Motivational and emotional states in self-directed language learning: a longitudinal study

The current longitudinal case study investigated emotional and motivational states in a self-directed and mobile-assisted language learning environment. The participant of the study was a highly motivated language learner who used the Busuu application for a period of one year to learn German. Tracing the participant’s emotional and motivational journey using solicited diaries and chat-based interviews, the study revealed that the process of self-directed language learning is associated with the experience of a variety of positive and negative emotions. Joy, hope, and interest were the prevailing positive emotions that were linked to learning vocabulary; and frustration, anger, and feeling annoyed were the negative emotions in relation to learning grammar and pronunciation. In general, the study found a strong connection between positive emotions and persistence in self-directed language learning. The findings also pointed to some possible causes of high dropouts in application based language learning, as lack of enough resources for learning some language skills (such as grammar) initiated and implicated extensive processes of negative emotions. The study contributes to the growing body of knowledge in mobile-assisted language learning, and calls for further research regarding the role of emotions in technology-based and self-directed learning environments.


Introduction
In recent years, the phenomenal growth in digital and communication technologies have transformed the face of language learning/teaching both inside and outside the classroom [1][2][3][4][5][6], and for a good number of language learners, the classroom is no longer the only environment for developing the knowledge of new languages [7].With increased connectivity to the Internet, more people around the world are using their smartphones and applications such as Duolingo, Rosetta Stone, Mondly, and Busuu for learning new languages [8][9][10][11][12].These environments provide learners with numerous opportunities to undertake learning activities anytime and anywhere, using different devices such as personal computers, tablets and smartphones [13][14][15][16][17].It has been argued that a considerable number of language learners rely on these applications as the primary source of language instruction [18].However, the dominant focus of much second language acquisition (SLA) research has remained on classroom-based learning [19], and the field has just recently developed an interest towards out-of-the-classroom and informal language learning conditions [2,[20][21][22].Knowing about the affordances of these new and emerging language learning environments is significant, as it helps us to better understand learners' behaviors and performances in relation to them [23]. 1 3 Moreover, while the majority of self-study language training programs are marketed as complete language learning solutions, our knowledge regarding their effectiveness and conditions under which they should be used are still limited and incomplete [10,24,25].In this regard, Heift and Chapelle [26] argued that "[t]he need exists to better understand the new conditions for second language acquisition (SLA) brought about by the real language-related capabilities of technologies that many learners have access to on a daily basis" (p.565).Furthermore, much research investigating the role of various technologies in language learning was concerned with describing the technologies themselves, paying scant attention to the impacts of technology use and associated affordances on language development and providing learners with optimal language learning opportunities [26][27][28][29].Previous research on distance education in general indicated that successful course completion by adult learners depends mostly on their ability to manage motivational and affective factors [30], and learners in such programs are expected to set personal goals, select learning trajectories, monitor their own learning and maintain their motivation [31].Nevertheless, studies investigating learners' motivation in informal and self-access language learning settings remained largely limited [32,33].To address these gaps in the literature, this study explored the emotional and motivational states in technology-assisted language learning, with a particular focus on self-study and informal language learning environments provided by mobile devices.
2 Literature review

Academic emotions
Academic emotions refer to the feelings that students experience in relation to their academic experiences, such as anxiety, enjoyment, boredom, or pride [34].Over the last few decades, researchers have increasingly recognized the importance of academic emotions in shaping students' motivation, learning, and achievement [35].One influential theory in this area is the control-value theory of achievement emotions [36,37].This theory proposes that academic emotions are influenced by two key factors: the perceived control that students have over their academic outcomes, and the perceived value or importance of those outcomes [38].For example, students may experience more anxiety if they feel that their performance on a test is high-stakes and beyond their control.Another important concept in the literature on academic emotions is emotional regulation, which refers to the strategies that individuals use to manage their emotions [35].A study by Pekrun et al. [39] found that students who were better at regulating their emotions tended to experience more positive academic emotions and better academic outcomes.In terms of the impact of academic emotions on student outcomes, a recent systematic review found that positive emotions such as enjoyment and pride were positively associated with academic achievement, while negative emotions such as anxiety and boredom were negatively associated with achievement [40].However, the relationship between emotions and achievement is complex and can depend on various factors, such as the type of task or subject [41].Accordingly, with the growing popularity of various forms of technology-enhanced language learning, there is a need to investigated the emotional and motivational landscapes of these emerging instructional environments.

Motivation in self-directed language learning
There is no universal agreement regarding the meaning of out-of-the-classroom and independent learning [42].In the field of language education, the concept is used synonymously with terms such as autonomous, self-directed, selfregulated [43], informal [7], and self-paced language learning [8].The current study uses 'self-directed language learning' as an umbrella term to capture the intentionality and awareness [3] involved in using the mobile applications such as Busuu for independent and out-of-the-classroom language learning.In this regard, self-directed learning reflects a paradigm shift from teacher-centered pedagogy to a student-centered one [44], and allows learners to take full responsibility for, and have more control over their learning trajectories and opportunities.It also requires learners to identify their learning needs, find their preferred ways of learning and reflect on their own progress, which is a way towards fostering autonomous learners [45].
Self-directed learning has been regarded as a method for organizing instruction and learning processes [46,47], with both cognitive and motivational aspects [48].Garrison [48] integrated cognitive, motivational, and social dimensions into a comprehensive theoretical model for self-directed learning, and believes that motivation is crucial for the initiation and maintenance of the self-directed learning.Moreover, motivation and persistence loom large in using applications for self-directed and out-of-the-classroom language learning with mobile devices [49].A recent study found that 73 out of 1 3 115 university students who used Duolingo as a supplement for their foreign language studies logged into the app for only 10 or fewer days over a year, being 'try-and-quit' users of the application [50].In a much larger study investigating 4000 users of Busuu, it was found that only a quarter of the participants used the application for at least six months, and long-term users were more likely to perceive improvements in their language abilities compared to short term visitors [18].The lack of persistence and severe dropouts in using similar applications was also highlighted in an earlier study focusing on the use of Rosetta Stone and Auralog's TELL ME MORE among the U.S. government employees [24].
Moreover, it has been shown that access to mobile devices and having positive attitudes towards using them for educational purposes do not necessarily guarantee the use of language learning applications [51,52].Moreover, according to Pishghadam et al. [53], the passive dimension of motivation can significantly impact individuals' willingness to be involved or not involved in various learning activities.Accordingly, learners' intrinsic motivation for language learning has been regarded as a crucial factor that interacts with the motivational dimension of using applications [10,54].It has been argued that mobile learning environments in general present particular issues related to motivational aspects of language learning that needs systematic attention [49].Nonetheless, our understanding of motivational potentials of mobile technologies for language learning has remained very limited, and there is a need for further empirical research focusing on such important factors [55].

Emotions and motivation interaction in SLA
Emotions have drawn a lot of attention from a variety of disciplines [56], however, attempts for creating a unified definition or a universal list of them have been less than successful [57].In fact, it has been argued that different theorists and researchers have used emotions in ways which imply different meanings, processes, and functions [58].From a traditional perspective, emotions have been defined as relatively short bodily responses compared to long lasting states such as moods [59].Within this paradigm, emotions are viewed as "short-lived, feeling-purposive-expressive-bodily responses that help us adapt to the opportunities and challenges we face during important life events" [60, p. 340].Nevertheless, recent studies have shown that people continue to feel and experience an emotion when they think about the past events [61].Indeed, in most current perspectives, emotions are considered as embodied feelings associated with action urges and responses that get individuals to change their situations or thoughts, rather than just feeling states [62].In this regard, it is now well established that students experience a variety emotions in the educational environments, ranging from anger to enjoyment [63,64], which further relate to a host of other variables influencing their motivations to learn [65].
Nevertheless, over the much history of the SLA, emotions have been given a secondary role in psychological accounts of the learner behavior, and the prevailing belief was that they constitute transient states that do not qualify as trait-like (and stable) individual differences factors [66].One notable exception is research on language anxiety which attracted considerable attention over the past decades, and developed its own research agendas [67].In this regard, there remained a gap in theorizing L2 motivation when it comes to the role of emotions and their connection to the key processes of motivation [67,68].As an example taken from the currently dominant theory in the field, the concept of possible selves in L2 Motivational Self System [69,70] refers to a vision of the self in future state including ideas of what a person might become, what they would like to become, and what they are afraid of becoming [71].If being proficient in target language is part of a learner's ideal or ought-to selves, this will be a very strong motivator, which might lead to arousal of some psychological desires with the aim of reducing the gap (or distance) between the current and future possible selves [70].However, motivational drives of L2 possible selves might be simply represented as cold cognitive goals [72], and cannot be fully appreciated without taking into account the role of emotional states that arise in relation to future self-guides [73,74].Recognizing this fundamental role of affective factors in language learning [75], and also because of the introduction of perspectives taken from positive psychology to the SLA research [76][77][78], the field is increasingly growing an interest into the role of emotions in language learning and teaching [79][80][81].
Moreover, given the multifaceted nature of emotions in terms of bodily and behavioral responses, feelings, and selfrelated appraisals, a growing number of studies in SLA are pointing the motivational potential of both negative and positive emotions [67,68,[82][83][84][85]. MacIntyre and Vincze [68] investigated the relationship among 19 positive and negative emotions, and well-established motivational constructs such as L2 Motivational Self System [69,70].The study found consistently strong correlations between positive emotions and motivation-related variables.For negative emotions, the study found less consistent and weaker relationships with motivational constructs.The findings of the study further highlighted the intricate relationship between emotions and the L2 motivation processes.In a recent study, MacIntyre et al. [82] examined the role of positive and negative emotions in relation to the long established construct of integrative motivation [86].The results showed a strong relationship between emotions and attitudes toward the learning situation, and in line with MacIntyre and Vincze [68], this study also found a stronger correlation with respect to positive emotions.Additionally, Pishghadam et al. [87] investigated the role of emotions in EFL learning.The authors highlighted the significance of emotions in learning, particularly in language learning, but argue that emotions associated with English language skills in listening, speaking, reading, and writing have not received adequate attention.The study involved interviews with 20 students to elicit their views on the emotions they experienced in EFL classes.Based on the findings, the EFL Skills Emotions Questionnaire was constructed, containing 20 items.A total of 308 students then completed the questionnaire, and confirmatory factor analysis was used to validate the scale.The results indicated that EFL learners experience different emotions associated with different language skills.For instance, anger is mostly related to listening skills, enjoyment and pride to speaking, shame to listening and speaking, hope, boredom, and hopelessness to writing and listening, and anxiety to all language skills.Within this emerging line of inquiry, research has explored the impact of specific emotions on learners' attitudes and motivation.For instance, Dewaele et al. [83] investigated the correlation between foreign language enjoyment (FLE) and foreign language classroom anxiety (FLCA), in conjunction with various internal and teacher/classroom-specific factors.The study found that greater FLE scores were linked to more positive attitudes towards learning foreign languages, while lower FLCA scores were also associated with positive attitudes.Another study by Dewaele and MacIntyre [84] employed a large-scale web-based survey to investigate FLE and FLCA across global classrooms.Results indicated that participants reported significantly higher levels of FLE than FLCA, and that both emotions were associated with various independent variables, such as participants' perceived proficiency levels and the number of known languages.Finally, Boudreau et al. [85] used an idiodynamic methodology to investigate the rapidly changing relationship between enjoyment and anxiety when using a second language on a moment-to-moment basis.The study revealed that there is a dynamic association between these two emotions.Furthermore, the study found that experiencing high levels of enjoyment was linked to increased interest and engagement with the task, resulting in better language acquisition.This expanding body of research suggests that emotions will play a crucial role in future investigations into L2 learning motivation [67].

Emotions in MALL
Mobile-Assisted Language Learning (MALL) provides learners with portable, authentic, social, and contextual environments [88], which further promotes inherent interests in the use of technology for language learning [54].However, since there is no external structure to shape the learning experience, it is both easy to start and easy to quit language learning with mobile applications [10].Consequently, learners' attitudes, beliefs, and affective feelings play an important role in the engagement with technology for self-directed language learning [89,90].Despite growing interest into the motivational dimension of self-directed learning with applications in MALL, the study of learner feelings, anxiety, and other emotional states that impact learning remained largely neglected in the existing literature [91,92].
In one of the studies which investigated the learners' affective responses in using mobile applications for learning English, it was found that participating in collaborative mini dialogues with a mobile application reduced speaking anxiety among the participants [93].Yet another study reported that a grammar-focused mobile application promoted positive affects like enjoyment, but did not significantly change negative emotions such as anxiety, boredom, and frustration in a Spanish course [92].In a recent study, Kohnke [94] explored perceptions towards mobile assisted vocabulary learning in an English-medium university Hong Kong.The participants were 14 university students, and the study collected data via semi structured interviews.The findings indicated that most students had positive views for using a game-based mobile application for vocabulary learning.More specifically, the study showed that the participants regarded their learning experience fun, engaging, and more motivating.However, the findings also pointed to some factors such as small screen size in mobile-assisted learning environment that resulted in negative emotions and stress.

The present study
As language learning in mobile and self-directed learning environments implicates new processes and cognitive demands on the learners, various emotions are likely to arise in such contexts.In this regard, there is a growing need to conduct studies to identify particular emotions linked to learning different language skills and sub-skills.Insights from such research contribute to our enhanced understanding with respect to the links between learners' motivation, emotions, and engagements with mobile applications designed for self-directed language learning.The current study contributed to this line of research, and aimed to investigate the motivational and emotional states in self-directed and mobile assisted learning of German using Busuu application over a period of one year.The findings add to our understanding with respect to emotional side of mobile assisted language learning, and has implications for materials development for distance and self-access language learning programs too.To this end, the following research questions were addressed: 1-What emotions are experienced in self-directed and mobile-assisted language learning with Busuu?2-What emotions are associated with different language skills in self-directed language learning?3-How do emotions impact motivation in self-directed language learning?3 Method

Participant
The participant of this study was Ramin (pseudonym), who was a 26-year-old native speaker of Farsi and a student of civil engineering at a university in Tehran, Iran.He worked part-time as an English language teacher at two private language schools in Tehran and had achieved proficiency in English, as evidenced by his IELTS band score of 7.5.Ramin aimed to become a competent and well-educated engineer in the future with the intention of finding employment in Western countries.He had been studying English for the past 10 years and, despite having no academic or educational background in English language teaching, was recognized as a highly successful language teacher at the private schools where he worked.At the time of the study, Ramin taught students of varying levels of proficiency in English.The study focused on this participant for the following reasons.First, given that the language used in the application for teaching German was English, we needed a participant with acceptable proficiency level in English which was the medium of instruction in this case.Second, as research on emotions inevitably leads to talking about personal feelings, having a close relationship and establishing a trusting environment among the researchers and the participant was essential for obtaining detailed accounts with respect to emotional states that might be considered sensitive data.Additionally, considering the longitudinal design of the study, we focused on one participant only by collecting detailed data during the research period.We have been acquainted with Ramin for three years before August 2019 when we discovered that he intended to learn German using the Busuu app on his Android smartphone.Due to our interest in exploring the motivational factors related to mobile-assisted language learning beyond traditional classroom settings, we invited Ramin to participate in a research study that aimed to investigate his language learning experience and motivation over a relatively long period.To ensure ethical compliance, we obtained Ramin's informed consent, explained the nature of the study, and guaranteed the confidentiality of any data collected through various means.

Busuu application
Being available in free and paid versions, Busuu application provides language learners with a variety of affordances for self-directed language learning.The learning environment is accessible via internet connection on a desktop computer, however, mobile users can install the application on their Android or iOS devices.The free version of the Busuu provides only a limited number of options for learning (with a focus on vocabulary).Most of the learning features are available in the paid version that include grammar lessons, Artificial Intelligence (AI) based vocabulary review, and offline use mode.After choosing a language to learn, the application asks learners about their goals for learning and guides them through steps for selecting their current level.Accordingly, they might start from beginner or upper levels.In order to keep learners engaged, the application provides them with personalized learning reminders for mini lessons (between 10 to 15 min per day).Busuu also gives learners some cultural tips and insights regarding the use of the target language appropriately.Finally, Busuu application has a social community in which learners might add friends and even connect with native speakers of the target language.
The study focused on Busuu application since it was the application selected by Ramin as he had a premium access to it for one year.The participant of the study started learning German from beginner level (A1) and his initial goal was spending at least 15 min every day for completing lessons.However, it should be noted that during a period of one year, his motivation and engagement with the application and self-directed language learning went through ups and downs with periods of intense motivational effort and times of declined motivations.By the end of the study, Ramin successfully finished three levels (B1) and earned his certificates for finishing this levels by completing built-in tests in the application (which are based on CEFR levels and are offered by McGraw-Hill education). 1 3

Procedures
The current case study examined the motivational and emotional states in self-directed learning of German as a foreign language over a period of one year (from August 2019) using diary method [95]- [97].In this method, the knowledge creation is a joint process between the researcher and the participants.The diary method provides insights into the participants' various experiences, views, and feelings [96].Most commonly, the participants are required to produce written diaries with some guidelines, although it is possible to use audio and photography as well [98].As self-directed and mobile assisted language learning takes place in different places and times [18,50], the use of diaries was intended to capture the temporal dimension [99] of the learning experience.To this end, the participant provided short answers for the following questions after completing lessons in the Busuu app: 1-When did you use the app to learn German?2-Describe any feelings (joy, anger, frustration, shame …) 3-Mention any strategies employed (such as note taking, using flash cards, …) 4-Mention places where you used the app.5-Mention the use of any extra resources or materials such as dictionaries, internet, … 6-Which language skills are easy to learn by Busuu?Which skills are difficult?Why?
A locally popular social media (i.e.Telegram) was used for writing diaries.In this regard, the participant was asked to write a diary after completing a lesson in Busuu, and send it to the researchers.The social media platform facilitated the research process in different ways.First, the researchers were able to observe any changes in the participant's learning behavior (for example increased or declined motivational states) and ask him to write about those fluctuations.Second, the platform was used to get more information regarding the emotional states of the participant, by conducting chatbased qualitative interviews [100].This helped in complementing the data obtained through solicited diaries by more in-depth understanding.And finally, the platform was used to send soft reminders to the participant to write diaries.As the study was longitudinal and lasted for a year, using the social media as explained above resulted in obtaining rich and detailed data regarding the motivational and emotional states in self-directed language learning.

Data analysis
The obtained diaries and conducted chats with the participant were saved into Microsoft word documents for further analysis.Over the course of a year, the participant produced more than 250 short diaries (8992 words), and 10 chat-based interviews were conducted.The study utilized Nvivo software (version 10) to analyze the data through an iterative process of coding and recoding, as described by Saldana [101].In the initial cycle, the in vivo, emotion, and holistic coding methods were employed to segment each interview or diary entry according to the various aspects of the learners' motivation and learning experience.During the subsequent cycle, the evolved codes from the first cycle were further refined through a process of recoding, categorizing, and discarding, while using cross-referencing across interview transcripts and diary reports.More specifically, the following procedures were followed in coding data in relation to research questions.To answer this first question, the study utilized emotion coding methods to identify and segment emotions expressed by the participant in their diaries and chat-based interviews.The emotions were then categorized into distinct categories and subcategories, such as positive and negative emotions.To answer the second question, the study utilized in vivo and emotion coding methods to segment interview or diary entries according to the various aspects of the learner's motivation and learning experience.The study then compared the emotions associated with different language skills, such as speaking, listening, reading, and writing, to determine the emotional impact of each skill on the learner's motivation and learning experience.As for the third question, the study utilized holistic and pattern coding methods to group related codes into categories to describe the learner's experiences.The study then analyzed the relationship between emotions and motivation, such as how positive or negative emotions influenced the learner's motivation to engage in self-directed language learning with Busuu.The study also looked at how emotions influenced the learner's self-efficacy and perceived competence, which, in turn, impacted their motivation to continue learning.
To ensure the credibility and rigor of the study's findings, several measures were taken to mitigate potential sources of bias and increase the trustworthiness of the interpretations.Firstly, the researchers engaged in member checking, a method in which the interpretations and findings are shared with the participants to ensure that the data is accurately and fairly represented.This process allowed the participant to provide feedback and corrections, if necessary, which helped to increase the credibility and transferability of the study's findings.Secondly, the researchers sought external review of their methods and findings by asking two colleagues familiar with qualitative research methods to review the transcriptions, coding process, and findings.This external review provided an additional layer of quality control and help in identify potential biases or inconsistencies in the data analysis process.Together, these measures helped to ensure that the study's findings are grounded in the data, accurately represent the perspectives and experiences of the participants, and are credible and trustworthy to the broader research community.

Findings and discussion
RQ1: What emotions are experienced in self-directed and mobile-assisted language learning?
The analysis of data obtained via solicited diaries and interviews revealed that self-directed language learning was associated with the experience of both positive and negative emotions.The results of emotion coding are provided in Table 1.The most frequent positive emotion mentioned by the participant was joy (88 times) which was expressed and elaborated on using terms like enjoyment (36 times), feeling good (30 times), and feeling great (22 times).Following joy, two negative emotions namely frustration and anger were mentioned 29 and 20 times respectively.Hope (17), interest (8), fascination (7), love (5) and pride (5) were the other positive emotions reported by the participant.Moreover, annoyed (15), confused (13), feeling bad (10), stressed (9), shame (8), guilt (5), and boring (3) were the reported negative emotions.Overall, the data analysis revealed that in diaries and interviews, the participant mentioned the experience of positive emotions 130 times, and negative emotions 112 times.
In line with previous studies that investigated the presence of negative and positive emotions in language learning [83][84][85], the current study also found a higher level of positive emotions in self-directed language learning.Although the present study had only one participants and the findings might not be comparable to the earlier studies that investigated large number of language learners, given the longitudinal nature of this research and the wealth of data collected over a year, such observations provide insights into the emotional side of self-directed language learning with mobile devices.In this regard, we believe that the relative stronger amount of positive emotions with respect to negative emotions contributed to the sustained effort and longer engagement with the Busuu application for the participant of the current study.The motivating potential for positive emotions is also highlighted in the research on emotion [62].For example, the 'broaden-and-build' theory posits that experiences of fleeting positive emotions have long lasting consequences on individuals' growth and social connection and transform people for better lives in the future.In other words, the experience of positive emotions broadens a persons' momentary mindset, which leads to building enduring personal resources [62].Another factor that might have contributed to the long lasting engagement on the part of the learner with the application is the restorative effects of positive emotions [102].Given the demotivating impacts of experiencing negative emotions in language learning [68], the experience of more positive emotions helped the learner in coping with the associated physiological and cognitive effects [102], and contributed to increased persistence in using the application over a long time.Moreover, as the participant of the study had a dream of working in a western country (i.e.Germany), learning German was in line with his ideal self [70], and the observed learning engagement might have resulted from his efforts to reduce the gap between his current condition and his desired vision for the future [103][104][105].Finally, although Busuu application provides learners with a social community as part of the learning experience, the participant of the study did not use this feature over the learning period, and this explains some aspects of his negative experiences such as being annoyed or confused.
Additionally, the results obtained might be interpreted in light of the concept of emotioncy [106].Emotioncy is a concept that refers to the combination of senses that are involved in an individual's experiences, which can evoke emotions and influence their perception and understanding of reality [107].Emotioncy highlights the importance of sensory experience in language learning, as the type and number of senses involved in a task can impact vocabulary acquisition and retention [108].In other words, learners who have more sensory experience with a word or concept will have a higher emotioncy level and may therefore be more likely to remember and understand it [109].The emotions and sensory inputs people receive from their environment also influence their perception of the future and their connections to themselves, others, and the world around them [110].Therefore, incorporating multisensory approaches to language learning can be beneficial in promoting inclusive and effective learning experiences [111].
RQ2: What emotions are associated with different language skills in self-directed language learning?Data analysis revealed that vocabulary, grammar, and pronunciation were associated with the most of the experienced positive and negative emotions in self-directed language learning with Busuu.In this regard, our data indicated that learning German vocabulary had a positive emotional loading by being associated with joy, interest, hope, and pride.The following excerpts from diary reports show how the participant viewed learning German vocabulary as a positive experience: 1. Vocabulary learning is the most interesting aspect of my learning experience.I learn new words, practice them and really enjoy when I learn a new word.

… As I was doing a new lesson after a while, I became exited and completed a series of vocabulary and grammar tasks.
3. … The learning experience of new vocabulary was fun and I enjoyed learning some useful words.4. Now that I have learned about 350 German words, I feel really excited.I also want to increase the number of words to 500 by the end of this month, hopefully.

…. And despite all these problems, I am really proud of myself for doing great on vocabulary items. If I keep using the app, I will be able to learn a lot of German words.
As highlighted by the participant in the first extract (#1), learning vocabulary items generated a sense of achievement that intensified learner engagement through positive emotions.Moreover, as the participant succeeded in learning vocabulary items (#4), he developed more confidence in learning language and set some short-term objectives for learning even more words.Additionally, it seems that positive emotionality resulted from learning vocabulary items acted as powerful driving motive for counterbalancing negative emotions triggered by other aspects of application-based language learning (#5).
In contrast, learning German grammar and pronunciation invoked a variety of negative emotions including anger, frustration, stress, shame, confusion, and being annoyed: 6.Even after practicing by myself, I thought that my pronunciation is not good at all and that was annoying.I think I need a safe and quiet place to practice my German pronunciation and when there are other people around me, I skip pronunciation exercises.

When a lesson is about vocabulary items or expressions, I really enjoy learning and want to continue!!! but when
the lesson is about grammar, most of the time it is frustrating.8.As usual the learning vocabulary was fun and joyful, but German grammar and sometimes spelling makes me crazy.9.I really fear to say some words!!! they are sometimes really difficult to even pronounce and writing them correctly is another problem.

Today, I completed a new lesson at home in about 20 min on the topic of bar, restaurants and grammar of giving instructions. Despite the fact that yesterday was full of joy and hope (because of the Gboard), today was full of disappointment and frustration. For most of the time, I was unable to understand what is going on in the lesson because of lots of new and confusing rules, and also difficult grammar. 11. Added to this frustration was grammar!!! All the way home (I think it took about 20 min to complete the last grammar review) I had anger, and I was really frustrated and disappointed. I made the same mistakes again and again.
Based on the above extracts from diaries, German pronunciation in particular was associated with feeling anxiety and fear as the participant mentioned skipping pronunciation tasks in the presence of others (#6).Data also pointed to a consistent dissatisfaction (and hence a range of negative emotions) connected to learning grammar.In most cases, the feeling of frustration was resulted from Ramin's inability to understand underlying grammatical rules presented in the Busuu lessons (#10).Relatedly, failure in completing grammar exercises negatively impacted the participant's sense of competence (#8) and provoked anger towards the learning material (i.e.Busuu application) and the participant himself (#11).In this regard, considering the complexity of learning the grammar of a new language, it seems that the mechanical and repetitive learning approach of the Busuu might not be able to develop the learners' grammatical competence effectively [112].
Moreover, the participant talked about the feelings of shame while doing speaking exercises in the presence of others: 12. When I am using the app at home, I skip the speaking exercises a lot.I think the reason is that I don't want to look stupid while trying to say German words out load and not being able to pronounce them correctly.The sad point is that nobody at our home knows German, but I don't want to practice it in front of others!
The above extract points to a different aspect of the participant's learning experience and related emotions.Given that the application provided some opportunities for practicing speaking, the feeling of shame impacted utilizing this important feature in the learning process.The participant clearly indicated that he mostly skipped such exercises even when his family members were around him and associated speaking German in the safe environment of his home as looking stupid.This feeling might have resulted from a number of factors.First, unlike classroom environment where learners are able to practice the sounds and words of a language with others, application-based language learning was a self-directed endeavor for Ramin.Accordingly, he had no support from peers or teachers which might have added to his insecurity with pronunciation and associated feeling of shame.Additionally, German language learning is much less popular in the study context compared to learning English.In this way, speaking German might sound odd and unusual even among family members.
According to MacIntyre et al. [67], research on the role of emotions in language learning motivation has a great potential to uncover the links between particular learning activities and specific emotions.In this regard, our data indicated that vocabulary learning continued to be associated with positive emotions over the research period (a year).The findings also revealed the affordances provided by mobile devices that facilitated the experiences of such positive emotions.For example, throughout the research project, the participant mentioned the experiences of positive emotions such as joy and hope brought about by Google Keyboard (Gboard) to the learning process.Gboard is a virtual keyboard that provides next word suggestion and spelling corrections in writing.Accordingly, Gboard helped Ramin with the spelling of German words (associated previously with negative emotions) and changed the negative emotional loading of spelling into joy and hope.Such affordances then can contribute significantly to self-directed language learning.Moreover, learning German grammar in particular remained linked to negative emotions and only in brief periods was associated with positive emotions [105].Those positive feelings only occurred during intense motivational periods featuring a strong and enduring motivational drive in the pursuit of a highly valued goal or vision which is referred to as Directed Motivational Currents [103,[113][114][115]. Describing this intensive motivational surge is beyond the scope of the current study, but it suffices to mention here that during such periods, the whole experience of learning was associated with positive emotional loading, that resulted in considerable developments and achievements in a short time period [105].This observation further highlights the complex nature of emotional experiences, and their links to specific aspects of language learning.RQ3: How various emotions impact motivation in self-directed language learning?Analyzing the diary and interview data also revealed the link between emotional states and language learning motivation in self-directed learning of German with the Busuu app.In this regard, a general pattern was found in the data that pointed to the impacts of positive emotional states on enhanced motivation, and negative emotional states on declined motivation (although we found some evidence for motivational capacities of some negative emotions such as guilt that will be discussed below).The experience of some positive emotions such as joy and hope was related to the participant's persistence and long-term use of the application: 13.I enjoyed learning new vocab items, and grammar was also easy.However, there was some frustrating points when I was unable to complete a dialog by filling in some vocabulary items.I'm pretty optimistic about the future, and I believe that I'm going to learn German and succeed in achieving my goal.14. ……Given all these difficulties, I reminded myself of the prospect of learning this language.I try to look at positive side.At least I'm going to learn a lot of new German words and phrases.15. …… For some moments I felt that maybe German is not my thing and even thought about giving it up altogether.Then I reminded myself that I know about 300 German words, and I am going to learn even more.This is really something to be happy about.
These extracts highlight the significant importance of negative emotions in sustaining self-directed language learning.First, such positive states contributed to the participant's optimistic view towards learning outcomes and made the distant goal of learning a new language by oneself accessible (#13).Second, positive feelings also resulted in motivational regulation processes by redirecting the learning effort and focusing on what is going well, vocabulary learning in this case (#14 and #15).
The data also indicated that when the participant reported experiencing more negative emotions, this was associated with declined motivation or demotivation for learning German: 16. ……As I progress and learn more words and grammar, it becomes really stressful to continue!There are too many rules to learn, I think I need to slow down for some time and go back to previous lessons or spend more time on grammar.I don't like it when things get out of control.17.Although I remain optimistic and remain committed to my goal of learning German, I must admit that my confidence and motivation have been somewhat shaken by the challenges I encounter when tackling new lessons.The learning process is no longer as enjoyable as it used to be and is instead riddled with frustration, occasional anger, and even confusion.I recognize the need to seek out additional resources, particularly for improving my grasp of grammar, which has proven to be a significant obstacle that threatens to derail my progress.
These accounts provided by the participant point to some of the possible impacts that negative emotions might exert on self-directed language learning.In this regard, although data collected for the current study as discussed above provided some evidence for the significant role of positive emotions on sustaining motivation, it seems that the impacts of negative emotions might be manifested sometimes later in the learning process.Accordingly, it might be argued that the motivating potential of positive emotions become less strong over time (#16).Moreover, the data also pointed to a ceiling effect in the learning process when lack of resources for learning some aspects of language (i.e.grammar) is aligned with more negative emotions (#17).For the participant of the current study, these feelings of negative emotions were linked to perceived inability to move forward with Busuu in learning German.
The obtained data via diaries also pointed to the experience of guilt as a negative emotion in the process of selfdirected language learning which resulted in action urges to resume the leaning.In this regard, this feeling was associated with periods of being inactive in using the application for some times and failing to achieve previously set goals:

Today I was thinking about the past week as I was totally away from my German learning. I don't want to blame anything or anyone but myself. I made some promises past weeks but really failed to keep trying. Tonight I am going to start working.
These accounts given by the participant in diaries and interviews indicated that the experience of negative emotions in self-directed language learning is further aggravated by the lack of resources for learning some aspects of language.Given the fact that learning the grammar of any language needs extensive efforts and ample chances to practice the newly learned rules, it seems that the way self-study applications present the grammar as packed in short lessons is leading to the experience of more negative emotions and hence high dropouts.This observation partially explains the lack of long-term engagements among a large group of 'try-and-quit' users with self-directed language learning applications [18,50].On other side, it is evident from our data that some positive emotions such as joy, hope, and interest can sustain motivation for a long time, and contribute to self-directed language learning in a significant way.Even some negative emotions such as guilt may have motivational impact as it brings the learner back on the learning path.

Conclusion
The study investigated emotional and motivational states in self-directed and mobile-assisted language learning.Tracing the participant's emotional and motivational journey over a year in using Busuu for learning German, this case study revealed that the process was associated with the experience of a variety of positive and negative emotions.Joy, hope, and interest were the prevailing positive emotions linked to learning German vocabulary; and frustration, anger, and feeling annoyed were negative emotions experienced in relation to grammar and pronunciation.The study highlighted the important role of emotions in self-directed language learning, and shed light on their links to enhanced or declined motivational states.In general, the study found a strong connection between positive emotions and persistence in self-directed language learning with mobile applications.The findings also pointed to some causes of high dropouts in application based language learning, as lack of enough resources for learning some language skills (such as grammar) initiated and implicated extensive processes of negative emotions.
The findings have implications for using applications like Busuu for self-directed language learning.First, as data analysis indicated, the experience of language learning with Busuu application for the study participant was associated with a range of both positive and negative emotions.Nevertheless, this emotional climate was mostly positive for learning vocabulary.Considering the importance of vocabulary knowledge in developing other language skills [116][117][118], Busuu application provides learners with a motivating and engaging resource for vocabulary learning, though it should be highlighted that this study did not investigate its effectiveness.Second, the findings of the study indicated that emotional vulnerability might limit the use of some features available in the language learning applications.In this regard, the as the participant investigated in the current study skipped speaking tasks, beside a need for a more emotionally sensitive learning environment in the applications, it should be acknowledged that such applications might not be able to replace the classroom language learning and the emotional support provided in traditional learning environments by peers and teachers.Third, although self-directed language learning with applications like Busuu might be much motivating in initial stages due to motivational impacts of new technologies [119], it seems that such novelty effects are counterbalanced with negative feeling in the long-term.More specifically, our data indicated that lack of resources for learning grammar in particular created a ceiling effect in the learning process that partly explains the high-drop-out rates in application based language learning.Accordingly, there is a need for a systematic and pedagogically sound approaches for dealing with learning grammar in self-directed language learning with mobile applications.
The study had some limitations that should be acknowledged.First, the study traced the emotional and motivational experiences of only one learner.This makes the generalizability of the findings limited.Nevertheless, the study aimed to describe the various positive and negative learning experiences in mobile assisted language learning that can inform the application users and learning content developers and raise their awareness on emotional side of mobile assisted and self-directed language learning.Second, the study investigated one learning environment namely Busuu application.As different language learning applications provide learners with different tasks and learning approaches (such as gamification), there is a need to conduct studies to further investigate learner motivation in those environments too.And finally, given the scope of the current study, we were unable to assess the language development occurred as the result of learning German with Busuu.Accordingly, in order to better appreciate the learning outcomes, there is a need for using standard and validated instrument for tracing language development in application-based language learning.Future studies might consider addressing such issues and investigate the emotional and motivational states in similar learning environments.Research could also evaluate the impact of interventions such as technology-based tools or targeted training programs on learners' emo-sensory intelligence and their ability to self-direct their language learning.