Studies in Second Language Learning and Teaching Another look at boredom in language instruction: The role of the predictable and the unexpected

Although recent years have seen a growing interest in positive emotions in second or foreign language learning and teaching, negative emotions are always present in the classroom and they deserve to be investigated in their own right. The article focuses on boredom, a construct that has been explored in educational psychology but has received only scant attention from second language acquisition researchers. It reports a study which examined the changes in the levels of boredom experienced by 13 English majors in four EFL classes and the factors accounting for such changes. Using data obtained from a few different sources (i.e., boredom grids, narratives, interviews, class evaluations and lesson plans), it was found that although boredom can be attributed to different constellations of factors, it was mainly traced to repetitiveness, monotony and predictability of what transpired during a particular class.

Keywords: boredom; English as a foreign language; L2 instruction; micro-perspective; negative emotions

Introduction
Learning a second or foreign language (L2) is an exceedingly complex process that somewhat inevitably encompasses a range of different affective states or emotions, many of which are negative. A good case in point is anxiety which mostly has a detrimental (debilitative) effect on learners' behavior inside and outside the classroom, their use of the target language (TL), and, as a consequence, the ultimate level of attainment (e.g., Gkonou et al., 2017;Gregersen & MacIntyre, 2015). In view of this, it is perhaps not surprising that research in second language acquisition (SLA), mirroring to some extent the dominant trends in educational psychology, has mainly concerned itself with negative facets of emotions, focusing on their pernicious effect on learning outcomes (see e.g., MacIntyre & Mercer, 2014). However, in recent years we can see a backlash against this tendency, as exemplified in the emphasis on positive psychology, which "rather than taking a palliative approach to reducing pain or coping with distressing experience, . . . seeks to develop tools to build positive emotions, greater engagement, and an appreciation of meaning in life and its activities (Seligman, 2006)" (MacIntyre & Mercer, 2014, p. 154).
Although there is clearly merit to exploring the positive aspects of the process of SLA, whether this occurs within the PERMA (i.e., positive emotions, engagement, relationships, meaning, and accomplishment) framework (cf. Seligman, 2011) or some other paradigm, we should by no means forget about the negative aspects of L2 learning which constitute part and parcel of all learners' educational experience. As Komorowska (2016) illuminates, positive emotions such as high self-esteem can bring with them their own share of problems, whereas negative emotions such as pessimism, fear or experience of difficulty can generate positive states related to, among others, reduced risk-taking, reliance on problem-focused strategies, or more effective functioning. In a similar vein, Dewaele and MacIntyre (2014) argue that while positive and negative emotions are independent and do not constitute two sides of the same coin, both of them are crucial in L2 learning, with anxiety, for example, "generating focus on the need to take specific action" (p. 262). All of this shows that we should avoid pendulum swings of the kind that have characterized L2 teaching methodology (cf. Brown, 2006), and attempt to strike a balance between the investigation of positive and negative emotions in SLA. In accordance with this rationale, this article explores the role that boredom plays in the learning of English by advanced students majoring in this language. Moreover, in line with recent tendencies in research on individual differences (see e.g., Mystkowska-Wiertelak & Pawlak, 2017), it adopts a micro-perspective, in which boredom is examined in specific contexts, also taking into account its dynamic dimension.

The challenge of defining the concept of boredom
Over the past three decades the phenomenon of boredom has been subject to considerable research interest in psychology and educational psychology, but it has remained an ignored and thus unexplored issue in L2 learning and teaching. The main reason for this neglect is that language teachers have a tendency to attribute the behaviors of bored students to laziness, anxiety, general passivity or personal characteristics (e.g., Chapman, 2013;Macklem, 2015). It is a concept that escapes concise, unambiguous explanations due to its multidimensional and situation-dependent character (Gordon et al., 1997). Thus, in order to define boredom, it is necessary to delve into its internal complexity with regard to the factors that generate it and the variables that it interacts with.
Boredom is referred to as an emotional or psychological state associated with an inner sense of emptiness experienced by individuals for whom it is difficult to find purpose and/or meaning in life (Goldberg et al., 2011). Bored students usually feel apathetic, disappointed, uninterested, annoyed and/or dissatisfied with their surroundings which they perceive as unchallenging and tedious, but at the same time they are reluctant to undertake a constructive effort to alleviate this situation (Eastwood et al., 2007). They exhibit impaired vitality, poor concentration skills and difficulty in paying attention (Fahlman, 2009).
Boredom is among the most frequently experienced academic emotions and, as such, it is vividly present in school settings affecting the quality of student learning, achievement and social interactions (Pekrun et al., 2009). It is clearly a negative emotion, which, next to anger, anxiety, helplessness and shame, influences various aspects of the learning process to a greater extent than positive emotions such as enjoyment and pride (Forgas, 2013;Sansone & Thoman, 2005). Boredom has been seen as having more or less the same impact on school performance and achievement as anxiety (Tze et al., 2013). It has also been shown to be related to learners' impulsivity and risk-taking as well as decreased self-regulated learning and diminished motivation (Daschmann et al., 2011;Preckel et al., 2010). The motivation of bored students is lower since they are disengaged from school subjects or tasks, which results in their inability to concentrate or simply manifest interest and joy (Pekrun & Linnenbrink-Garcia, 2012).
There are several theories that can be referred to in order to shed light on the causes of boredom. One of them, the under-stimulation model (Larson & Richards, 1991), highlights the paucity of challenging incentives to learn as the reason for deactivating students and discouraging them from eager participation in school activities. In the so-called forced-effort model (Hill & Perkins, 1985), boredom is conceptualized as a consequence of imposing on students tasks into which they have to put excessive cognitive effort, although they perceive these tasks as monotonous and unattractive. The attentional theory of boredom proneness (Eastwood et al., 2012;Harris, 2000), in turn, posits that the primary cause of boredom is an individual's inability to self-regulate attention. Attentional problems are related to individual differences between learners and can arise as a result of the cognitive demands of a task but can also be traced back to interests, concerns or values (Mercer-Lynn et al., 2014). It is also worth mentioning the emotion theory (Eastwood et. al., 2007(Eastwood et. al., , 2012, according to which boredom may stem from difficulty in recognizing, accessing and communicating one's own feelings, a phenomenon that has been diagnosed as alexithymia. Yet another attempt to explain why students get bored has been made within the framework of the control-value theory of achievement emotions (Pekrun, 2006;Pekrun et al., 2010). In this view, students' appraisals of control over the task at hand and the value they attribute to this task determine whether they will experience boredom. The final theoretical perspective is the dimensional model (Pekrun et al., 2010), which views boredom as both a deactivating and activating emotion. This means that, perhaps quite paradoxically, in certain circumstances the negative state of emotional ennui may bring about arousal behavior aimed at searching for change. Seen in this way, boredom assumes the role of a functional negative emotion that may push students into setting new goals or reformulating those they are pursuing (cf. Komorowska, 2016).
In light of such a variety of approaches to the occurrence of boredom and the reasons which underlie this occurrence, the concept, whether conceived of in the field of psychology or SLA, is exceedingly difficult to define. However, as the focus of the present study is on naturally-occurring EFL classes, the concept is equated in the present article with some form of disengagement from the classroom activities implemented by the teacher, although adopting such a definition surely runs the risk of oversimplification. At the same time, it has to be noted that the connection between boredom and disengagement is not at all a novel idea since the former is often described as the experience of being disengaged from the task at hand or just as a kind of engagement problem (Macklem, 2015). What also deserves attention here is that disengagement is one of the five factors included in the Multidimensional State Boredom Scale (Fahlman et al., 2013), the others being high arousal, low arousal, inattention and time perception, which shows how complex and unobvious the concept of boredom is. In fact, it could as well be argued that boredom is intricately tied to a number of other factors such as motivation, demotivation, engagement, involvement, interest as well as disinterest (Ainley, 2012;Pekrun et al., 2010), to name but a few. Even though this issue will be touched on and illuminated in the discussion section, it must be borne in mind that relationships of this kind can only be successfully teased apart through future research.

The intensity of boredom
Numerous manifestations of boredom have been taken into consideration in several typologies attempting to shine a light on its underlying mechanisms and to pinpoint its most distinctive features. However, given the focus of this article, the most noteworthy is the five-subtype taxonomy (see Table 1) concerning the intensity of boredom because it tells us more about how this emotional state might change depending on specific circumstances and how it might influence student behavior. In a nutshell, the typology indicates that in some instances the experience of boredom may serve as a motivator enabling students to realize what inhibits the performance of the task at hand, thereby encouraging them to switch goals.

Research on boredom in L2 teaching and learning
As mentioned before, research into boredom in the L2 classroom is extremely scant. To the best of the authors' knowledge, there are just five studies directly related to this issue (i.e., Chapman, 2013;Kruk, 2016aKruk, , 2016bKruk & Zawodniak, 2017, 2018. Several other publications can be seen as indirectly connected with the concept of boredom, either because they explore its flipside in the form of flow in task performance (Aubrey, 2017) or interest and engagement in classroom activities (see e.g., Peacock, 1997;Tin, 2016), or they focus on the broader notion of demotivation (Kikuchi, 2015). However, due to space limitations, the discussion in the present section is confined to studies that have specifically addressed the issue of boredom. Chapman (2013) investigated the beliefs about this negative emotion manifested by learners of German as a foreign language and their teachers. It turned out that the best predictor of boredom were learners' feelings towards the teacher which were much more influential than the nature of tasks and activities performed. Kruk (2016a), in turn, conducted a study which explored changes in the level of boredom experienced by senior high school students learning English as a foreign language over the period of three weeks. The use of several data collection tools yielding both quantitative and qualitative results allowed the researcher to identify fluctuations in boredom levels both in single classes and sequences of classes. The reasons for such changes were both more general, such as the students' proneness to boredom, and situational, connected with the nature of the activities. Kruk (2016b) also conducted a longitudinal research project in which he examined the changes in the level of boredom, alongside motivation and anxiety, over one semester in a virtual environment constituted by Second Life. Using a mixed-methods approach, he demonstrated that while the levels of boredom changed together with those of motivation and were likely to increase over time, anxiety remained relatively stable. He showed as well that lower levels of boredom could be attributed to participants' enthusiasm about the opportunity to use Second Life as a means of practicing the L2, but also as a kind of entertainment.
Another two studies dealing with the role of boredom in L2 learning were carried out by Zawodniak (2017, 2018), also in the Polish educational context. The first (Kruk & Zawodniak, 2017) examined the relationship between overall boredom experienced by English majors and the boredom they exhibited in EFL classes (e.g., speaking, writing, grammar). Quantitative analysis revealed a significant, positive correlation between general boredom proneness and boredom proneness specific to EFL classes, with the intensity of this negative emotion increasing over time. Thanks to qualitative data, it was also possible to pinpoint causes of boredom, the most important of which included activities unadjusted to participants' L2 proficiency, the repetitive character of these activities, the teacher, and form(s) of work applied. Kruk and Zawodniak (2018) also investigated the experience of boredom in EFL classes, focusing only on a sample of the participants of the previous study (i.e., Kruk & Zawodniak, 2017) and relying solely upon qualitative procedures. Aside from investigating the reasons for student boredom and changes in its intensity over time, the researchers probed into the manner in which students manifested their boredom, the ways they coped with it, and the differences between this kind of experience in English classes and other academic subjects. The students were more active in their attempts to overcome boredom outside of school compared to in-school situations. Participants also proved to be more bored with theoretical subjects (e.g., descriptive grammar) and electives (e.g., lectures) than with EFL classes which they simply perceived as more useful, attributing their advantage to small groups and the use of English as the only language of instruction.
The common thread running through most of the recent studies is the changing nature of L2 student boredom. Based on the empirical evidence collected to date, this variable can be regarded as dynamic as well as temporally and spatially situated, intertwined and constantly interacting with other systems (Larsen-Freeman, 2016), such as, for example, motivation, anxiety, pedagogical procedures or group dynamics. Although the investigations reported before have somewhat illuminated this issue, the insights could only be limited, mainly because of the methodology employed, which did not allow a more in-depth look into fluctuations in levels of boredom as a function of contextual and individual factors. The present study seeks to fill this gap by investigating the evolving nature of boredom in real time, during regularly scheduled EFL classes, also aiming to shed light on the complex interplay of factors responsible for such fluctuations.

Research questions
The study aimed to explore changes in the levels of boredom of advanced learners of English in four EFL classes that were taught by one of the present authors. As mentioned before, for the purpose of this study boredom is defined as a state of disengagement caused by lack of interest and involvement. The following research questions were addressed: 1. How do levels of boredom change over the course of a single class and from one class to the next? 2. What factors are responsible for the changes in boredom levels?

Participants
Participants were 13 Polish university students (12 females and 1 male) majoring in English, enrolled in the final year of a three-year BA program. The students were on average 21.92 (SD = 0.76) years of age and their mean experience in learning English, their TL, amounted to 12.92 (SD = 3.52) years. Overall, participants' command of English could be characterized as ranging from B2 to C1 according to the Common European Framework of Reference for Languages (Council of Europe, 2001). However, some individual variation could be observed regarding overall mastery and the command of different TL skills and subsystems. The students' average grade in the end-of-the-year examination in English, typically used for assessment purposes in Polish universities, was 3.63 (SD = 0.65) on a scale from 2 (fail) to 5 (very good). Such evaluation largely corresponded to participants' own perception of their TL ability, which is evident in their self-assessment that was only a little higher and stood at 3.73 (SD = 0.63) on the same scale. It should be noted that students gave their consent to take part in the study.

Data collection instruments and procedures
The study comprised four naturally occurring EFL classes scheduled one per week. They were conducted by one of the present researchers in a group of the 13 participants described before. Table 2 provides a short description of the activities included in the four successive classes under investigation, with a focus on the activities applied, the skills practiced, and the modes of class organization employed.

Table 2 Tasks and activities included in the classes under investigation
Class 1 Class 2 Class 3 Class 4 · organization · speaking: memory; pairwork · reading: memory techniques; individual work · speaking: mnemonic techniques; whole-group discussion · grammar: modal verbs; individual work, pair-work · listening: instructions; individual work · vocabulary: sorting out words; individual work · conclusion · organization · speaking: formal letters; pair-work · reading: matching true/false questions; individual work · vocabulary: classifying words, text completion; individual work · grammar: relative pronouns; individual work · speaking: discussing dangerous sports; pair-work · listening: an interview, completing sentences; individual work · conclusion · organization · vocabulary quiz; individual work · listening: answering questions, completing notes; individual work · grammar: phrasal verbscompleting and matching sentences; individual work · vocabulary: collocations; individual work · speaking: work and business; pair-work · reading: answering questions; individual work · conclusion · organization · vocabulary: filling gaps, literal and metaphorical meanings of words; individual work, pair-work · grammar: language of cause and effect, text completion; individual work · reading: answering questions; individual work · listening: answering questions; individual work · conclusion Drawing upon previous research on the dynamics of motivation and willingness to communicate (e.g., Pawlak, 2012;Pawlak et al., 2016), four instruments were used to collect the data on fluctuations in boredom levels: · a background questionnaire was filled out by the students at the start of the study and provided information needed for the description of participants included in Section 3.2; · an in-class boredom questionnaire, which consisted of four parts: (1) Part One, filled out at the beginning of each class, where students provided their names and the date of the class; (2) Part Two, completed during each class, that consisted of a boredom grid in which participants self-rated the level of their boredom on a scale from 1 (minimum) to 7 (maximum) at 5-minute intervals in response to a prerecorded sound; the Cronbach alpha value for this instrument was 0.95, which speaks to high internal consistency reliability; (3) Part Three, completed towards the end of the class, which included seven items (they were chosen on the basis of the relevant literature, for example, Pawlak, 2012;Peacock, 1997) on a semantic differential scale (i.e., meaningless vs. meaningful; dull vs. exciting; useless vs. useful; unsatisfying vs. satisfying; usual vs. unusual; unappealing vs. appealing and monotonous vs. absorbing); this in fact constituted a 7-point Likert scale as the extreme responses were accorded the value from 1 (e.g., dull) to 7 (e.g., exciting); Cronbach alpha equaled 0.86 and can also be deemed satisfactory; (4) Part Four, which required students to write a short paragraph about the experience of boredom during each class ("Please write a few sentences related to the experience of boredom during the class"); · a semi-structured interview, which was held immediately after each class; since the interview was held immediately after classes and since all the students who participated in this study had other classes scheduled after the one investigated in the study, the decision was made not to interfere with the participants' desire to take part in them; thus the interview was conducted with four volunteers who were different individuals each time; the interviews were audio-recorded; interviewees were asked questions regarding reasons for changes in boredom levels depicted in the grid and their overall opinions about a class (e.g., activities performed, materials used); · lesson plans, which provided information related to the conducted classes, their stages and the language activities performed.
Importantly, Polish was used in all the data collection instruments to avoid misunderstanding or misinterpretation, and participants were allowed to use Polish or English when responding to open-ended questions. For the same reason, Polish was used to conduct the four interviews.

Data analysis
Depending on the nature of the data, they were analyzed quantitatively and qualitatively. In the case of the former, descriptive statistics in the form of means and standard deviations were calculated for some of the items in the background and in-class questionnaires. As for the latter, it involved transcription of the participants' narratives (Part Four of the in-class questionnaire) by means of a computer word processor and partial transcription of the individual interviews (Dörnyei, 2007). Each of the three researchers went through the transcripts with a view to pinpointing the factors accounting for the experience of boredom. To be more specific, notes and annotations were made to record any immediate observations. Next, the themes were identified and labeled (e.g., some themes concerned tasks and some were related to the characteristics of a specific class). This was followed by a discussion during which the dominant themes and tendencies were agreed upon. Such data were juxtaposed against the lesson plans to arrive at the factors that may have impacted the levels of boredom experienced during the classes in question.

Overall variation across classes
As can be seen from Table 3, which presents the mean level of boredom in each class, both overall and at 5-minute intervals, the participants experienced the most boredom in Class 1 (C1) and the least in Class 4 (C4), which is evident in the total means of 3.57 and 2.87, respectively. In addition, overall, the students tended to be the most interested at the start of each class, that is in Minute (M) 10 and M15, when the average boredom level stood at 2.23 and 2.42, respectively, and they felt the most boredom towards the end of the class, in M60 and M65, when its mean levels oscillated around 3.90. It is also clear that the changes in boredom levels were the most pronounced in C1, with the difference between the lowest and the highest mean values equaling 2.46. Conversely, the difference between the extreme values was the least prominent in C4, in which it equaled 1.30. The amount of variation in boredom levels may be the corollary of the degree to which individual participants were involved in a particular class, as shown by standard deviation values which proved to be the highest for C1 (SD = 0.77), followed by C2 (SD = 0.66), C3 (SD = 0.54) and C4 (SD = 0.51).

Variation in boredom levels during individual classes
Now that the overall patterns of change in boredom levels have been described, a more detailed analysis is provided in order to take into account the activities that each of the classes comprised. This is elaborated subsequently with respect to the four classes under investigation.

Class 1
When it comes to C1, Figure 1 shows that the participants were quite interested in the lesson in its first 15 minutes. This is evident in the fact that the initially reported boredom level equaled 2.36 points and it was even lower in M15 (2.18 points, a difference of 0.18). This was the time when the teacher explained the aims of the class, commented on activities to be included and provided feedback on the homework assignment. The situation started to change during the subsequent speaking activities in which students were requested to ask and answer questions about memory (S1), discuss tips on how to improve it (S2), and talk about techniques they would like to try out (S3), where an increase in boredom levels from 2.18 (M15), first to 2.91 (M20) and then to 3.36 (M30 and M35), could be detected. The level of boredom continued to rise throughout the reading (R) and speaking (S4) activities (4.00 in M40 and 4.27 in M45), in which students were asked to read about mnemonic techniques useful in language learning and discuss them. Students' boredom kept increasing during the next two grammar activities which focused on matching the meaning of the modal verb can with its use (G1) and explaining the difference in meaning between sentences containing modal verbs (G2). Somewhere in the middle of the third grammar activity (G3) in which students were instructed to look at pictures and speculate about what might have happened, however, the level of boredom fell to some extent (4.00 in M55), only to rise again at the beginning of a listening activity (L), reaching its maximum in M60 (a difference of 2.46 in comparison to the lowest value in M15). It should be noted, though, that during this listening activity the level of boredom decreased a little, with the difference between M60 and M65 equaling 0.28. The experience of this academic emotion kept decreasing in the final stage of the class (a difference of 0.09 from M70 to M75) as a vocabulary activity (V) was assigned in which students grouped vocabulary items.

Figure 1
Changes in the level of boredom together with language activities performed in C1

Class 2
As illustrated in Figure 2, in C2 the reported level of boredom was on the rise from the beginning until M55 and M50, which was followed by a drop, a slight increase, and a final decrease. A closer look at Figure 2 reveals that the students were the least bored with the part of the class devoted to introductory issues and feedback on homework (O+H; 2.29 -the first 15 minutes). The activities related to composing formal letters (i.e., S1 -reading pertinent extracts and discussing their purpose; S2 -discussing the layout) resulted in a slight rise in the feelings of boredom (an increase of 0.11 from M15 to M20) and then a levelingoff at 2.71 (M20 and M25), when participants had to read a text and match headings to the "secrets" of writing business letters (R1). This was followed by an increase in reported levels of boredom, which began to manifest itself in the second half of R2 (i.e., the activity connected with reading a text and answering true/false questions) and was maintained during the subsequent two vocabulary activities (V1 -grouping words under appropriate headings; V2 -completing a text with given words) and two grammar exercises (G1 -working with relative pronouns; G2 -rewriting sentences in a formal style). The change in the levels of boredom from M30 (2.86) to M55 (4.29) was quite substantial and equaled 1.43. The intensity of boredom fell during a speaking activity (S1) in which students were requested to discuss questions about dangerous sports (a difference of 0.58 between M60 and M65), and then it rose slightly (a change of 0.15 between M65 and M70) in the middle of a listening task (L -an interview with a successful Formula One driver) where participants were asked to complete sentences with correct words, only to fall again by 0.15 at the end (M75).

Figure 2
Changes in the level of boredom together with language activities performed in C2

Class 3
When it comes to C3, Figure 3 shows that, yet again, the students were the least bored during the initial phase devoted to organizational issues (2.18 in M10). Then the level of boredom increased by 0.64 during a vocabulary quiz (M15) and remained the same until its completion (M20). This was followed by another increase in boredom levels by 0.45 at the end of a listening activity (L1) in which participants were asked to listen to a short anecdote and answer three questions, followed by a slight decrease of 0.18 and a slight increase of 0.27 in the middle and at the end of another listening task (L2), respectively, which involved listening to phone messages and completing notes. This relatively low level of boredom (i.e., 3.36) was maintained throughout two grammar activities on phrasal verbs (G1 -completing sentences; G2 -matching sentences). A slight increase to the level of 3.55 was subsequently observed at the start of a vocabulary task (V) dealing with collocations (putting words under correct headings and then choosing collocations to complete sentences), which was followed by a speaking activity (S) in which the students discussed in pairs questions related to work and business. While this speaking task inspired more involvement and interest (a drop of 0.28 in M55), these faded away rather quickly, as is evident in the fact that average boredom levels bounced back to the previous levels towards the end of the task in M60 (3.55, a change of 0.28). The intensity of boredom was even higher in the first half of the last reading activity (R) (a change of 0.54 from M60 to M65) in which students were asked to read a rather long text and answer four questions. The mean self-ratings then leveled off and remained at 4.09 for the last two consecutive measurements (i.e., M70 and M75).

Figure 3
Changes in the level of boredom together with language activities performed in C3

Class 4
Figure 4 shows that, as was the case with the remaining classes, the participants' self-ratings of boredom in C4 were the lowest in its initial stages, dedicated to organization and checking homework (2.08 in M10 and M15). Somewhere in the middle of the first vocabulary activity involving filling out gaps in sentences a minor rise in boredom was registered (a change of 0.05 from M15 to M20), followed by a major one at the end of the task (a change of 0.31 from M20 to M25). The boredom levels did not change during the second vocabulary task (V2) dealing with literal and metaphorical meanings of words from the text covered in the previous class (i.e., C3) but rose steeply at the end of the activity, increasing by 0.69 from M30 to M35. Boredom continued to rise during the first two grammar activities in which students read a short text and underlined expressions linked with the language of cause and effect (G1), and completed a text with suitable words (G2), reaching the maximum level in M40 (3.38). The third grammar activity (G3) in which participants were requested to listen to an actor and indicate the words and expressions related to cause and effect led to a slight drop in boredom (a change of 0.30 from M40 to M45). Its levels did not alter in the first part of the subsequent task (R) devoted to reading a text about work experience and answering questions related to that text. However, a slight increase was detected in the second part of the task, followed by a minor decrease (a change of 0.16 from M55 to M60) when the understanding of the text was checked. The group then reported a slight increase in boredom (a change of 0.23 from M60 to M65), a stable period between M65 and M70, and a final small drop at the time of the last measurement (a change of 0.30 from M70 to M75). These ups and downs in boredom levels happened during the last activity (L) when students were requested to listen to eight people talking about their jobs and answer related questions.

Figure 4
Changes in the level of boredom together with language activities performed in C4

Students' evaluation of the four classes
As can be seen from Table 4, the participants' overall evaluations of the four classes by means of the semantic differential scale (i.e., meaningless-meaningful, dull-exiting, etc.) changed from one class to the next, with the caveat that, in this case, a higher mean indicates a lower level of boredom. More precisely, they showed that students were less engaged in the first two classes than in the last two, a finding that only partly overlaps with the results presented before, where the overall levels of boredom in C2 and C3 were comparable (3.36 vs. 3.35). The students were the most positive about C4 and the least about C1 (a major difference of 0.83), which corresponds to the lowest and highest overall levels of boredom, respectively (2.87 vs. 3.57)

Narratives and interviews
Apart from juxtaposing the participants' self-evaluations with what transpired during the four classes, further insights into factors responsible for changes in boredom levels emerged from the qualitative analysis of the narratives, the interviews and the related notes.
When it comes to C1, the main factor generating boredom was the reading task. This is because it was not considered "a creative activity" as it dealt with "an uninteresting topic," involved reading a long text ("boredom was caused by reading a lengthy passage") and was accompanied by an exercise that was perceived as boring ("monotonous exercises related to checking comprehension of the text"). In addition, the reading task was generally regarded as too demanding, thus evoking the feeling of boredom. Another factor responsible for boredom in C1 pertained to the performance of unchallenging language activities, their predictability and repetitiveness. Some students were also bored with speaking ("the speaking activities which I don't like very much and which take a lot of time") and listening tasks ("an uninteresting listening exercise"). One student mentioned the length of the class and another noted that lack of sleep made her feel tired and bored. When asked about the increase in the self-reported levels of boredom, the student interviewed afterwards said that she always felt bored during speaking tasks because she did not like to talk about "imposed topics" and participate in "artificial conversations." The student's level of boredom was also higher during the second grammar activity (G2) in which participants were supposed to explain the difference in meaning between pairs of sentences involving modal verbs. The interviewee was simply confused with it since, as she said: "I didn't know what to do." With respect to C2, the students listed the following factors responsible for the experience of boredom: (1) reading a boring text and listening to a boring interview, (2) doing easy grammar exercises, (3) a limited number of speaking tasks, and (4) the recurring pattern of language activities. However, according to the student interviewed after the class, boredom was not caused by the factors just mentioned, but it was related to the length of the class. This is because she became disengaged somewhere in the middle and then boredom increased with time and was the highest towards the end. As she commented: "I spent half an hour and I wasn't bored at all. Then I started to feel a bit bored . . . Only after 50 minutes did I start to feel really bored . . . 50 minutes is a long time . . . sitting all the time in one place." As regards C3, the factors inducing boredom were similar to those in the previous classes and included reading a text, completing easy and monotonous language activities ("We could do more diverse exercises") and performing only one speaking activity ("If there had been more discussion, I'd have been bored even less"). It should be noted, however, that some students did not mention any boredom-evoking factors but pointed to the vocabulary quiz conducted during the first part of the class as a positive influence on their engagement. According to two participants, "the unexpected test for sure 'revived' the group and made us more motivated" and was "a good way for waking up 'sleepy' heads." The student interviewed after the class said the she felt a little bored during the speaking task performed in pairs, which focused on work and business, since, as she said: "I personally like to discuss a topic with the whole group." Another factor that, in her opinion, contributed to the experience of boredom was external in nature, being related to noise she could hear in one part of the class. She described this as follows: "I heard a chainsaw or something . . . I felt deconcentrated and I wasn't able to focus for a while . . . [Did the noise cause boredom?] Yes, to some extent . . . I started thinking about something else and I needed some time to focus again." Finally, the analysis of relevant data in C4 revealed that the reading and listening tasks were the major factors responsible for the lack of involvement for six students. In the case of the former, boredom resulted from the length of the activity while, in the case of the latter, it was triggered by an uninteresting topic. In addition, two students mentioned the absence of a speaking task (e.g., "The only missing task during the lesson was speaking in pairs. In my opinion, such an activity can wake us up and prevent us from becoming bored"). Another learner was bored because "the language activities did not differ much from those I did in my high school" and yet another saw C4 and the activities it included as "carried out in the usual manner." The student who was interviewed also complained about the length and monotony of the listening task and perceived it as a factor generating boredom. This can be seen in the following comment: "Also the listening . . . very extensive . . . and because it was not very absorbing, it made me less attentive." Surprisingly, she pointed to "low blood sugar" as another factor which may have had an impact on her boredom in that class.

Discussion
The study was guided by two research questions, one concerning the nature of changes in the intensity of boredom and the second focusing on factors responsible for such dynamics. When it comes to the first issue, the analysis demonstrated that, similar to motivation (e.g., Pawlak, 2012) or willingness to communicate (e.g., Mystkowska-Wiertelak & Pawlak, 2017;Pawlak et al., 2016), the intensity of boredom is subject to major fluctuations over different timescales. For one thing, the students' average boredom changed considerably over the four classes, being at its height in C1, leveling off in C2 and C3, and reaching the lowest intensity in C4. Secondly, ups and downs in its levels were also evident in the four classes under investigation, although the scope and amplitude of such fluctuations varied and so did the overall pattern of the changes. For example, while the general experience of boredom was the highest in C1 and it is here that the greatest amplitude of fluctuations was revealed, the opposite was found for C4, in which disengagement was not only the lowest but also the differences in its levels were the least pronounced. It should also be noted that while all the classes shared similar characteristics with respect to boredom, such as its lowest levels in initial stages, there were also differences related to the frequency of rises and falls as well as the overall tendencies concerning the intensity of this experience. For instance, a relatively steady rising pattern in C2 contrasts to some extent with a more intermittent one in C1, whereas the persistently high level of boredom in C3 at the end of the class stands in contrast to the drops in its intensity in the final stages of the remaining three classes. Such findings are more or less in line with the results of previous studies, in particular those involving senior high school students and English majors (Kruk, 2016a(Kruk, , 2016b. Two crucial observations should be made, however. First, the highest intensity of boredom corresponded with the greatest extent of individual variation, as expressed in SD values (C1), which testifies to a considerable role of learners' personal agendas, individual profiles, and preferences. Second, although students' post-hoc assessment of the classes mirrored in most cases their self-evaluation of the intensity of boredom, there is a discrepancy in the case of C3, which may indicate a certain degree of ambivalence as to whether it was overall interesting (and thus motivating) or boring.
As regards the factors accounting for students' experience of boredom and changes in this regard, the analysis yielded some interesting insights which, on the one hand, go beyond and expand on the findings of previous studies, and, on the other, can be related to theories referred to earlier in this article. It should be emphasized from the outset that pinpointing the influences that shaped the intensity of boredom in the four classes, let alone ordering them according to the magnitude of their importance, poses a formidable challenge since the role of various factors overlapped, creating clusters of variables which could have played a very different role in each class. To give but one example, at first blush, it would appear that the level of boredom was a function of the skill or subsystem being practiced as both the self-ratings of this academic emotion and participants' comments in the narratives and interviews indicate that, on the whole, activities focused on reading or grammar were more boredom-inducing than those involving speaking or vocabulary. On closer inspection, however, it becomes clear that the reality is much more complex and the factor that may trump all the other influences is the nature of the activity. In particular, what seemed to matter the most was the challenge posed by the task (e.g., too easy), its duration, its relationship to other activities (e.g., several similar activities in a row were likely to enhance boredom), excessive reliance on the coursebook, and, by far most importantly, the degree of repetitiveness and predictability, or, conversely, the degree of novelty that a specific task entailed. For example, it was not the inclusion of a speaking activity as such that played the decisive role but, rather, the demands it placed on the students and the topics discussed, particularly when they were simply imposed by the teacher rather than negotiated, but also the mode in which a task was performed (i.e., whole class, group, pair). Other factors that may have impacted boredom levels were more idiosyncratic and tied to the length of the class, external distractions (e.g., noise) or the beliefs, expectations or preferences of individual students. Although the results mirror to some extent those obtained in prior studies (e.g., Kruk, 2016aKruk, , 2016bKruk & Zawodniak, 2017), conspicuous by its absence is the teacher, who was previously identified as a major influence on the occurrence and intensity of boredom (e.g., Chapman, 2013). A possible explanation is that the study took place in classes taught by one of the authors, which may have dissuaded students from commenting on this source of boredom. It is also possible to account for causes of boredom through the lens of some of the theoretical stances mentioned earlier in this article, such as the under-stimulation model (Larson & Richards, 1991), since the participants clearly lacked the right kind of challenge, the forced-effort model (Hill & Perkins, 1985), as the monotonous and repetitive activities did not warrant the amount of cognitive effort invested, control-value theory of achievement emotions (Pekrun, 2006;Pekrun et al., 2010), since the participants had little influence on the choice and implementation of the tasks, but also the dimensional model (Pekrun et al., 2010), as on some occasions boredom was overcome by the onset of a new activity, even when it was not appealing as such. As regards the intensity of boredom, using the classification proposed by Goetz et al. (2014), the participants seemed to have experienced mainly its calibrating and searching types. This is because, on the one hand, they could not change what transpired in class and thus may have engaged in off-task thoughts and, on the other hand, they wanted to combat their disengagement, thus seeking more interesting things to do.
There are several strengths of the study that should be emphasized here: (1) it followed the same group of students during four classes, which allowed minimizing the mediating impact of individual difference variables; (2) multiple data collection tools were employed, thus enhancing the validity of the findings; and (3) the data were collected during regularly scheduled classes, which increases the ecological validity of the investigation. This said, the study also suffers from some weakness. First, similarly to other research projects drawing on this methodology (e.g., Pawlak, 2012;Pawlak et al., 2016), a question arises as to the extent to which indicating the levels of boredom every five minutes may have interfered with cognitive processing and task performance as well as may have caused disruptions or even evoked more boredom. Even though this is clearly a valid concern, similar tools have been successfully employed in other studies and it is difficult to see an alternative that would have allowed obtaining requisite real-time data (see e.g., Pawlak, 2012 or some of the papers included in the edited collection by Dörnyei et al., 2015). Second, the analysis focused upon overall levels of boredom, which might have masked students' individual trajectories, an issue that surely deserves investigation but could not be accommodated in this article for reasons of space. Third, the data were collected during classes taught by one of the researchers, which may have impacted the honesty of participants' responses to some extent. However, the students were reassured that their candid responses would not negatively influence their assessment and the results reveal that most of them did not have second thoughts about indicating their increased boredom or pointing to factors responsible for its occurrence. Fourth, and perhaps much more fundamentally, it could be argued that investigating boredom may be superfluous because it constitutes little more than the flip side of motivation. In our view, however, this reasoning is flawed since a lack of motivation (i.e., demotivation or amotivation, see Kikuchi, 2015 for a discussion) does not have to involve the presence of boredom and the other way around. For example, a learner who is demotivated may in fact be engaged in some parts of a lesson that he or she finds interesting, and a student who is overall motivated may still get bored with the performance of repetitive and monotonous activities. Even if we narrow down the concept of motivation to engagement and involvement in a particular class, as Pawlak (2012) did, a lack of such engagement may not be an indicator of boredom but, for example, limited enjoyment, the presence of anxiety or scant willingness to communicate. All of this indicates that boredom is a distinct construct which deserves to be examined in its own right.

Teaching implications
Since research on the role of boredom in the L2 classroom is still in its infancy, it is clearly premature to offer concrete pedagogical recommendations, and the results of the present study should also be taken with circumspection. Nonetheless, it is possible to offer a handful of suggestions that might help teachers avoid situations in which learners, especially those more advanced, such as the English majors involved in the present study, experience excessive levels of boredom in the classroom. First and foremost, disengagement from classroom tasks and activities is most likely to take place as a result of monotony, repetitiveness and lack of challenge. Therefore, teachers should strive to introduce a modicum of novelty into their classes, which might require abandonment of strict adherence to the coursebook and reliance on materials that originate from other sources, such as the internet. Second, teachers must be cognizant of the fact that, at least based on the data collected from the participants of this study, the feeling of boredom is largely an individual issue, which surely testifies to the importance of catering to the needs of particular students, even though this is clearly bound to produce considerable challenges. Third, it is essential that activities are sequenced appropriately in the sense that particular components of a lesson proceed progressively from easier towards the ones that may pose some difficulties. Fourth, it may be a good idea to encourage students to ask for clarification anytime they find themselves at a loss as to how to cope with a given task. Fifth, specifically at more advanced levels, teachers could be advised to offer their students more opportunities for choosing what classes should focus on and make sure that the tasks and activities included represent the right kind of challenge.

Conclusion
The study has without doubt provided valuable insights into the changes in boredom levels in the course of regularly-scheduled English classes taught to advanced learners and has shed light on constellations of factors responsible for these changes, also linking the findings with theoretical perspectives in educational psychology. At the same time, it has to be recognized that, given the overall paucity of research in this area, the findings constitute just one piece of the puzzle concerning the role of boredom in L2 learning. Within the micro-perspective that this study represents, it would be instructive, for instance, to examine how boredom affects students with diverse individual profiles (e.g., in terms of motivation, learning styles, learning strategies, personality, beliefs), to what extent it is determined by the composition of a class and group dynamics (e.g., by exploring groups following similar lesson plans), or what kind of pedagogical intervention can be used to handle this problematic emotion and whether it should involve appropriate motivational strategies (see Dörnyei, 2001). However, as emphasized by Mystkowska-Wiertelak and Pawlak (2017), the investigation of individual difference factors is the most effective when a micro-perspective is adeptly combined with a macro-perspective, where a variable is investigated by means of meticulously designed tools with much larger populations. This is also a direction that future research on boredom should take if we seek to provide more global insights on its role in L2 learning and teaching and the forces shaping it. It can only be hoped that future research will provide evidence of not only why boredom sets in and changes in language classes, but also of effective interventions that can alleviate it.