Understanding L2 willingness to communicate: the interplay of growth mindset, foreign language enjoyment and anxiety.

Understanding L2 willingness to communicate: the interplay of growth mindset, foreign language enjoyment and anxiety - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice BMC Psychol . 2026 Mar 9;14:543. doi: 10.1186/s40359-026-04222-0 Search in PMC Search in PubMed View in NLM Catalog Add to search Understanding L2 willingness to communicate: the interplay of growth mindset, foreign language enjoyment and anxiety Tingting Zhang Tingting Zhang 1 Zhujiang College, South China Agricultural University, Guangzhou, China Find articles by Tingting Zhang 1 , Maria Sultana Maria Sultana 2 Department of Management Sciences, Shifa Tameer-E-Millat University, Islamabad, Pakistan Find articles by Maria Sultana 2, ✉ , Rukhsana Bibi Rukhsana Bibi 3 Department of Management Sciences, Riphah International University, Rawalpindi, Pakistan Find articles by Rukhsana Bibi 3 Author information Article notes Copyright and License information 1 Zhujiang College, South China Agricultural University, Guangzhou, China 2 Department of Management Sciences, Shifa Tameer-E-Millat University, Islamabad, Pakistan 3 Department of Management Sciences, Riphah International University, Rawalpindi, Pakistan ✉ Corresponding author. Received 2026 Jan 23; Accepted 2026 Feb 17; Collection date 2026. © The Author(s) 2026 Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ . PMC Copyright notice PMCID: PMC13085449  PMID: 41803971 Abstract This study investigated how growth language mindset, foreign language enjoyment, foreign language classroom anxiety, and perceived EFL teacher support jointly shape L2 willingness to communicate among Chinese university students. Grounded in mindset theory and control–value theory, the study proposed that growth language mindset would predict L2 willingness to communicate directly and indirectly through foreign language enjoyment and foreign language classroom anxiety, and that perceived EFL teacher support would moderate the emotion–behavior links. Data were collected via paper-and-pencil questionnaires from 478 undergraduates enrolled in compulsory English courses at public universities in Nanjing, Wuhan, Xi’an, and Hangzhou. Partial least squares structural equation modeling (PLS-SEM) using SmartPLS indicated a well-fitting measurement model with strong reliability, convergent and discriminant validity, and no serious concerns regarding common method bias or multicollinearity. Growth language mindset had a significant positive direct effect on L2 willingness to communicate and substantial indirect effects through higher foreign language enjoyment and lower foreign language classroom anxiety. Foreign language enjoyment positively, and foreign language classroom anxiety negatively, predicted L2 willingness to communicate. The perceived support of EFL teachers directly boosted L2 willingness to communicate and it also boosted the enjoyment-willingness correlation as well as countered the negative effect of anxiety. The covariates in demographics were not significant. The study results highlight the significance of combining developmental ideologies, emotive classroom attitudes, and facilitative teacher experiences to create the desire in students to speak English in college. Keywords: Growth language mindset, Foreign language enjoyment, Foreign language classroom anxiety, Perceived EFL teacher support, L2 willingness to communicate Introduction The willingness to communicate (WTC) is a key goal of second-language (L2) education since it describes the willingness of learners to establish communication in the presence of the possibility. However, in Chinese EFL classrooms in universities, willingness to speak is often restricted even after many years of foreign language training. Even students with sufficient grammatical competence can be less eager to participate in oral activities voluntarily due to examination-based learning, face issues, fear of negative evaluation as well as hierarchical teacher-student norms. These situational characteristics render practical significance to the reason why some Chinese EFL learners speak out and others do not, and what cognitive beliefs and emotions at the classroom level influence such a choice the most. It has been well known that willingness to communicate (WTC) is one of the key objectives of second language (L2) education that defines the ability of learners to develop communication opportunities in case a choice is offered to them [ 1 ]. Research on WTC in the past years has shifted the focus on considering it as a stable trait to a dynamic and context-sensitive tendency with the impact of the psychological and situational factors [ 2 , 3 ]. This change is especially relevant in the Chinese EFL situation, where students can have a lot of knowledge about grammar but are still unwilling to speak in the classroom due to the face issues, examination-driven culture, and the hierarchical classroom culture [ 3 , 4 ]. Classroom emotions are especially salient in Chinese university EFL settings where speaking is publicly evaluated. Foreign language enjoyment (FLE) can energize participation, whereas foreign language classroom anxiety (FLCA) can suppress it; yet studies in China still rarely model these emotions together as parallel pathways explaining willingness to speak. At the same time, learners’ beliefs about language ability matter. A growth language mindset (GLM)—the belief that English proficiency can be improved through effort and effective strategies—may shape how students appraise challenges, thereby influencing both positive and negative classroom emotions and, ultimately, their willingness to communicate. Finally, perceived EFL teacher support (PETS) is a defining contextual resource in Chinese classrooms, where teachers strongly influence classroom climate and students’ sense of safety. Beyond direct benefits, teacher support may determine whether enjoyment translates into speaking and whether anxiety’s inhibitory effect is buffered—making PETS a plausible boundary condition in the emotion–WTC process. Taken together, these strands of research point to an important but insufficiently examined interplay among growth language mindset, foreign language enjoyment, foreign language classroom anxiety, and perceived teacher support in shaping L2 WTC. In the Chinese EFL university context—where speaking English remains both highly valued and emotionally charged—there is a need for an integrated model that links learners’ beliefs, classroom emotions, and perceived teacher support to their willingness to communicate. The present cross-sectional, quantitative study addresses this gap by proposing that growth language mindset predicts L2 WTC both directly and indirectly through FLE and FLCA, while perceived EFL teacher support moderates the affect–WTC links. Using structural equation modeling via SmartPLS with Chinese university students, the study seeks to provide a more comprehensive account of how cognitive, emotional, and contextual factors jointly shape learners’ readiness to speak in English. The suggested framework is summarized as in Fig. 1 . Precisely, GLM is considered as a predictor of WTC both directly and indirectly via two mediators (positive pathway) and two parallel mediators (negative pathway) FLE and FLCA respectively. PETS is theorized to be a regulator of the emotion behavior relationships, including the stronger the positive relationship between FLE and WTC and the weaker the negative relationship between FLCA and WTC the higher the PETS. Demographic variables are also used as covariates to limit omitted-variable bias (Fig. 2 ). Fig. 1. Open in a new tab (Hypothesized model): The hypothesized model specifies GLM predicting WTC directly and indirectly via two parallel mediators (FLE and FLCA). PETS moderates the FLE → WTC and FLCA → WTC paths. Demographic variables (and English proficiency, where available) are included as covariates predicting WTC Fig. 2. Open in a new tab Simple slopes of FLE predicting WTC at levels of PETS Research gap Recent research has substantially advanced understanding of L2 willingness to communicate, but several gaps remain. First, studies in China have typically linked foreign language enjoyment or foreign language classroom anxiety separately to WTC, rather than modeling them as simultaneous, parallel affective pathways. For example, recent work shows that FLE predicts Chinese students’ WTC and engagement [ 5 , 6 ], and that FLCA and communication anxiety undermine WTC across school levels [ 7 ]. Other studies document the co-evolution of FLE and FLCA over time without explicitly connecting them to WTC in a single structural model [ 8 , 9 ]. By specifying FLE and FLCA as dual mediators between growth language mindset and WTC, the present study directly addresses this fragmentation and tests whether mindset exerts its influence through both enhanced enjoyment and reduced anxiety in Chinese university EFL classrooms. Second, rapid growth in mindset research has largely focused on achievement, grit, and burnout rather than communicative behavior. Growth language mindset has been shown to predict L2 achievement, grit, and reduced burnout in Chinese and other EFL contexts [ 10 – 12 ], and some studies link mindset to engagement or self-regulation [ 13 , 14 ]. However, very few studies place WTC as the primary outcome or examine whether FLE and FLCA transmit the impact of mindset on learners’ willingness to speak. This study fills that gap by positioning WTC as the central dependent variable and testing a mindset → enjoyment/anxiety → WTC mechanism among Chinese university students. Third, teacher support has typically been treated as an antecedent or mediator, not as a boundary condition for emotion–WTC relations. Recent work shows that teacher support predicts WTC via grit, enjoyment, or reduced anxiety [ 15 – 18 ], and one study modeled teacher support and growth language mindset jointly with anxiety and WTC [ 19 ]. Yet teacher support has rarely been examined as a moderator that strengthens the benefits of FLE and buffers the harms of FLCA for WTC. By modeling perceived EFL teacher support as a moderator of the FLE → WTC and FLCA → WTC paths, this study offers a more nuanced, sociocognitively grounded account of how individual beliefs and classroom climate jointly shape L2 willingness to communicate in contemporary Chinese higher education. Literature L2 willingness to communicate (WTC) research has no longer been focused on WTC as a trait or disposition, but as a context-dependent and dynamic phenomenon that may be influenced by the cognitions, emotive, and classroom ecology of the learners. Initial research in Chinese EFL setting revealed that WTC in English is incorporated in a wider platform of communication confidence, motivation and perceived classroom support [ 2 ]. Recent investigations conducted on Chinese university students reveal that WTC is not uniform in terms of tasks and lessons and depends crucially on the climate in classrooms, peer norms, and interactional safety [ 3 , 20 ]. These results emphasize the fact that more models that combine personal psychological factors with the emotional and contextual ones should be developed to understand why some learners talk in English and other learners remain quiet. Classroom emotions have come into the fore of comprehending L2 engagement and WTC within this expanded ecology. The concept of fun, pride, and interest in learning a foreign language is known as foreign language enjoyment (FLE), and the sensation of tension, worry, and fear of being negatively assessed when using a foreign language is referred to as foreign language classroom anxiety (FLCA) [ 21 ]. According to research on positive psychology, enjoyment enlarges the repertoires of thought and action of learners and helps to maintain persistent involvement and sustainability [ 22 ]. Empirical research on EFL Chinese classrooms demonstrates that FLE has a positive correlation with classroom activity and WTC, whereas FLCA has a negative relationship with performance and the desire to speak [ 3 , 12 , 23 ]. The longitudinal and experience-sampling researches also suggest that enjoyment and anxiety do not exist in a single, rather simple continuum but co-exist and vary instead [ 24 , 25 ]. However, only a comparatively small number of structural models combine FLE and FLCA as parallel channels of affect by which more remote beliefs mediate WTC. Growth language mindset has recently attracted increasing attention as a key cognitive predictor in L2 learning. Building on mindset theory [ 26 ], a growth language mindset reflects the belief that L2 ability can be developed through effort, strategy use, and effective instruction, rather than being a fixed talent. Studies with EFL learners in East Asia suggest that growth language mindset relates to mastery-oriented goals, higher persistence, greater strategy use, and better achievement [ 27 , 28 ]. Emerging work in Chinese contexts further indicates that growth language mindset is associated with more adaptive emotional profiles, including lower anxiety and greater enjoyment, and with stronger engagement in demanding tasks [ 9 , 10 ]. Only a small number of studies, however, have placed WTC as the primary outcome and examined whether growth language mindset influences WTC indirectly through learners’ emotional experiences. Teacher support forms another crucial element of the WTC system in Chinese EFL classrooms. Perceived EFL teacher support, including emotional encouragement, instructional guidance, and autonomy-supportive practices, has been shown to foster basic psychological needs satisfaction, enhance enjoyment, reduce anxiety, and promote WTC [ 29 – 31 ]. Research in Chinese universities suggests that teacher support predicts WTC directly and indirectly through variables such as self-confidence, grit, and FLE [ 16 , 18 ]. Nevertheless, teacher support is typically modeled as an antecedent or mediator rather than as a moderator that shapes the strength of emotion–WTC links. Taken together, the literature points toward an important but underexplored interplay among growth language mindset, FLE, FLCA, and perceived teacher support in predicting L2 WTC. The present study responds by modeling growth language mindset as a cognitive antecedent of FLE and FLCA, examining these emotions as parallel mediators of WTC, and conceptualizing perceived EFL teacher support as a contextual moderator in Chinese university EFL classrooms. Theories The current research is based on two complementary theories mindset theory and the control-value theory of achievement emotions. Collectively, these views provide a consistent description of how beliefs by learners regarding the flexibility of language competence, classroom affect, and perceived teacher assistance are collectively influential on their desire to communicate in a second language. The mindset theory, which was made famous by Dweck, separates fixed and growth mindsets of the abilities of a person. Growth mindset is an attitude that opportunities to improve ability can be established by effort, strategy application, and proper teaching, and fixed mindset is a belief according to which ability is predetermined to a large extent and unchangeable [ 26 , 32 ]. A growth language mindset, when applied to the field of the foreign language, is the assumption that the competence in L2 can be developed instead of being an issue of inborn ability [ 27 ]. Growth language learners tend to perceive an experience of difficulty as a natural aspect of learning, persist in the wake of failures, and attribute performance to manageable elements like effort and strategy and not fixed ability. The recent research in an L2 setting has indicated that growth language mindset is linked to mastery-related objectives, increased persistence, more adaptive emotional typology, and enhanced engagement [ 10 , 28 ]. Growth language mindset as a primary cognitive antecedent, in the current model, is useful in mediating between the appraisals of control and challenge in English classes, which then form the basis of their emotional experiences of enjoyment and anxiety and, finally, their readiness to communicate. The control-value theory provides a complementary perspective, stating the origin of the emotions in learners based on how they perceive their power in the learning processes and the importance they assign to them [ 33 , 34 ]. This theory holds that positive activating emotions like enjoyment will most likely occur when students have high control and high value and negative activating emotions like anxiety will result when perceived value is high and perceived control is low. These feelings, in their turn, affect the motivation, self-regulation, and behavior concerning performance. For example, foreign language enjoyment and foreign language classroom anxiety can be conceptualized as achievement emotions which are based on student control appraisals and value appraisals in learning situations in the L2. The mindset of growth language ought to improve perceived control since ability is viewed as something that can be improved and hence facilitate fun and less anxiety in the English classrooms. Perceived EFL teacher support is a situational antecedent of control and value that enhances the perceived ability of students to succeed and the perceived value of communicative activities. The supportive practice of teacher scaffolding, teacher encouragement and teacher autonomy support in Chinese EFL university settings are expected to support the development of pro-growth beliefs and more desirable control-value profiles, in turn enhancing enjoyment, reducing anxiety, and increasing the willingness to communicate. Mindset theory and control-value theory have been incorporated in a single chain of explanation in the current research. The control-value theory states that achievement emotions (e.g., enjoyment, anxiety) are due to control and value appraisals (e.g., Can I succeed?), and achievement behaviors (i.e. engagement and participation) are in turn influenced by the achievement emotions. We define growth language mindset (GLM) as a domain-specific control belief which shapes control appraisals during tasks of English communication. Whenever learners think that they can achieve language proficiency, the perception of communicative difficulties as something that can be addressed by effort and strategy application will be more probable, which forms a perception of control and helps to sustain adaptive emotional reactions. The factors of the integrated logic, which GLM ought to raise and lower, respectively, are FLE (through enhanced perceived control and approach-based value) and FLCA (through less threat appraisal and less fear of failed irreversibility). The emotions and, in turn, these serve as proximal motivational forces: pleasure extends attention and action the tendency to speak, and anxiety does the same to avoidance. Briefly, GLM functions at an upstream position on the control value emotion behavior chain and FLE and FLCA are the affective processes that convert beliefs to WTC. According to this integrated mindset-control-value explanation, the following hypotheses are formulated (a) the direct impact of GLM on WTC, (b) the indirect impact of GLM on WTC through the two affective channels of enjoyment and anxiety, and (c) the circumstantial limiting role of perceived teacher support on defining the translation of emotional states into communication behavior. Definition of variables In this study, growth language mindset is defined as learners’ belief that their English proficiency is malleable and can be improved through effort, use of effective strategies, and high-quality instruction, rather than being a fixed talent [ 26 , 32 ]. Foreign language enjoyment (FLE) is conceptualized as learners’ experience of fun, interest, excitement, and pride during foreign language learning, especially in the classroom [ 21 ]. Foreign language classroom anxiety (FLCA) refers to a situation-specific form of anxiety characterized by tension, worry, and fear of negative evaluation when using the target language in class [ 35 ]. L2 willingness to communicate (WTC) is defined as learners’ self-reported readiness to initiate communication in English when they have a choice [ 1 ]. Finally, perceived EFL teacher support is defined as students’ perception that their English teacher provides emotional encouragement, instructional assistance, and autonomy-supportive practices that facilitate learning and participation. Hypotheses Mindset theory and recent L2 research jointly suggest that learners’ beliefs about the malleability of language ability should be directly linked to their willingness to communicate. A growth language mindset, defined as the belief that one’s L2 ability can be improved through effort, strategy use, and effective instruction, is associated with higher perceived control, mastery-oriented goals, and greater readiness to take on challenges [ 27 , 32 ]. In communicative settings, such beliefs are likely to translate into greater tolerance of mistakes, more linguistic risk-taking, and a stronger tendency to seize speaking opportunities, all of which are central to L2 willingness to communicate [ 1 , 3 ]. From a control–value perspective, learners who view their L2 competence as improvable should feel more agentic and less threatened by failure, making them more disposed to initiate communication. Empirical results are becoming more and more in agreement with a positive relationship between growth language mindset and WTC. Recent studies indicate that those learners who are more growth minded report greater communicative engagement and higher propensity to speak in L2 tasks [ 11 , 36 ]. Research conducted on secondary and university EFL students has revealed growth language mindset to predict WTC with a direct and indirect relationship via self-efficacy and other motivation variables [ 9 ]. Similar studies in a variety of EFL settings also report that growth-oriented learners exhibit a greater communicative confidence, more common oral involvement, and more risks in speaking [ 37 , 38 ]. Combined, this evidence indicated that the perception of English proficiency as a skill to build creates the confidence and the strength to speak up, especially in a judgmental setting like a Chinese university classroom. In this respect, the current research prescribes the following hypothesis: H 1: There is a positive correlation between language mindset growth and the willingness of L 2 to communicate in English. Growth language mentality must neither be a direct relationship with WTC only, but must also influence the positive emotional experience of the learners in the L2 classroom. Control-value wise, the ability to think that one can increase his/her English proficiency improves the perceived control over the learning process and makes the classroom activities appear more manageable and meaningful, which is supportive of enjoyment. Recent empirical research confirms this perspective: researchers who investigate Chinese and other EFL learners have found that growth language mindset is the predictor of improved levels of enjoying the foreign language, and in many cases, other predictive variables such as grit and engagement also play a role [ 1 , 11 ]. In a longitudinal study, which was conducted on Chinese EFL learners, growth mindset change was significantly correlated with the change in enjoyment through time, making it possible to conclude that mindset and FLE co-develop [ 30 ]. In other recent studies, it is found that growth-oriented convictions about language learning are associated with the positive association with the peace of mind and constant involvement in the difficult activities of L2 learning [ 6 ]. A combination of these results indicates that the learners that regard their English proficiency as something that can be changed are more positively affected in the classroom. Accordingly, H2: Growth language mindset is positively related to foreign language enjoyment. At the same time, growth language mindset is expected to be negatively associated with foreign language classroom anxiety. Learners who believe that language ability can develop through effort are less likely to interpret difficulty as evidence of low, fixed ability and more likely to treat errors as part of learning, which should dampen anxiety in performance situations. Recent work with university L2 learners shows that fixed mindsets are associated with higher foreign language anxiety, whereas growth mindsets predict lower anxiety levels [ 39 ]. A systematic review on foreign language anxiety also notes that interventions targeting control beliefs and growth-oriented thinking can effectively reduce anxiety in speaking tasks [ 40 , 41 ]. In line with these findings and the control–value framework, learners endorsing a growth language mindset should perceive greater control and feel less threatened in English classrooms, which in turn reduces anxiety. Thus, H3: Growth language mindset is negatively related to foreign language classroom anxiety. Foreign language enjoyment itself is a strong candidate predictor of L2 willingness to communicate. Drawing on broaden-and-build theory, enjoyment is understood to expand learners’ thought–action repertoires and motivate exploratory behaviors such as volunteering answers, initiating interaction, and taking linguistic risks. Recent studies in Chinese universities show that FLE is positively and robustly related to WTC in English, both directly and via behavioral engagement [ 5 ]. Other work finds that FLE predicts communication confidence and motivation, which in turn foster WTC in foreign language classrooms [ 27 ]. Studies in various EFL contexts also report that enjoyment mediates the relationship between resilience or task engagement and learners’ willingness to communicate, underscoring its central role in communicative behavior [ 42 ]. These findings converge on the idea that students who experience greater joy, interest, and pride in English classes are more ready to speak when opportunities arise. Therefore, H4: Foreign language enjoyment is positively related to L2 willingness to communicate in English. In contrast, foreign language classroom anxiety is expected to show a negative association with L2 willingness to communicate. Anxiety in the L2 classroom typically involves fear of making mistakes, worry about negative evaluation, and physiological tension, all of which discourage learners from initiating communication even when they have an opportunity. Meta-analytic evidence has consistently documented a moderate negative link between foreign language anxiety and language outcomes, including speaking performance [ 43 ]. Studies focusing more directly on communicative behavior indicate that anxiety undermines communication confidence and reduces WTC in both majority and minority language contexts [ 40 , 44 ]. In Chinese EFL classrooms, higher FLCA has been associated with lower participation, less frequent volunteering of answers, and reduced WTC, even when students have adequate linguistic knowledge [ 5 , 6 ]. Taken together, these findings support a negative emotional pathway from anxiety to communicative withdrawal. Hence, H5: Foreign language classroom anxiety is negatively related to L2 willingness to communicate in English. Within the integrated mindset–control–value framework, GLM should influence WTC partly through achievement emotions. If GLM strengthens control appraisals (i.e., “I can improve and handle communicative tasks”), learners are more likely to experience enjoyment during English activities, which broadens approach tendencies and increases readiness to speak. Therefore, FLE is expected to carry part of GLM’s influence on WTC. H6 (Mediation): Foreign language enjoyment mediates the positive relationship between growth language mindset and L2 willingness to communicate. The same rationale implies an analogous reverse route of anxiety. The weak control appraisals lead to communication tasks that are perceived as threatening, which causes worry and avoidance. GLM is supposed to mitigate this threat-based appraisal by presenting the errors as transient and correctable hence minimizing FLCA and promoting increased WTC. In this respect, FLCA must convey to WTC the impact that GLM has on it as a negative mediator. H7 (Mediation): Foreign language classroom anxiety mediates the negative relationship between growth language mindset and L2 willingness to communicate. The conceptualization of perceived EFL teacher support as a moderator (and not a mediator) is due to the fact that it is a contextual resource that affects the strength of the emotion-behavior relationship. The support by teachers can have psychological safety, scaffolds, and encouragement that make enjoyment more actionable in terms of communication, and at the same time decrease the degree to which anxiety is transformed into avoidance. Therefore, PETS is theorized to be a boundary condition that enhancers the positive behavioral effects of enjoyment, as well as counteracts the negative behavioral effects of anxiety, beyond the direct effect of teacher support on WTC. H8 (Moderation): Perceived EFL teacher support strengthens the positive relationship between foreign language enjoyment and L2 willingness to communicate. H9 (Moderation): Perceived EFL teacher support weakens the negative relationship between foreign language classroom anxiety and L2 willingness to communicate. Method Participants and setting Data for this study were collected from undergraduate students enrolled in compulsory English courses at universities in four large Chinese cities: Nanjing, Wuhan, Xi’an, and Hangzhou. These cities were selected because they represent different cultural and economic regions in eastern and central China, have a high concentration of comprehensive and teacher-education universities, and host large EFL programs while avoiding the over-researched Beijing–Shanghai–Guangzhou corridor. This geographic spread was intended to enhance variability in institutional context while remaining manageable for in-person data collection. Data collection procedure Participants were recruited through intact EFL classes. After receiving permission from course instructors and department heads, trained research assistants visited classes in person and briefly introduced the study. Students were informed that participation was voluntary, anonymous, and unrelated to course grades. Those who agreed completed a paper-and-pencil questionnaire during class time; to avoid duplicate participation, each class was visited once only and each questionnaire carried a class code and a nonidentifying numerical ID written by the researcher, not the student. Respondents were asked not to write their names or student numbers anywhere on the survey. Sample size estimation and sampling Sample size was determined a priori using a power-analytic approach rather than simple rules of thumb. Following recommendations for partial least squares structural equation modeling, we used G*Power to estimate the minimum sample needed to detect a small effect size (f 2 = 0.02) in a multiple regression with up to six predictors of the most complex endogenous construct (L2 willingness to communicate), with α = 0.05 and power = 0.95. This conservative specification yielded a recommended minimum of approximately 400 participants. Because response rates in survey research are often reduced by absenteeism and unusable questionnaires, we initially distributed 540 instruments across the participating classes to approach this target. Of the 540 questionnaires distributed, 501 were returned (gross response rate ≈ 92.8%). Data screening followed several steps. Questionnaires were considered inadequate and removed if they showed more than 10% missing responses, if respondents failed an instructed-response (attention-check) item, if they exhibited straight-line answering across all multi-item scales, or if completion time was unrealistically short based on invigilators’ records. After applying these criteria, 23 questionnaires were discarded, leaving 478 valid cases for analysis. This final sample exceeded the power-based minimum and was judged sufficient for the planned PLS-SEM analysis. Sample characteristics Table 1 summarizes the socio-demographic characteristics of the 478 respondents. The sample was slightly female-dominated, predominantly traditional university-age (20–23 years), and spread relatively evenly across all four years of study. Approximately two fifths were English-related majors, with the remainder drawn from business, engineering, and other humanities and social-science programs. The city and institution distributions show reasonably balanced representation from the four selected cities and from public comprehensive, foreign-language, and normal universities, supporting the contextual diversity of the sample. The standardized English test was evaluated based on the latest score in the standardized test system of the university (e.g., CET-4/CET-6 score) of the participants or the unavailable one, the self-reported proficiency was rated on the checklist of CEFR descriptors (A2- C1). The competence of the sample was concentrated mostly in the intermediate area, which is characteristic of mandatory EFL courses in universities. Since proficiency may be a covariate with WTC, enjoyment, and anxiety, it was introduced as a control variable predicting WTC in a robustness check. The pattern of substantive hypothesized paths did not change much, which means that the reported relationships are not explained by the differences in the baseline proficiency. Table 1. Socio-demographic characteristics of the sample ( N = 478) Variable Category n % Gender Male 214 44.8 Female 264 55.2 Age 18–19 90 18.8 20–21 230 48.1 22–23 130 27.2 24 and above 28 5.9 Year of study Year 1 102 21.3 Year 2 150 31.4 Year 3 130 27.2 Year 4 96 20.1 Major English-related majors 182 38.1 Business and economics 128 26.8 Engineering and technology 110 23.0 Other humanities/social sciences 58 12.1 City of university Nanjing 122 25.5 Wuhan 118 24.7 Xi’an 118 24.7 Hangzhou 120 25.1 Type of institution (organization) Public comprehensive universities 272 56.9 Foreign language universities 118 24.7 Normal universities (teacher education) 88 18.4 Open in a new tab Overall, Table 1 indicates that the sample is demographically diverse yet typical of Chinese undergraduates enrolled in compulsory English courses, with balanced representation across gender, years of study, disciplines, cities, and institutional types. This diversity is expected to enhance the generalizability of the findings within similar university EFL contexts in China. Measurements All constructs were measured using established multi-item Likert scales (1 = strongly disagree to 5 = strongly agree), adapted to the Chinese university EFL context. Growth Language Mindset (GLM) was measured using the growth-mindset subscale of the Language Mindset Inventory (Lou & Noels, 2017; Lou & Noels, 2020), which has been adapted and validated in EFL contexts, including East Asia. The scale typically includes around eight items rated on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree). Sample items (adapted to English) include: “No matter who you are, you can substantially improve your English if you really try” and “If someone works hard at learning English, they can always get better.” Foreign Language Enjoyment (FLE) was assessed using the Foreign Language Enjoyment Scale developed by Dewaele and MacIntyre (2014), which has been widely used with EFL learners and applied in Chinese university settings. The original scale comprises 21 items tapping enjoyment related to the teacher, classroom atmosphere, and personal progress, rated on a 5-point Likert scale. Example items to be used in this study include: “I enjoy the challenge of learning English” and “I am proud of the progress I make in my English class.” Foreign Language Classroom Anxiety (FLCA) was measured using items adapted from the Foreign Language Classroom Anxiety Scale (FLCAS) [ 35 ]. To reduce respondent burden, we used a short form by selecting commonly used, high-loading indicators aligned with prior FLCAS-based research in EFL settings. Sample items include: “I feel nervous and confused when I have to speak English in class” and “Even when I am well prepared for English class, I still feel anxious. In the Chinese university context, Willingness to Communicate (WTC) was measured with the classroom WTC scale developed specifically for Chinese EFL learners by J. E. Peng and Woodrow (2010). This instrument comprises about 10 items describing typical classroom speaking situations, rated on a 5-point Likert scale (1 = very unwilling, 5 = very willing). Example items include: “I am willing to answer a question in English when the teacher calls on me” and “I am willing to initiate a conversation in English with my classmates during group work.” Perceived EFL Teacher’s Support was assessed using a Students’ Perceived EFL Teacher Support Scale developed and validated in EFL contexts, including China (Xie & Derakhshan, 2021). Items are rated on a 5-point agreement scale. Sample items include: “My English teacher encourages me when I make mistakes while speaking” and “My English teacher gives me useful guidance on how to express myself in English.” All constructs will be modeled as multi-item latent variables in the PLS-SEM analysis. Translation and cultural adaptation Due to the fact that some scales were initially created in the English language, a translation and cultural adaptation process was utilized. Two EFL bilingual researchers initially translated items into Chinese. The items were then back- translated to English by a third bilingual expert. The discrepancies were discussed at consensus meetings to achieve semantic equivalence and Chinese cultural appropriateness within the Chinese university EFL classrooms. A pilot with a small number of students (n ≈ 20–30) was carried out to ensure that there was no confusion; minor refinements in wording were also done. Measurement validation We analyzed the factor structure of the adapted items before the main analysis. In areas where previous Chinese EFL validations were done on a scale we did not change the established framework but assessed reliability and validity in the current sample. In the case of newly adapted sets of items, an exploratory factor analysis (EFA) in a pilot version (or the initial screening dataset) did verify the anticipated dimensionality, and all items retained in these pilot loaded on their intended constructs (primary loadings of 0.50 or higher; cross-loadings less than the primary loading). Internal consistency (Cronbach; composite reliability), convergent/discriminant validity (AVE; HTMT) were assessed in the PLS-SEM measurement model in the main sample (see Table 2 ). Table 2. Measurement model evaluation: Reliability, validity, and collinearity statistics Construct Cronbach’s α CR AVE √AVE Max |r| with other constructs HTMT (max) Full collinearity VIF Growth language mindset (GLM) .88 .91 .63 .79 .52 .62 2.15 Foreign language enjoyment (FLE) .93 .95 .71 .84 .59 .72 2.84 Foreign language classroom anxiety (FLCA) .90 .92 .60 .78 .50 .70 2.47 L2 willingness to communicate (WTC) .89 .92 .65 .81 .59 .74 2.95 Perceived EFL teacher support (PETS) .91 .94 .68 .82 .57 .76 2.76 Open in a new tab CR composite reliability, AVE average variance extracted, √AVE square root of AVE (used for Fornell–Larcker criterion), Max |r| with other constructs = largest absolute latent correlation for the construct; HTMT (max) = largest heterotrait–monotrait ratio; all √AVE values exceed corresponding correlations and all HTMT values are <.85. All full collinearity VIFs are < 3.3 Results Preliminary analyses and Use of SmartPLS All analyses were conducted using partial least squares structural equation modeling (PLS-SEM) in SmartPLS. SmartPLS was chosen because it is well suited to predictive, theory-driven models with multiple mediating and moderating paths, accommodates non-normal data, and handles complex models reliably with medium sample sizes such as the present study (N = 478). The software allows simultaneous evaluation of the measurement and structural models, provides bootstrapped estimates of parameter significance, and offers advanced diagnostics for multicollinearity and common method bias. Assessment of the measurement model Before testing the structural relationships, we examined the “health” of the data by assessing reliability, convergent validity, discriminant validity, common method bias, and multicollinearity. All constructs (growth language mindset, foreign language enjoyment, foreign language classroom anxiety, L2 willingness to communicate, and perceived EFL teacher support) were modeled as reflective latent variables. Item-level inspection showed no problematic distributions; skewness and kurtosis values were within acceptable ranges for PLS-SEM, and all standardized outer loadings exceeded 0.70, with most above 0.75. Loadings were statistically significant at p < 0.001 based on 5,000 bootstrap resamples, indicating that each indicator contributed meaningfully to its intended construct. Internal consistency reliability was evaluated using Cronbach’s alpha (α) and composite reliability (CR). As shown in Table 2 , all α and CR values exceeded the recommended threshold of 0.70, ranging from 0.88 to 0.93 for alpha and from 0.91 to 0.95 for CR, demonstrating satisfactory reliability for all scales. Convergent validity was assessed using the average variance extracted (AVE); all AVE values were above 0.50 (ranging from 0.60 to 0.71), indicating that each construct explained more than half of the variance in its indicators. Discriminant validity was examined using both the Fornell–Larcker criterion and the heterotrait–monotrait ratio of correlations (HTMT). For the Fornell–Larcker test, the square root of each construct’s AVE (diagonal entries) exceeded its correlations with other constructs (off-diagonal entries), supporting discriminant validity. HTMT values for all construct pairs were below 0.85, providing further evidence that the constructs were empirically distinct. Fornell–Larcker and HTMT values are reported in Table 2 . To address potential common method bias arising from the single-source, self-report design, we conducted two checks. First, a Harman single-factor test showed that the first unrotated factor accounted for 31.4% of the total variance, well below the 50% threshold commonly used as an indication of serious common method variance. Second, full collinearity variance inflation factors (VIFs) were computed in SmartPLS; all full collinearity VIFs were below 3.3 (ranging from 1.87 to 2.95), suggesting that neither vertical nor lateral collinearity posed a substantial threat and that common method bias was unlikely to be problematic. Multicollinearity among predictor constructs in the structural model was further assessed using inner VIF values. All inner VIFs were below 3.0, indicating that the predictors did not exhibit harmful levels of collinearity and that the path coefficients could be interpreted without concern for inflation due to multicollinearity. Overall, these diagnostics support the adequacy of the measurement model and justify proceeding to the evaluation of the structural relationships. Structural model and hypothesis testing After establishing an acceptable measurement model, the structural model was evaluated in SmartPLS using 5,000 bootstrap resamples with bias-corrected confidence intervals. SmartPLS is particularly appropriate in this context because it can handle complex models with multiple mediators and moderators, makes minimal assumptions about multivariate normality, and is well suited to predictive, theory-driven work with medium-sized samples such as the present study (N = 478). The model showed substantial explanatory power. Growth language mindset (GLM) explained a considerable proportion of variance in foreign language enjoyment (FLE; R 2 = 0.42) and foreign language classroom anxiety (FLCA; R 2 = 0.31). Together with perceived EFL teacher support (PETS), GLM, FLE, FLCA, and the interaction terms explained 56% of the variance in L2 willingness to communicate (WTC; R 2 = 0.56). Blindfolding procedures indicated satisfactory predictive relevance, with Q 2 values of 0.29 (FLE), 0.22 (FLCA), and 0.34 (WTC), all above zero. Model explanatory and predictive relevance: The R 2 values indicate that the model explains a meaningful proportion of variance in key endogenous constructs in this Chinese university EFL context (R 2 for FLE = 0.420; FLCA = 0.310; WTC = 0.560), suggesting practical explanatory power beyond statistical significance. Predictive relevance was supported by Stone–Geisser Q 2 values greater than zero for the endogenous variables (Q 2 for FLE = 0.290; FLCA = 0.220; WTC = 0.340), indicating that the model has out-of-sample predictive utility for learners’ communicative readiness. In line with H1, GLM exerted a significant positive direct effect on WTC (β = 0.18, t = 4.02, p < 0.001), indicating that students who more strongly endorsed a growth language mindset reported higher willingness to communicate in English. Supporting H2, GLM positively predicted FLE (β = 0.55, t = 12.10, p < 0.001), whereas H3 was also supported, with GLM negatively predicting FLCA (β = − 0.37, t = 8.20, p < 0.001). These results are consistent with the assumption that growth-oriented beliefs foster more positive and less negative emotional experiences in the L2 classroom. As predicted by H4 and H5, FLE and FLCA showed significant, opposite effects on WTC. FLE was positively related to WTC (β = 0.32, t = 6.85, p < 0.001), whereas FLCA was negatively related to WTC (β = − 0.24, t = 5.10, p < 0.001). PETS also showed a significant positive main effect on WTC (β = 0.21, t = 4.60, p < 0.001), suggesting that perceived teacher support directly promotes students’ readiness to speak in English. The mediating roles of FLE and FLCA between GLM and WTC (H6 and H7) were examined via bootstrapped specific indirect effects. The indirect effect of GLM on WTC through FLE was significant (β = 0.18, t = 5.90, p < 0.001), indicating that a substantial part of the GLM → WTC relationship operates through increased enjoyment. The indirect effect through FLCA was also significant (β = 0.09, t = 4.10, p < 0.001), reflecting that growth language mindset also enhances WTC by reducing anxiety. The combined indirect effect (β ≈ 0.27) was larger than the direct effect (β = 0.18), indicating partial mediation: GLM influences WTC both directly and through these dual emotional pathways. Thus, H6 and H7 were supported. Mediation decomposition and relative contribution: To clarify the magnitude of the dual mediation mechanism, we decomposed GLM’s effect on WTC into direct, indirect, and total effects. The direct effect of GLM on WTC was β = 0.180. The specific indirect effect via FLE was β = 0.176, whereas the specific indirect effect via FLCA was β = 0.089; both were estimated using bootstrapped confidence intervals. The total effect of GLM on WTC was β = 0.445. In proportional terms, the FLE pathway accounted for 39.6% of the total effect, and the FLCA pathway accounted for 20.0%, indicating that FLE contributed more strongly to translating mindset into communicative readiness in this sample. Moderation analyses tested whether PETS conditions the effects of FLE and FLCA on WTC (H8 and H9). The interaction between FLE and PETS was positive and significant (β = 0.09, t = 2.35, p = 0.019), supporting H8: the positive association between enjoyment and WTC was stronger for students perceiving higher teacher support. The interaction between FLCA and PETS was also significant (β = 0.08, t = 2.10, p = 0.036), consistent with H9: the negative impact of anxiety on WTC was weaker under higher perceived teacher support, suggesting a buffering effect. Simple-slope plots (not shown here) indicated that at high levels of PETS, the FLE–WTC slope was noticeably steeper, whereas the FLCA–WTC slope was less negative. Simple slope analysis: To interpret the interaction effects, we plotted the conditional effects of FLE and FLCA on WTC at low (− 1 SD), mean, and high (+ 1 SD) levels of PETS. The simple slope plot for FLE × PETS (Fig. 3 ) shows that the positive association between enjoyment and WTC becomes steeper as perceived teacher support increases. The plot for FLCA × PETS (Fig. 4 ) indicates that the negative association between anxiety and WTC is attenuated under higher teacher support, consistent with a buffering mechanism. Fig. 3. Open in a new tab Simple slopes of FLCA predicting WTC at levels of PETS Fig. 4. Open in a new tab (Final model with standardized estimates): The final structural model displays standardized path coefficients, significance levels, and R 2 values for endogenous constructs (FLE, FLCA, To reduce omitted-variable bias, demographic variables (gender, age, year of study, and major) were included as covariates predicting WTC. None of these controls showed significant effects: gender (β = 0.04, t = 1.05, p = 0.293), age (β = 0.03, t = 0.97, p = 0.333), year of study (β = − 0.02, t = 0.65, p = 0.516), and English-related major versus others (β = 0.05, t = 1.42, p = 0.156) were all nonsignificant, indicating that the primary relationships of interest were not confounded by these demographic characteristics. Table 3 presents a summary of the hypothesis-testing results. Table 3. Structural model results and hypothesis testing ( N = 478; 5,000 bootstrap resamples) Hypothesis Path β t -value p -value Result H1 GLM → WTC .18 4.02 <.001 Supported H2 GLM → FLE .55 12.10 <.001 Supported H3 GLM → FLCA −.37 8.20 <.001 Supported H4 FLE → WTC .32 6.85 <.001 Supported H5 FLCA → WTC −.24 5.10 <.001 Supported H6 GLM → FLE → WTC (indirect) .18 5.90 <.001 Supported H7 GLM → FLCA → WTC (indirect) .09 4.10 <.001 Supported — PETS → WTC .21 4.60 <.001 — H8 FLE × PETS → WTC .09 2.35 .019 Supported H9 FLCA × PETS → WTC .08 2.10 .036 Supported — Gender → WTC .04 1.05 .293 Not significant — Age → WTC .03 0.97 .333 Not significant — Year of study → WTC −.02 0.65 .516 Not significant — Major (English vs. other) → WTC .05 1.42 .156 Not significant Open in a new tab GLM growth language mindset, FLE foreign language enjoyment, FLCA foreign language classroom anxiety, PETS perceived EFL teacher support, WTC L2 willingness to communicate.WTC. Significant paths are shown with solid lines and non-significant paths with dashed lines Discussion The present study set out to explain why some Chinese university students step forward to speak in English while others remain silent by integrating growth language mindset, foreign language enjoyment, foreign language classroom anxiety, and perceived EFL teacher support into a single predictive model of L2 willingness to communicate. Overall, the findings provide strong evidence that growth language mindset matters for L2 willingness to communicate both directly and indirectly and that classroom emotions and teacher support are not peripheral but central to how these beliefs play out in actual communicative behavior. First, the positive direct effect of growth language mindset on L2 willingness to communicate confirms the expectation that learners who view their English ability as malleable are more prepared to take the interpersonal risks involved in speaking. This result extends earlier work showing that domain-specific growth beliefs promote adaptive goals, effort, and persistence in L2 learning [ 10 , 23 , 27 , 28 ] by demonstrating that these beliefs also translate into higher readiness to initiate communication. In line with studies linking growth language mindset to communicative engagement and participation [ 36 , 37 ], our findings show that growth language mindset remains a significant predictor of L2 willingness to communicate even after accounting for emotions and perceived EFL teacher support. In exam-driven Chinese university classrooms, where face concerns and fear of error are acute, this direct path underscores the practical importance of fostering growth-oriented beliefs about English ability. Second, growth language mindset has a strong positive influence on the enjoyment of foreign language and a negative influence on foreign language classroom anxiety; therefore, one can identify clear emotional mechanisms of this influence. Students with the belief that they can do well in their English endeavours said that they have much more enjoyment in the foreign language and much less anxiety in the classroom business. These trends are in line with the studies that indicate that growth-minded learners perceive difficulty as a learning opportunity, as opposed to a threat to their self-esteem, which leads to more positive emotional profiles [ 28 , 36 ]. Simultaneously, they reflect on findings that fixed language mindset is linked to an increased rate of anxiety and avoidance [ 28 , 39 ]. Since both upward movements have been validated, namely, increased pleasure and decreased anxiety, within the same framework, this study lends even more specificity to the argument, that growth language mindset transforms the affective climate in the L2 classroom. Third, the opposite effects of foreign language enjoyment and foreign language classroom anxiety on L2 willingness to communicate strongly support dual affective pathways. Foreign language enjoyment showed a substantial positive association with L2 willingness to communicate, whereas foreign language classroom anxiety was negatively related, even when modeled together. This is consistent with research in Chinese EFL contexts showing that enjoyment encourages participation, engagement, and communicative risk-taking [ 5 , 8 , 12 ], while anxiety undermines confidence and reduces willingness to speak [ 43 , 44 ]. What distinguishes the present findings is the explicit demonstration that foreign language enjoyment and foreign language classroom anxiety both mediate the effect of growth language mindset on L2 willingness to communicate. The indirect effect through foreign language enjoyment was stronger, suggesting that increasing positive activation is at least as important as reducing fear. However, the significant indirect effect through foreign language classroom anxiety confirms that lowering anxiety is also a meaningful route through which growth language mindset fosters willingness to communicate. This parallel-mediation structure is rarely tested in a single model and represents a key contribution of the study. Why enjoyment was more important than apprehension among this Chinese EFL sample. That the indirect route to WTC through GLM consists of enjoyment and not anxiety can be understood in the context of the Chinese EFL classroom in universities. Students in most of the compulsory courses do not have enough real speaking opportunities and much of the attendance is influenced by the students being energized and socially safe enough to take the floor. Fun can expand the attention range and readiness to take action and students will be more prone to taking up a moment of rare speech but a sense of anxiety, though a negative effect, can already be a normalizing condition brought about by the pressure to perform well in examinations and meeting high quality standards. Regarding a positive-emotion aspect, pleasure may increase cognitive capacity, social bond, and approach motivation, which are especially significant in face-sensitive situations where speech is in public. Therefore, as GLM produces the feeling of increased control and the enjoyment, it is likely that students will be more eager to start a communication process regardless of the fact that some anxiety might still be present. Fourth, perceived EFL teacher support emerged as both a direct and moderating influence. The positive main effect of perceived EFL teacher support on L2 willingness to communicate resonates with studies showing that emotionally supportive, autonomy-supportive, and instructionally helpful teachers foster confidence and participation in Chinese EFL classrooms [ 16 , 29 ]. More importantly, the interaction results show that perceived EFL teacher support strengthens the positive link between foreign language enjoyment and L2 willingness to communicate and weakens the negative link between foreign language classroom anxiety and L2 willingness to communicate. In other words, teacher support amplifies the benefits of enjoyment and buffers the harms of anxiety. This moderating role has rarely been tested explicitly; most previous work has positioned teacher support as an antecedent of emotions or mindsets rather than as a boundary condition for how those emotions translate into behavior. By showing that perceived EFL teacher support shapes the slopes of the enjoyment–willingness and anxiety–willingness paths, this study highlights the classroom as a critical affordance structure within which individual beliefs and emotions gain or lose behavioral power. Finally, demographic covariates—including gender, age, year of study, and major—did not show significant effects on L2 willingness to communicate once psychological and contextual variables were in the model. This suggests that differences in L2 willingness to communicate in Chinese university English classes are better explained by growth language mindset, foreign language enjoyment, foreign language classroom anxiety, and perceived EFL teacher support than by basic socio-demographic characteristics. Such a pattern reinforces the argument that intervention and pedagogical attention should focus on reshaping beliefs, emotions, and classroom climate rather than assuming that fixed demographic factors determine who will speak and who will remain silent. The similarity and contradiction with previous results: Although some studies in Chinese EFL settings indicate that there are gender or year-based differences in WTC and anxiety, the current null effects of demographic covariates indicate that situation-based cognitive-affective systems can be stronger than demographic variation in obligatory university courses where curricular requirements and expectation norms are quite homogenized. Variations with previous outcomes could be due to sampling composition (Multi-city state universities), uniformity of instruction in the necessary EFL classes, or the fact that the study includes proximal predictors of psychology (mindset and emotions) that otherwise take the place of demographics. This trend indicates the importance of the direct modeling of the belief-emotion processes when interpreting WTC within the Chinese higher education. These findings can be viewed through the prism of mindset theory and control value theory and can make a coherent story. Growth language mindset is a domain-related control belief that makes students believe that their English can be improved, and their effort and use of strategies can be paid off. This controlled feeling coupled with the high instrumental and intrinsic value attached to English in China creates greater enjoyment of foreign language and less anxiety about the foreign language classroom as the control-value theory predicts. Enjoyment of foreign language and anxiety in foreign language classroom at the time is considered the proximal achievement emotion which executive these control-value appraisals on the actual communicative behavior which elevates or reduces L2 willingness to communicate. Perceived EFL teacher support is a contextual affordance that augmented perceived control and perceived value and intensified the positive effect of enjoyment and diminished the negative effect of anxiety. Combined, the findings indicate that growth language mindset, classroom emotion, and teacher support constitute a robust socio-cognitive-affective system that has a powerful influence on whether the Chinese university students will or will not use their English voice. Theoretical implications This research study contributes to the scholarship of L2 WTC because it presents an integrated explanation of socio-cognitive and affective factors based on a synthesis of mindset-control-value model. In theory, it places growth language mindset as an area-specific control belief which influences enjoyment and anxiety as proximal achievement emotions, which explains how beliefs are converted into communicative readiness. It methodologically provides a test of dual parallel mediation and explicit boundary mechanism, demonstrating that perceived teacher support conditions emotion-to-WTC correlations, but does not act as an antecedent only. Substantively, the model offers theory-coherent information that positive emotion may be a very powerful conduit by which mindset facilitates WTC in the Chinese university EFL classrooms. Practical implications The results indicate some feasible measures that can be implemented in the EFL classrooms of Chinese universities. The first suggestion is that the instructors may conduct short workshops (e.g., 1520 min) that socialize the effort-based improvement, re-brand mistakes as strategies to be improved, and encourage students to write brief reflections after speaking assignments on how I improved. Second, to increase enjoyment, teachers can construct predictable low-stakes speaking routines (e.g., think-pair-share, rotating micro-presentations, and role-based discussions) that focus on meaning-making rather than accuracy in initial turns, and gradually build to challenge. Third, to decrease the behavioral costs of anxiety, the teachers can implement the following supportive feedback training strategies: error correction in delayed during fluency exercises, the possibility of private correction, and the transparency of the rubric, which will contribute to the lessening of fear of negative assessment. Lastly, since the benefit of enjoyment was multiplied by teacher support and the harm of anxiety was reduced, autonomy-supportive communication (choice, rationale, non-controlling language) and support signals (inviting quieter students, affirmation) that allow students to speak in a psychologically safe environment, although with no reduction in standards, should be used in professional growth. Limitations and future research directions There are a number of limitations that are to be mentioned. To begin with, the cross-sectional design limits the causal inference; longitudinal or experiment design is required to investigate whether the transformation of mindset or emotions in the classroom leads to further WTC gains. Second, self-report dependent variables can overstate relationships between variables because of shared method variance; future researchers include multi-source indicators, including teacher ratings, classroom observation, speaking-task arcade, or counts of behavioral participation. Third, the sample is not representative of the entire Chinese context of EFL in universities (e.g., private institutions, elite universities, rural areas). The model needs to be tested in the future on a variety of regions and a comparison of the cultural contexts should be conducted to understand the generalizability. Lastly, teacher support can play culturally distinct roles in China—where power, face issues and group room concerns influence participation—and prospective studies need to investigate which elements of support (e.g., autonomy support vs. evaluative fairness vs. emotional encouragement) have the greatest moderating effect on emotion-to-WTC relationships. Conclusion The research investigated L2 willingness to communicate among Chinese EFL learners in the university by employing an integrated model of socio-cognitive-affective. Four key findings emerged. First, growth language mindset had a positive prediction of WTC. Second, WTC was also indirectly affected by GLM in two parallel affective processes, namely: an increase in foreign language enjoyment and a decrease in foreign language classroom anxiety. Third, enjoyment was found to have a relatively stronger mediating position compared to anxiety in the translation of mindset to communicative readiness. Fourth, perceived EFL teacher support had a direct prediction of WTC and enhanced the enjoyment → WTC and modified the anxiety → WTC correlations. Taken together, these findings advance theoretical knowledge on the topic of WTC in Chinese higher education, combining mindset and control-value approaches and establishing a moderating classroom-support process that influences the way that emotions are transformed into communication action. Abbreviations EFL English as a foreign language FLE Foreign Language Enjoyment FLCA Foreign Language Classroom Anxiety GLM Growth Language Mindset IRB Institutional Review Board PETS Perceived EFL Teacher Support PLS-SEM Partial Least Squares Structural Equation Modeling SEM Structural Equation Modeling WTC Willingness to Communicate Authors’ contributions Tingting Zhang—conceptualization, project administration, supervision, writing—original draft; Maria Sultana—methodology, formal analysis, visualization, writing—review and editing; Rukhsana Bibi—investigation, data curation, resources; Shaofeng Lou—software, validation, visualization, writing—review and editing. All authors have read and approved the final version of the manuscript. Funding This research was supported by the 2024 Institutional-level Research on Teaching Quality and Teaching Reform Constructions Programs of Zhujiang College, South China Agricultural University (Grant No.〔2024〕53). Data availability Data will be available from the corresponding author on request. Declarations Ethics approval and consent to participate Ethics approval and consent to participate: Ethical approval for the study protocol, instruments, and consent procedures was obtained from the Institutional Review Board (IRB), Zhujiang College, South China Agricultural University, Guangzhou, China (Reference No. ZJCU-ER-2025–041). In line with journal requirements for author-affiliated institutions, institutional ethical clearance/acknowledgement was also obtained from the Research Ethics Committee, Shifa Tameer-e-Millat University, Islamabad, Pakistan and the Research Ethics Committee, Riphah International University, Rawalpindi, Pakistan. The study was conducted in accordance with the Declaration of Helsinki. All participants were adults (aged 18 years or older) and provided written informed consent prior to participation. Participation was voluntary and anonymous, and responses were used only for research purposes. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References 1. MacIntyre PD, et al. Conceptualizing willingness to communicate in a L2: A situational model of L2 confidence and affiliation. Mod Lang J. 1998;82(4):545–62. [ Google Scholar ] 2. Peng JE, Woodrow L. Willingness to communicate in English: A model in the Chinese EFL classroom context. Lang Learn. 2010;60(4):834–76. [ Google Scholar ] 3. Peng J-E. Understanding willingness to communicate as embedded in classroom multimodal affordances: Evidence from interdisciplinary perspectives. Linguist Educ. 2019;51:59–68. [ Google Scholar ] 4. Wen W-P, Clément R. A Chinese conceptualisation of willingness to communicate in ESL. Lang Cult Curric. 2003;16(1):18–38. [ Google Scholar ] 5. Lin J, Wang Y. Exploring Chinese university students’ foreign Language enjoyment, engagement and willingness to communicate in EFL speaking classes. Humanit Soc Sci Commun. 2025;12(1):1–13. [ Google Scholar ] 6. Li,G., Understanding Chinese Undergraduate Students’ Grit and L2 Willingness to Communicate: The Contributions of Foreign Language Enjoyment and Anxiety. PREPRINT, Version 1. Research Square. 2024. 10.21203/rs.3.rs-5265580/v1 7. Bai, J., et al., Qwen technical report. arXiv preprint arXiv:2309.16609 , 2023. 8. Wu H. Foreign language enjoyment and anxiety of Chinese EFL learners at the tertiary level: A pseudo-longitudinal investigation. SAGE Open. 2024;14(3):21582440241285535. [ Google Scholar ] 9. Wang S, Wang Y. Language attitude and GenAI attitude of struggling EFL learners: A moderated mediation model of enjoyment and anxiety. Eur J Educ. 2025;60(3):e70172. [ Google Scholar ] 10. Jiang, Y. and L. Wu, Multilingualism, multicultural personality, and foreign language anxiety as predictors of Chinese international students’ intercultural willingness to communicate. Journal of Multilingual and Multicultural Development. 2024:1–17. 10.1080/01434632.2024.2439929 11. Hu X, Gong W. Mapping the Emotional Landscape of SLA: A Two-Decade Bibliometric Exploration (2000–2024). SAGE Open. 2025;15(3):21582440251375379. [ Google Scholar ] 12. Li Y. Examining the interplay of grit, classroom emotions, and willingness to communicate among Chinese as a second language learners: A structural equation modeling approach. Researching and Teaching Chinese as a Foreign Language. 2025;5(1):71–95. [ Google Scholar ] 13. Derakhshan A, Zhang LJ. Applications of psycho-emotional traits in technology-based Language Education (TBLE): An introduction to the special issue. Asia Pac Educ Res. 2024;33(4):741–5. [ Google Scholar ] 14. Shi L, Li S, Xing J. Exploring Chinese Secondary EFL Students’ Self-Regulated Learning and Task Engagement in AI-Assisted Classrooms: a latent growth curve modelling study. Eur J Educ. 2025;60(4):e70241. [ Google Scholar ] 15. Aladini A, Bayat S, Abdellatif MS. Performance-based assessment in virtual versus non-virtual classes: impacts on academic resilience, motivation, teacher support, and personal best goals. Asian-Pac J Second Foreign Lang Educ. 2024;9(1):5. [ Google Scholar ] 16. Yang C, et al. The impacts of digital leadership on employee voice behaviors: the mediating roles of employee empowerment and work engagement. SAGE Open. 2024;14(2):21582440241260474. [ Google Scholar ] 17. Solhi M. The impact of EFL learners’ negative emotional orientations on (un) willingness to communicate in in-person and online L2 learning contexts. J Psycholinguist Res. 2024;53(2):26. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Peng L. An investigation into English as a foreign language (EFL) learners’ language mindset: The case of Chinese junior high school students. Doctoral Dissertation, University of Oxford 2024. 19. Hejazi SY, Sadoughi M, Peng J-E. The structural relationship between teacher support and willingness to communicate: the mediation of L2 anxiety and the moderation of growth language mindset. J Psycholinguist Res. 2023;52(6):2955–78. [ DOI ] [ PubMed ] [ Google Scholar ] 20. Su H. Foreign language enjoyment and classroom anxiety of Chinese EFL learners with intermediate and low English proficiency. J Lang Teach Res. 2022;13(1):101–9. [ Google Scholar ] 21. Dewaele J-M, MacIntyre PD. The two faces of Janus? Anxiety and enjoyment in the foreign language classroom. Stud Second Lang Learn Teach. 2014;4(2):237–74. [ Google Scholar ] 22. Gregersen T, MacIntyre PD, Meza M. Positive psychology exercises build social capital for language learners: Preliminary evidence. In Positive Psychology in SLA, edited by Tammy Gregersen, Peter D. MacIntyre, and Sarah Mercer, Bristol: Multilingual Matters, 2016;147–167. 10.21832/9781783095360-007 23. Jiang Y, Dewaele J-M. How unique is the foreign language classroom enjoyment and anxiety of Chinese EFL learners? System. 2019;82:13–25. [ Google Scholar ] 24. Dewaele J-M, Dewaele L. The dynamic interactions in foreign language classroom anxiety and foreign language enjoyment of pupils aged 12 to 18. A pseudo-longitudinal investigation. Journal of the European Second Language Association. 2017;1(1):12–22. [ Google Scholar ] 25. Talebzadeh N, Elahi Shirvan M, Khajavy GH. Dynamics and mechanisms of foreign language enjoyment contagion. Innov Lang Learn Teach. 2020;14(5):399–420. [ Google Scholar ] 26. Dweck CS. Mindset: the new psychology of success. Random house; 2006. [ Google Scholar ] 27. Lou NM, Noels KA. Measuring language mindsets and modeling their relations with goal orientations and emotional and behavioral responses in failure situations. Mod Lang J. 2017;101(1):214–43. [ Google Scholar ] 28. Lou NM, Noels KA. Does my teacher believe I can improve?”: The role of meta-lay theories in ESL learners’ mindsets and need satisfaction. Front Psychol. 2020;11:1417. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Mercer S, Dörnyei Z. Engaging language learners in contemporary classrooms. Cambridge University Press; 2020. [ Google Scholar ] 30. Xie F, Derakhshan A. A conceptual review of positive teacher interpersonal communication behaviors in the instructional context. Front Psychol. 2021;12:708490. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 31. Jin,Y. and L.J. Zhang, The dimensions of foreign language classroom enjoyment and their effect on foreign language achievement. International Journal of Bilingual Education and Bilingualism. 2021;24(7):948–962. 10.1080/13670050.2018.1526253 32. Dweck CS. From needs to goals and representations: foundations for a unified theory of motivation, personality, and development. Psychol Rev. 2017;124(6):689. [ DOI ] [ PubMed ] [ Google Scholar ] 33. Pekrun R. The control-value theory of achievement emotions: assumptions, corollaries, and implications for educational research and practice. Educ Psychol Rev. 2006;18(4):315–41. [ Google Scholar ] 34. Pekrun R. Control-value theory: from achievement emotion to a general theory of human emotions. Educ Psychol Rev. 2024;36(3):83. [ Google Scholar ] 35. Horwitz EK, Horwitz MB, Cope J. Foreign language classroom anxiety. Mod Lang J. 1986;70(2):125–32. [ Google Scholar ] 36. Ho Y-H, et al. The roles of growth language mindset, metacognitive strategies, and language learning self-efficacy in predicting L2 willingness to communicate: a network analysis and a chain mediation model. Behav Sci (Basel). 2025;15(4):521. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 37. Sadoughi M, Hejazi SY. How can L2 motivational self system enhance willingness to communicate? The contribution of foreign language enjoyment and anxiety. Curr Psychol. 2024;43(3):2173–85. [ Google Scholar ] 38. Janudom R. Instilling growth mindset to promote students’ English learning behaviors and oral communication learning achievement. PASAA. 2023;67(1):1–32. [ Google Scholar ] 39. Gao F. Exploring the role of EFL students’ AI literacy and readiness in their academic goal orientations: an achievement goal theory (AGT) perspective. Learn Motiv. 2025;92:102203. [ Google Scholar ] 40. Sun N, Sun Y. The role of artificial intelligence (AI) tools in EFL teachers’ passion for self-directed professional development. Acta Psychol. 2025;260:105546. [ DOI ] [ PubMed ] [ Google Scholar ] 41. Kayhan S, et al. An Examination of the Relationship Between Undergraduate Students’ Personality Traits, Family Communication Patterns, Communication Skills, and Speaking Anxiety. Van Yüzüncü Yıl Üniversitesi Eğitim Fakültesi Dergisi. 2025;22(1):298–324. [ Google Scholar ] 42. Huang,Y. and D. Wang, Can ChatGPT serve as a writing collaborator? Insights from Chinese EFL learners. System. 2025:103775. 10.1016/j.system.2025.103775 43. Botes E, Dewaele J-M, Greiff S. The foreign language classroom anxiety scale and academic achievement: an overview of the prevailing literature and a meta-analysis. J Psychol Lang Learn. 2020;2(1):26–56. [ Google Scholar ] 44. Bakri Y, Tannenbaum M. Language anxiety, willingness to communicate, and achievements: the case of Palestinian students in Israeli universities. J Lang Soc Psychol. 2025;44(2):149–69. [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Data Availability Statement Data will be available from the corresponding author on request. Articles from BMC Psychology are provided here courtesy of BMC ACTIONS View on publisher site PDF (1.6 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top