when teaching (‘I enjoy foreign language teaching’) but it does not provide information regarding whether or not the teacher expresses this feeling of enjoyment to learners. The distinction between felt and expressed emotions can be linked to the Basic Emotion Theory (Ekman, Reference Ekman1992), where emotions are experienced (in a mostly universal way) in order to react to certain stimuli and instigate action (Frijda, Reference Frijda1986) and emotional expression is the communication of these felt emotions, often through facial expressions, body movements, and vocalizations (Keltner et al., Reference Keltner, Sauter, Tracy and Cowen2019).
This distinction between feeling an emotion and expressing that emotion is especially relevant to teachers, as the need for emotional regulation, emotional labour, and emotional display rules are often self-reported by teachers (Chang, Reference Chang2020). Teachers may not always display their true felt emotions, as the expression of negative emotions is often seen as a hindrance to learning (Sutton et al., Reference Sutton, Mudrey-Camino and Knight2009) and the display of positive emotions is seen as conducive to learning as it creates a supportive atmosphere (Pekrun, Reference Pekrun2006) and can lead to emotionally competent students (Poulou & Denham, Reference Poulou and Denham2023). These perceptions of positive versus negative emotional displays can lead to teachers choosing to suppress negative emotions, as found in the qualitative study of Kimura (Reference Kimura2010), where teachers exerted more control over negative emotions than positive emotions. This selective display of emotions can be termed emotional labour, which is defined as the expression or non-expression of felt emotions during the course of job-relevant interactions (Hochschild, Reference Hochschild1983). Emotional labour is especially relevant to the context of FL teaching, as FL ‘teachers end up being emotional laborers who find themselves often having to reconcile tensions between expression of feelings – as dictated by professional circumstances – and their actual feelings’ (De Costa & Nazari, Reference De Costa and Nazari2024, p. 1159). Furthermore, emotional labour of FL teachers can adversely affect well-being (Geng et al., Reference Geng, Zhou and Yu2024) and is positively linked to teacher burnout (Blake & Dewaele, Reference Blake and Dewaele2023). Furthermore, the expression of emotions and the performance of emotional labour are linked to emotional display rules, which are social norms that define whether or not the display of certain emotions is appropriate in a given context (Chang, Reference Chang2020). For example, it may be seen as appropriate to display irritation or disappointment towards students in some schools, but not anger (Levering, Reference Levering2000). These emotional display rules can be highly culturally dependent, for example, some cultures and schools may deem the display of anger by a teacher to be appropriate, whilst in others such behaviour may be considered inappropriate (Cole et al., Reference Cole, Bruschi and Tamang2002; Schutz, Reference Schutz2011, as cited in Chang, (Reference Chang2020)). These emotional display rules and the suppression of naturally felt emotions can have a negative effect on teachers, as found in Chang (Reference Chang2020), where emotional display rules and expressive suppression were associated with burnout in teachers, with increased expressive suppression positively linked to exhaustion, depersonalisation, and inefficiency in teachers.
In addition to impacting the well-being of teachers, the emotional expression of FL teachers can also have an impact on the learning of FL students. Emotions can be viewed as social constructs, with emotional expressions communicating an individual’s internal state to others, which can be interpreted as behavioural intentions (Van Kleef, Reference Van Kleef2009). The interpretation of emotional expressions of others and their possible behavioural intentions lead to a reaction in the observer, which can also take the form of internally experienced emotions, emotional expressions, and behavioural intentions, as detailed in the emotions as social information (EASI) theory (Van Kleef, Reference Van Kleef, Ashkanasy and Cooper2008, Reference Van Kleef2009). The EASI theory posits that emotional expressions influence others via (1) inferential processes, where observers interpret the meaning of emotions, and (2) affective reactions, where observers experience emotions in response (Van Kleef, Reference Van Kleef, Ashkanasy and Cooper2008). For example, in the language classroom, a language teacher feeling pride due to high exam scores will express pride through voicing the emotion and through facial expressions. Students, in turn, will interpret this emotion and make a behavioural judgement regarding their teacher and may experience joy and pride due to the pride expressed by their teacher. As such, the inherently social experience of expressing emotions ought not to be disregarded in language emotion research, especially considering the inherently social context of a language classroom. Previous studies have found that the emotional expression of teachers can impact student learning, with Poulou and Denham (Reference Poulou and Denham2023) finding that positive emotional expression led to increased school adjustment and emotional competence development in young learners. In contrast, van Doorn et al. (Reference Van Doorn, van Kleef and van der Pligt2014) found that negative emotional displays can motivate change. In a series of experimental studies, where undergraduate students had to learn word pairs, the researchers found that the display of anger by an instructor led to better learning performance than the display of happiness. However, this finding needs to be treated with caution as cultural (in-)appropriateness related to teachers’ expression of anger might play a crucial role in determining whether and to what extent a teacher is allowed to express anger in the classroom.
Given the effect of emotional expression, emotional labour, and emotional display rules on teachers, and the possible effect of emotional expression by teachers on students’ learning performance, there is a need to examine emotional expression within the context of FL teaching (see Research task 4).
Research task 4: Examine the interrelationships between FL teachers’ expressed emotions, emotional display rules, and student learning outcomes.
Following the precedent set by Poulou and Denham (Reference Poulou and Denham2023) in including emotional expressiveness of teachers in relation to student outcomes and the study by Chang (Reference Chang2020) where emotional display rules were linked to expressive suppression, researchers in SLA can embark on studies examining the relationships between expressed emotions, emotional display rules, and student outcomes within the context of FL learning. These studies will need to measure expressed emotions in the FL class, the prevalence of emotional suppression, and the perceived emotional display rules by teachers. These studies can further integrate concepts such as emotional labour (Blake & Dewaele, Reference Blake and Dewaele2023), burnout (Wang, Reference Wang2022), and well-being (Mercer, Reference Mercer2017) of teachers to further understand the nomological network of emotional expression of teachers. Ideally, studies examining emotional expression of teachers would also incorporate student outcomes, with student emotions (possibly distinguishing between students’ felt and expressed emotions), language learning motivation, and classroom engagement especially being of interest, due to the strong links these variables have with teacher behaviours (see Dewaele et al., Reference Dewaele, Saito and Halimi2022; Henry & Thorsen, Reference Henry and Thorsen2018; Zhou, Hiver et al., Reference Zhou, Hiver and Al-Hoorie2023). These studies are particularly well-suited for self-report research where, for example, emotional expressiveness can be measured quantitatively via the Self-Expressiveness in the Classroom Questionnaire (SEQ; Poulou & Denham, Reference Poulou and Denham2023). Longitudinal designs will be a boon, as conclusions regarding directionality and causality can be made in terms of emotional expression in teachers impacting their own and student outcomes. Furthermore, studies can consider the possible impact of demographic variables in the emotional expression of teachers, as age (Gross et al., Reference Gross, Carstensen, Pasupathi, Tsai, Götestam Skorpen and Hsu1997), culture (Mesquita & Walker, Reference Mesquita and Walker2003), and gender (Simpson & Stroh, Reference Simpson and Stroh2004) may play a role. As with the previous research tasks, mixed-methods designs are especially encouraged, as qualitative data will enrich research insights made into the effect of the expressed emotions of teachers in the FL classroom.
3.3. Diversifying research contexts
3.3.1. Research task 5: The need for more research on languages other than English (LOTE)
Since Dewaele and MacIntyre’s (Reference Dewaele and MacIntyre2014) seminal work in the field, much knowledge has been gained on positive and negative learner emotions in in-person FL classes, mostly EFL classes, across the globe based on data gathered from FL learners from a wide range of L1 backgrounds, countries, ages, different educational levels, and proficiency levels. Studies from the Chinese context, for instance, already pointed to the context-sensitivity of learner emotions by demonstrating that Chinese EFL learners tended to report lower FLE scores than found in other contexts (e.g. Li, Reference Li2020). Despite the great variety of learners studied so far, what is observable in most studies on FL learners’ emotions is their disproportionate focus on one specific target language, which is English. Hence, a similar trend to the one outlined by Plonsky (Reference Plonsky2023) for LX research more broadly (LX referring to any language acquired after the age of 3 [Dewaele, Reference Dewaele2018]) is noticeable in studies on the role of emotions in FL learning and teaching as well. This is problematic to a certain extent for the following reason: ‘From a practical standpoint, we are simply unable to speak to and inform the many matters of an applied nature that pertain to understudied contexts and demographics when our research does not sample them’ (Plonsky, Reference Plonsky2023, p. 9). While this critique should not downplay the great value of previous contributions in the field, we feel it is high time to put a much stronger focus on yet unexplored populations and contexts in order to gain deeper insights into the generalizability of the findings gained so far and to represent the multilingual world in all its facets in order to help every FL learner to thrive and flourish.
At this point, it is important to highlight that some studies have been conducted on, for instance, languages other than English (LOTE), meaning they have not been ignored completely so far. Dewaele and MacIntyre (Reference Dewaele and MacIntyre2022), for instance, compared 761 EFL learners’ and 825 LOTE learners’ FLE, FLCA, and the proportion of time these learners experienced a state of flow. Statistical analyses revealed that the LOTE group reported experiencing significantly higher levels of FLE and lower levels of FLCA than the EFL group and they reported spending more time in a state of flow, too. A study by Dewaele and Saito (Reference Dewaele and Saito2024), in which 360 tertiary-level learners from Kuwait who all knew English but studied a variety of FLs participated, showed that students studying LOTE reported higher levels of FLE, similar levels of FLCA, and lower levels of motivation than students who studied English. Taken together, these findings suggest that the prestigious role of English as a lingua franca does not necessarily boost learners’ enjoyment. Additionally, Dewaele and Proietti Ergün (Reference Dewaele and Proietti Ergün2020) found in their comparative study on 110 Turkish learners’ emotions, motivation, and course marks in two FLs, Italian and English, that FLE was positively linked across both FLs, while FLCA and motivation were unrelated in both FLs. The predictors of course marks differed, too: while FLCA was the strongest predictor in EFL classes, motivation played a much bigger role in Italian FL classes. Overall, while some knowledge has already been gained in FL contexts other than English, we feel much is still to be explored regarding potential causes and outcomes of such differences.
Building on prior research that revealed contextual differences in different types of FL classes, we believe future research should further explore the role of language learners’ emotions in migration contexts. In education settings, a deeper understanding of the emotional climate is not only crucial for the language classroom, but learners’ emotions linked to LX learning and/or use might significantly influence learners’ academic achievements, their wellbeing, and thriving in other subjects, too. Foreign language classes might prove particularly fruitful to explore though, as they often offer immigrants the rare possibility to be on equal (or even superior) ground with their peers on the level of language competence. As Mercer and Gregersen (Reference Mercer and Gregersen2023, p. 11) put it, it is high time to use the transformative potential of PP and combine it with a ‘[l]ens of social criticality,’ with the ‘[i]ntention for positive social change.’ Research task 5 provides concrete ideas for future research designs in this respect.
Research task 5: Explore the emotional experiences of language learners with a migration background and those raised in the target country and identify similarities and differences in their (foreign) language learning trajectories.
Example studies to draw on would be, for instance, the study by Dewaele and Proietti Ergün (Reference Dewaele and Proietti Ergün2020) on English and Italian FL classes mentioned above or a study conducted by Resnik and Dewaele (Reference Resnik and Dewaele2020) on learners’ emotional experiences in L1 and FL classes. The data collection instrument could be a web survey, in which widely used scales to measure learner emotions (see, e.g. Botes et al., Reference Botes, Dewaele and Greiff2021 for FLE, Dewaele & MacIntyre, Reference Dewaele and MacIntyre2014 for FLCA, Li et al., Reference Li, Dewaele and Hu2023 for FL boredom) could be selected and adapted to investigate the contexts concerned. Data could be collected from two groups of learners: those who migrated to the country concerned (a country with a language other than English as the official language) and study the language of the host country as one of their additional languages, and learners who grew up with the L1 language spoken in the country. The opportunities are manifold here. Due to the previously mentioned context-sensitivity of learner emotions (see, e.g. Resnik & Dewaele, Reference Resnik and Dewaele2020), data could be, for instance, collected from primary-, secondary- or tertiary-level learners on learners’ emotions in language classes in which the official language of the host country is taught, FL classes, or their language enjoyment and anxiety in open-content classes. The suggested group comparison is not necessary as data collected solely from immigrants would yield rich insights, too. Including open-ended questions in the survey (see, e.g. Dewaele & MacIntyre, Reference Dewaele and MacIntyre2014, Reference Dewaele, MacIntyre, MacIntyre, Gregersen and Mercer2016), such as participants’ description of sources of the specific emotion(s) studied (e.g. describing an episode they found anxiety-provoking or which they enjoyed in class), or data gathered from interviews would, furthermore, provide an in-depth understanding of the similarities and differences in learners’ emotional experiences and it would give learners the opportunity to share their experiences (see, e.g. Resnik et al., Reference Resnik, Dewaele and Knechtelsdorfer2023a, Reference Resnik, Dewaele and Knechtelsdorfer2023b). Focusing on younger learners (i.e. those at the primary or secondary level) in this context might be particularly fruitful due to the observed predominant focus on learners at the tertiary level in applied linguistics (Plonsky, Reference Plonsky2023).
Additionally, by collecting background information on the participants, such as their language learner history and demographics, potential sources of differences in their emotional experiences in the different contexts could be investigated, too, and so could be potential links to their achievements, such as learners’ grades in these classes or, for instance, their willingness to communicate (see, e.g. Botes et al. [Reference Botes, Resnik, Moskowitz, Dewaele, Mercer and Greiff2025a] as a sample study investigating such a network of potentially interwoven elements in a different context, namely online FL learning). Gaining a deeper understanding of migrants’ emotional experiences would be beneficial for teachers, too, as they would be able to consider the insights gained for their teaching. Eventually, the findings would yield rich insights into ways to better support migrants in their (foreign) language learning trajectories.
3.3.2. Research task 6: The need for more research on different FL learning settings
Linked to the need to diversify research contexts, we feel another important future direction is to explore the role of positive and negative learner emotions in different FL learning settings. One of the pioneering studies in this respect was conducted by De Smet et al. (Reference De Smet, Mettewie, Galand, Hilingsmann and Van Mensel2018) who investigated 896 learners’ emotions in Content and Language Integrated Learning (CLIL) and non-CLIL classes in primary and secondary schools in French-speaking Belgium. Their research showed that the type of instruction mattered as students who had CLIL instruction experienced significantly lower levels of FLCA. Additionally, target language-related differences were observable, too: learners studying Dutch reported significantly higher levels of FLCA and lower levels of FLE than those studying English, which was attributed to the tense political relationship with the Dutch-speaking community in the country.
More recently, comparative studies into learner emotions during in-person versus online EFL classes throughout the pandemic (Dewaele et al., Reference Dewaele, Albakistani and Kamal Ahmed2024; Resnik et al., Reference Resnik, Dewaele and Knechtelsdorfer2023a, Reference Resnik, Dewaele and Knechtelsdorfer2023b) revealed that positive and negative learner emotions seem to be dulled in emergency remote teaching (ERT), possibly due to a lower emotional resonance of online classes, which is partly linked to less interaction with the teacher and peers, as well as fewer opportunities to form social bonds. The well-established moderate negative correlation between FLE and FLCA for in-person FL classes (e.g. Dewaele & MacIntyre, Reference Dewaele and MacIntyre2014) was absent in ERT, meaning FLE and FLCA were no longer significantly linked in the online mode. This suggests that FL learner emotions seem to function differently in different contexts (Resnik & Dewaele, Reference Resnik and Dewaele2023). Insights gained from in-depth interviews with learners have furthermore revealed that the sources of FLE and FLCA differed in both contexts as they were, in the case of ERT, related to the online mode as such (Resnik et al., Reference Resnik, Dewaele and Knechtelsdorfer2023a, Reference Resnik, Dewaele and Knechtelsdorfer2023b).
The advantages of outdoor education in comparison to staying indoors throughout the pandemic have also been explored by means of an intervention study in Norway spanning over six weeks (Myhre et al., Reference Myhre, Dewaele, Fiskum and Holand2023). The quantitative analysis of data gathered with a survey revealed no significant differences between the FLCA and FLE experienced by the 67 teenagers who were in the experimental group and received FL instruction outdoors and the 39 students in the control group who received the FL instruction in school. However, qualitative insights revealed several benefits of outdoor FL education, such as reduced pressure to perform, which was linked to lower FLA, fun, and a relaxed atmosphere, which is likely to allow students to thrive.
A context that has received increased attention since the outbreak of the COVID-19 pandemic, is computer-assisted language learning (CALL) and the role of affective variables in this context. Pre-pandemic CALL research mostly examined the negative emotion of anxiety, mirroring the original trend observed for in-person FL classes (Kruk & Pawlak, Reference Kruk and Pawlak2022). However, the role positive emotions and negative emotions other than anxiety play is still under-researched in this context (for an overview of studies on CALL, see Kruk & Pawlak, Reference Kruk and Pawlak2022) and so is their role in informal language learning settings (see Kruk and Pawlak [Reference Kruk and Pawlak2022] for pioneering work on, e.g. the experience of FLE, FLCA, FL boredom, and curiosity in informal FL learning in virtual worlds). The era of digitalisation offers numerous possibilities in this respect, and learners often take advantage of these digital worlds to enhance their language learning. However, the vast majority of studies on learner emotions so far have investigated them inside the FL classroom – be it online or on site. Future research on the role of affective variables in language learning outside the classroom walls, especially in informal digital learning, is much needed to gain a deeper understanding of these settings and to better understand their potential benefits (Li, Wei, & Jiang, Reference Li, Wei and Jiang2024). Research by Lee and Lee (Reference Lee and Lee2021), for example, has already pointed to a positive link between engagement with informal digital learning of English and learners’ FLE among Korean high school and middle school students. Research into eTandems, where two language learners mutually support each other in their FL learning via online tools by being both expert (meaning in this case an L1 user of the language their partner is learning) and FL learner (of the partner’s L1) (Stickler & Lewis, Reference Stickler, Lewis, Hurd and Lewis2008), revealed similar trends among the participating tertiary-level students from the UK and Austria, who collaborated on FL learning activities on a voluntary basis in their free time (Resnik & Schallmoser, Reference Resnik and Schallmoser2019). Research task 6 provides concrete ideas for much needed future research in this respect.
Research task 6: Explore the emotional experiences of learners in out-of-class digital FL learning and their relationship to learners’ perceived and actual proficiency in the different skills.
An exemplar for measuring learner emotions in online informal FL learning would be the abovementioned research by Kruk and Pawlak (Reference Kruk and Pawlak2022) on the experience of enjoyment, anxiety, boredom, and curiosity among tertiary-level students in Poland who majored in English. These learners were asked to use the virtual environment Second Life (SL) informally in their free time throughout the pandemic, which granted them autonomy in terms of when to learn as well as what to learn. The students were instructed how to use the virtual world (e.g. how to create an avatar, how to move around in SL, and how to communicate), and they were offered some pre-designed language activities, including casual chats with other users in the FL or writing an article on SL users’ perception of the quality of life in the virtual world, which involved interviewing the residents. Additionally, the learners were encouraged to design their own FL learning activities as they used the virtual world informally. Data were collected using a variety of instruments, including a background questionnaire, a survey on the measured learner emotions, as well as an in-World session log, in which the students provided a description of each session with reference to the emotions concerned and filled out an emotion grid (Kruk & Pawlak, Reference Kruk and Pawlak2022, pp. 54–55). The data provided rich insights into learners’ emotional experiences, the sources underlying them, and their dynamic development. Future research avenues to explore here would be the impact FL learners’ emotional experiences in such environments have on learners’ perceived and actual achievements (possibly in the different skills), depending on, for example, the extent of engagement with gaming. Such research on informal FL learning could be of cross-sectional nature and include data collection from a larger sample (see, e.g. Lee & Lee, Reference Lee and Lee2021) or involve a longitudinal focus. Developing specific scales for specific activities would be a valuable approach to researching emotions in informal environments (see also Kruk & Pawlak, Reference Kruk and Pawlak2022), too, as it is likely their sources are to a certain extent unique, meaning they could not be captured fully with scales which were originally developed for instructed settings. Hence, studies in which the decision is made to draw on or modify pre-existing scales should thus at least include a qualitative aspect to uncover aspects which would otherwise be hidden or lost.
It should, however, be noted that the example above is just one of the many contexts which deserve to be explored in greater detail in the future. People have many possibilities to engage with FLs online in their free time and they do tend to do so in many different ways (e.g. learning apps such as Duolingo, video platforms such as YouTube and Twitch, online gaming communities, and social media platforms such as Reddit, Facebook, and Twitter). In order to be able to reap the benefits offered by such comparatively new opportunities which opened up, future research into the emotional dynamics of such settings is much needed.
3.4. Bridging the research-practice gap
3.4.1. Research task 7: The need for more PP-based intervention studies and practice
Since its inception, PP has been overtly focused on practical interventions to improve the happiness, well-being, and flourishing of people. Early intervention studies in non-SLA contexts include the work of Seligman et al. (Reference Seligman, Steen, Park and Peterson2005), which involved five activities to increase happiness, including keeping diaries with daily entries of three good things that happened during the day (later named the Three Good Things exercise), writing a letter of gratitude to a loved one, and writing a personal essay reflecting on personal strengths. Another early intervention study includes the work of Emmons and McCullough (Reference Emmons and McCullough2003), which involved an experimental study of daily and weekly gratitude reflections that resulted in increased well-being. The early PP interventions laid the foundations of future work to come, with studies focusing on improving well-being through simple, accessible activities. These activities have since taken root in both general education research and SLA research.
A review of evidence-based PP interventions in schools without an explicit focus on FL classrooms found five predominant constructs that interventions are designed to improve, namely: hope, gratitude, mindfulness, resilience, and character strengths (Waters, Reference Waters2011). Interventions framed around increasing hope and hopeful thinking in students resulted in reduced anxiety (Green et al., Reference Green, Grant and Rynsaardt2007) and increased life satisfaction and self-worth (Marques et al., Reference Marques, Lopez and Pais-Ribeiro2011). Gratitude interventions frequently involve counting blessings or the Three Good Things exercise and can result in increased optimism and life satisfaction (e.g. Froh et al., Reference Froh, Sefick and Emmons2008). In turn, mindfulness interventions can involve mindfulness practice (e.g. Peixoto et al., Reference Peixoto, Guedes Gondim and Pereira2022) and mindfulness meditation (e.g. Broderick & Metz, Reference Broderick and Metz2009), leading to positive outcomes such as increased positive emotions, the development of new strategies to cope with stress, and increased feelings of calmness and acceptance. Resilience interventions often include the implementation of resiliency training programs at schools, such as the Healthy Kids Program (Lee et al., Reference Lee, Heberlein, Pyle, Caughlan, Rahaman, Sabin and Kaar2021) and ‘Learning to BREATHE’ (Felver et al., Reference Felver, Clawson, Morton, Brier-Kennedy, Janack and DiFlorio2019). Lastly, character strengths interventions usually focus on the broader grouping of positive traits deemed as character strengths (Peterson, Reference Peterson, Csikszentmihalyi and Csikszentmihalyi2006), where these traits are universally valued and can be seen as contributing to flourishing. Peterson (Reference Peterson, Csikszentmihalyi and Csikszentmihalyi2006) identified 24 character strengths that include traits such as courage, curiosity, creativity, and humility. An example of a character strengths intervention is the Strathaven Positive Psychology Program (Seligman et al., Reference Seligman, Ernst, Gillham, Reivich and Linkins2009), where n = 347 students in language arts classes were taught with a PP emphasized program. Students who participated in the program showed increased character strengths, greater enjoyment, and improved engagement in comparison to the control group.
Although the majority of school-based interventions in PP has been domain-general, some studies within SLA have utilized PP interventions in the language learning classroom. Gregersen et al. (Reference Gregersen, MacIntyre and Meza2014a) employed three PP-based activities in order to improve the emotional experiences and emotional intelligence of participants. The first, Three Good Things, encouraged learners to reflect on three positive experiences each day related to language learning and their reasons for positivity. The second activity, Savoring, asked students to notice and log one positive experience per day, whether related to language learning or life in general, focusing on their feelings and potential for future positive experiences. The third activity, Learned Optimism, guided learners to identify negative experiences, challenge their beliefs about these events, and find counter-evidence to reframe their perceptions. Inspired by Gregersen et al.’s (Reference Gregersen, MacIntyre, Finegan, Talbot and Claman2014b) qualitative study, Li and Xu (Reference Li and Xu2019) conducted a follow-up study that combined correlational and experimental methods rooted in PP theories. They investigated the relationship between trait emotional intelligence, FLE, and FLCA. In the correlational part, questionnaires were administered to 1,718 English learners from three high schools in China, revealing moderate correlations between TEI, FLE, and FLCA. The experimental part used a pre-test, treatment, and post-test design to evaluate a six-week PP-based emotional intelligence intervention. This intervention, applied to 56 students in the experimental group, included the ‘ARGUER’ training model and PP diary activities, while 52 students in the control group did not receive the intervention. Quantitative and qualitative findings demonstrated that the intervention effectively enhanced TEI, increased positive emotions, and reduced negative emotions in the classroom.
Other SLA-specific intervention studies include the work of Jin et al. (Reference Jin, Zhang and MacIntyre2020), who aimed to reduce FLA through increased contract speaking and diary entries in n = 42 FL students. The experimental group showed reduced anxiety and increased engagement and self-efficacy in comparison to the control group. In a second study, Jin et al. (Reference Jin, Dewaele and MacIntyre2021) utilized a reminiscing exercise to reduce anxiety, where the experimental group engaged in reminiscing about FL achievements and in a post-test did have significantly less anxiety than the control group. Furthermore, Abdolrezapour and Ghanbari (Reference Abdolrezapour and Ghanbari2021) conducted an intervention based on hope, gratitude, and emotion regulation and found a positive effect on listening comprehension scores. In turn, Hui et al. (Reference Hui, Chow, Chan and Leung2020) found positive results in a picture book reading study on n = 78 kindergarten children. The children in the experimental group were reading in their FL using positive psychology informed dialogue reading techniques and as a result, showed increased complex syntactic structures and verbal creativity.
Although some headway has been made in introducing PP interventions into FL learning contexts, the scope of these interventions is limited, sample sizes are often small, and the duration of studies is often short (see Aydın & Tekin, Reference Aydın and Tekin2023, for review). Therefore, studies using PP interventions in SLA are needed, with a focus on quasi-experimental designs and longitudinal follow-ups (see Task 7).
Research Task 7: Positive psychology interventions in foreign language classrooms using a quasi-experimental design.
Following the lead from Gregersen et al. (Reference Gregersen, MacIntyre and Meza2014a) and Li & Xu (Reference Li and Xu2019), we encourage researchers to further introduce PP interventions into FL learning classrooms. Researchers may use the review of evidence-based PP interventions in schools by Waters (Reference Waters2011) as a further guide. The effects of several interventions have not yet been examined within the context of FL learning, such as resiliency training (e.g. Lee et al., Reference Lee, Heberlein, Pyle, Caughlan, Rahaman, Sabin and Kaar2021), emotion regulation training (e.g. Raccanello & Hall, Reference Raccanello and Hall2021), and character strengths training (e.g. Seligman et al., Reference Seligman, Ernst, Gillham, Reivich and Linkins2009). Researchers are also further encouraged to design studies to include cognitive, behavioural, and emotional mechanisms that may facilitate well-being, such as emotional intelligence (e.g. Li & Xu, Reference Li and Xu2019), emotion regulation (e.g. Raccanello & Hall, Reference Raccanello and Hall2021), and self-esteem (Marques et al., Reference Marques, Lopez and Pais-Ribeiro2011). Beyond the PP interventions, researchers are also encouraged to expand the nomological network and consider FL learning outcome variables in relation to these interventions, such as willingness to communicate, FL engagement, and FL motivation.
In the design of these studies, researchers can examine in-class and out-of-class FL learning contexts, consider age, gender, and culture effects, and the linguistic background of participants. Given the past criticism regarding study design and generalizability of results of PP research (see Singal, Reference Singal2021), we encourage researchers to strictly adhere to a quasi-experimental design with experimental groups and control groups (i.e. one class is exposed to the intervention, while the other class serves as a control group), pre- and post-tests, and longitudinal follow-ups. Carefully considered study designs will also conform to the ongoing methodological reform in applied linguistics as a whole, where more emphasis is being placed on methodology (see Marsden et al., Reference Marsden, Plonsky, Gudmestad, Edmonds, Gudmestad and Edmonds2018).
3.4.2. Research task 8: The need for exploring a greater variety of emotions as both dependent and independent variables in diverse TBLT contexts
Task-based language teaching (TBLT), originating from the communicative language teaching approach (Long, Reference Long, Hyltenstam and Pienemann1985), prioritizes meaning while not overlooking form (Ellis et al., Reference Ellis, Skehan, Li, Shintani and Lambert2020). It encourages learners to engage in incidental language acquisition for its meaning-oriented focus. In contrast to form-oriented methods (e.g. grammar-translation), which emphasize mechanical memorization of language structures, TBLT is considered more emotionally engaging due to its focus on real life communication, meaningful interactions in authentic and personally relevant contexts, and the potential for greater perceived value and control (Abdi Tabari et al., Reference Abdi Tabari, Khajavy and Goetze2024; Li, Reference Li2025).
This difference between form-oriented methods, such as the grammar-translation method, and TBLT also highlights the potential of integrating TBLT and diverse emotions in both research and practice (Abdi Tabari et al., Reference Abdi Tabari, Khajavy and Goetze2024; Aubrey, Reference Aubrey2025; Li, Reference Li2025). However, research in TBLT has predominantly focused on cognitive factors (e.g. working memory) and negative emotions, such as long-term, task-independent anxiety (Li, Reference Li2025), while neglecting the role of positive emotions, including task-specific enjoyment (Li & Dewaele, Reference Li, Dewaele and Li2024; Li, Wei, & Lu, Reference Li, Wei and Lu2024). This narrow focus risks misrepresenting learners as emotionless and tasks as solely anxiety-inducing (Li & Dewaele, Reference Li, Dewaele and Li2024). To offer a more holistic view, future research could incorporate diverse emotions into TBLT research, emphasizing positive emotional factors like task enjoyment, which is believed to not only enhance task motivation, engagement, and performance, but also improve learner well-being during task performance and language learning in the long run (Abdi Tabari et al., Reference Abdi Tabari, Khajavy and Goetze2024; Aubrey, Reference Aubrey2025; Chang et al., Reference Chang, Zhou and Zhang2024; Li, Reference Li2025; Li, Wei, & Lu, Reference Li, Wei and Lu2024).
Recognizing the need for a broader affective lens, recent TBLT studies have begun exploring beyond anxiety, examining a wider spectrum of emotions, primarily in speaking and writing tasks, including task enjoyment in L2 speaking (Aubrey, Reference Aubrey2025; Chang et al., Reference Chang, Zhou and Zhang2024; Li & Dewaele, Reference Li, Dewaele and Li2024), task enjoyment in L2 writing (Abdi Tabari et al., Reference Abdi Tabari, Khajavy and Goetze2024; Li, Reference Li2025; Li, Wei, & Lu, Reference Li, Wei and Lu2024), and task boredom in L2 writing (Li, Reference Li2025).
Research task 8: Explore a variety of emotions as both dependent and independent variables in diverse TBLT contexts.
To advance task emotion research, we propose three key priorities: (1) investigating a broader range of emotions, including both individual and social dimensions, beyond the current focus on anxiety, enjoyment, and boredom; (2) examining emotions as both independent variables (their influence on learning) and dependent variables (their outcomes in learning); and (3) expanding research beyond L2 production (speaking/writing) to include L2 reception (listening/reading) contexts.
First, current research on task emotions has primarily examined discrete affective states, particularly anxiety, enjoyment, and boredom, within L2 productive skills (speaking and writing). While these studies have established important foundations, significant opportunities exist to expand the emotional spectrum under investigation. Future research should consider incorporating social emotions (e.g. shame, guilt, gratitude, embarrassment) in collaborative task contexts, as well as exploring more complex emotional experiences in receptive skills (listening and reading). For such emerging constructs, rigorous conceptualization and measurement validation remain fundamental prerequisites.
The following two studies exemplify the essential processes of proposing, conceptualizing, measuring novel emotional constructs in tasks, and distinguishing them from existing related constructs. Wang et al. (Reference Wang, Li and East2024) investigated task anxiety in L2 oral performance among Chinese EFL learners completing video narration tasks. Through exploratory factor analysis, they identified four task anxiety sub-facets (language-related difficulties, environment-related anxiety, setting-related anxiety, and physiological symptoms), with language-related difficulties showing the strongest influence. Structural equation modeling confirmed task anxiety as a distinct construct, directly influenced by foreign language anxiety and indirectly by trait anxiety. Similarly, Li and Dewaele (Reference Li, Dewaele and Li2024) developed a 23-item Task Enjoyment Scale for L2 speaking contexts through interviews with Chinese EFL learners. Their subsequent exploratory and confirmatory factor analyses revealed and confirmed three key dimensions (self, task characteristics, and social), measured by the final 10 items of the scale. Correlation analyses then identified task enjoyment as a distinct yet related construct of general foreign language enjoyment.
These studies demonstrate the value of investigating specific emotional constructs in L2 speaking contexts. Future research could extend this line of explorations to other understudied emotions, such as shame and embarrassment as they are reported to be frequently experienced in L2 speaking (Khajavy et al., Reference Khajavy, Smid, Mercer and Murillo-Miranda2025), guilt and love in collaborative tasks (Teimouri, Reference Teimouri2018), and boredom in L2 reading and listening as they are inclined to occur in less interactive L2 contexts (Li et al., Reference Li, Dewaele and Hu2023; Pawlak et al., Reference Pawlak, Zawodniak and Kruk2020a).
Another aspect to note is that current research has predominantly treated task emotions as independent variables that have the potential to impact task performance rather than dependent variables influenced by task environments or other learner-related or teacher-related factors. Future research should explore how different task features affect emotional responses, as demonstrated by two recent studies. Li, Wei, and Lu (Reference Li, Wei and Lu2024) examined 412 Chinese 8th-grade EFL learners and found that increased task complexity significantly enhanced task motivation and enjoyment, leading to improved performance in content and organization (though not language accuracy). Similarly, Abdi Tabari et al. (Reference Abdi Tabari, Khajavy and Goetze2024) investigated task sequencing effects among 75 L2 English students completing nine versions of decision-making tasks over nine weeks. Their results showed that simple-to-complex sequencing reduced anxiety and increased enjoyment while improving writing performance, highlighting sequencing’s crucial role in emotional and cognitive outcomes. These findings underscore the importance of emotional wellbeing in task-based learning and suggest multiple avenues for future research. Building on these studies, researchers could examine how additional task features (e.g. planning, control, creativity, format, time limits), learner factors (e.g. self-control, achievement goals), teacher factors (e.g. enthusiasm), or assessment factors (e.g. feedback systems) influence emotions, followed by intervention studies manipulating these variables based on empirical evidence.
4. Conclusion
The present paper has reported on the quickly growing field of research on both positive and negative emotions in FL classrooms. We argue that this research can expand in different ways and in different directions. There is a need to broaden the spectrum of learner and teacher emotions and to gain a deeper understanding of their complex interactions. Also, in order to avoid the assumption that findings from EFL research are universally applicable, there is an urgent need to diversify research contexts and target languages. Finally, continuous efforts are needed to bridge the research-practice gap, as findings about the causes and effects of learner and teacher emotions can inform pedagogical practice and boost both the performance and wellbeing of FL learners and their teachers.
Authorship and contributorship
PR, EB, and CL participated in the conceptualization. PR, EB, CL, and JMD drafted the manuscript and revised the article. All authors read and approved the final manuscript. PR administered the project.
Competing interests
The authors declare none.
Pia Resnik is Professor of ELT Research and Methodology at the University College of Teacher Education, Vienna|Lower Austria, Austria. Her research interests include all aspects surrounding multi-competent LX users of English, with a particular focus on emotions in multilingual contexts and the psychology of language learning and teaching. She has written and co-authored several publications in these areas, including the book Multilinguals’ Verbalisation and Perception of Emotions published by Multilingual Matters. She is currently serving on the leadership team of the Emotion Special Interest Group and the board of the International Association for the Psychology of Language Learning.
Jean-Marc Dewaele is former president of the International Association of Multilingualism, the European Second Language Association and the International Association for the Psychology of Language Learning. He is General Editor of the Journal of Multilingual and Multicultural Development. He won awards from the British Association for Counselling and Psychotherapy (2013), the International Association of Language and Social Psychology (2016) and the European Second Language Association (2022). He was ranked as the second most influential linguist in the world in the Stanford rankings 2025.
Chengchen Li is a full professor of applied linguistics at the School of Foreign Languages, Shanghai Jiao Tong University, China. Her research interests include positive psychology, L2 writing, task-based language teaching, and bilingualism and cognition. She has published more than 60 articles in high-ranking international journals. She is Co-Editor of the Journal of Multilingual and Multicultural Development and Associate Editor of Studies in Second Language Learning and Teaching. She is Vice President of the International Association for the Psychology of Language Learning. She has been included in Clarivate’s Highly Cited Researcher list (2024–2025) and Stanford/Elsevier Top 2% Scientists list (2022–2025).
Elouise Botes is a post-doctoral researcher in psychology at the Cognitive Science and Assessment Institute at the University of Luxembourg, Luxembourg. She is an inter-disciplinary researcher examining the role of individual differences in language learning, using concepts from both psychology and applied linguistics in her work. She was recently awarded with the Early Career Researcher Award from the International Association for the Psychology of Language Learning and is currently on the editorial board for the Journal of Multilingual and Multicultural Development.