A rollercoaster of emotions: An integrative review of emotions and its impact on health professional students' learning in simulation‐based education

Abstract Aim Simulation‐based education establishes a specific learning environment capable of activating emotions before, during and after the task. Research has identified stress and anxiety related to simulation. However, the role of various emotional experiences in a simulation that favour learning is still unclear. This review describes, interprets and synthesizes the current research findings on health professional students' experience of emotions and the effects on student learning in simulations. Design This study design was guided by integrative review method. Methods Databases were systematically searched for articles. 9,323 records were screened and 16 studies met the inclusion criteria. The study protocol was reported in Prospero. Results Three themes emerged from the analysis: (a) simulation as a fearful and stressful situation, (b) variability in emotions experienced during simulation as a rollercoaster of emotions and (c) emotions wide‐ranging effects on students' learning in the simulation.


| INTRODUC TI ON
Emotions can influence memory, attention, cognitive and metacognitive thinking strategies, judgement, decision-making, cognitive resources and both intrinsic and extrinsic motivation (Artino et al., 2012;Gluck et al., 2016). Research into how emotions affect learning has increased in the past decade (Pekrun, 2019;Pekrun & Linnenbrink-Garcia, 2014). However, the ability of the context and learning activities to sway students' emotions have not been investigated thoroughly, especially for simulated learning. Simulationbased education (SBE) is a learning activity in which the clinical work environment is simulated in the laboratory. Simulation has attracted attention in health-care education and is an important method for teaching clinical skills in a safe learning environment (Campbell & Daley, 2018;Levine et al., 2013;Motola et al., 2013). Education in nursing, medicine, pharmacology, physiotherapy, dentistry and occupational therapy are some of the health areas currently using SBE (Bethea et al., 2014;Blackstock & Jull, 2007;Harder, 2010;Hattingh et al., 2018;Perry et al., 2015).
By working in teams during SBE, students learn to manage patient situations in a realistic environment but away from real patients (Campbell & Daley, 2018;Jeffries, 2020). The benefits for health professional students are the opportunity to practise in a safe environment and make mistakes without harming patients. Teachers can adapt the situations to achieve specific learning outcomes. In an academic situation, emotions are often provoked by the learning environment itself, and research has shown that emotions can impact learning (Pekrun & Linnenbrink-Garcia, 2014). Because students are highly activated both mentally and physically, SBE is a specific learning environment that can lead to the activation of emotions before, during and after the task.
Historically, research on cognition and emotions has been conducted separately, although there has been a shift towards a more integrative approach to provide a holistic understanding of students' learning processes. Studies have focused on student learning and reflection during simulation (Cant & Cooper, 2010, 2017Hegland et al., 2017;Husebo et al., 2013). Student learning outcomes include acquisition of patient care skills and clinical competency in addition to improved knowledge, consider alternative solutions to a clinical problem and critical thinking abilities. SBE has been found to boost student self-confidence and satisfaction (Cant & Cooper, 2010;Issenberg et al., 2005;Olaussen et al., 2020). However, despite these learning outcomes and student satisfaction, students participating in SBE also report anxiety and stress (Al-Ghareeb et al., 2017;Cantrell et al., 2017).
There is a need to study the role of emotions in different contexts to understand how emotions can affect student learning (Artino et al., 2012;Pekrun, 2019). Health education has ignored the important role emotional experiences have for learning (LeBlanc et al., 2015). Given the current importance of SBE for health professional students, a greater understanding of the importance of emotions to learning in this context is needed.
This review describes, interprets and synthesizes the current research findings on health professional students' experience of emotions and the effects on student learning in simulations.

| Simulation-based education
Simulation-based education usually follows a template (Jeffries, 2020) starting with a briefing session which students are informed about the patient's situation and learning goals and are allowed to become comfortable with the equipment. In second session, the scenario unfolds, and students become engaged and active in solving the patient's problems. Finally, a debriefing session provides the opportunity for students to reflect on and discuss their actions. Debrief has been identified as the aspect in which most learning occurs (Dufrene & Young, 2014;Levett-Jones & Lapkin, 2014). Debriefing provides the opportunity for complex learning by identifying problems that occurred during the simulation and how these were solved, making comparisons and explaining their actions as a way to demonstrate their learning.

| Learning in SBE
Students add new information to their already existing knowledge while learning. This reconstruction of knowledge takes place through the interplay between cognition, emotions and social dimensions (Biggs & Collis, 2014;Illeris, 2015). This process is also relevant to SBE. During simulation, students are asked to apply their existing knowledge and skills in a new setting. To do so students create new connections with varied solutions and dilemmas, and this process challenge them to learn on a deeper level. The intention of SBE is to facilitate the extension of the current understanding into new contexts, in this case, health professional practice. Through this process, learning is extended to more abstract and complex levels (Biggs & Collis, 2014).

| Academic emotions in SBE
Participating in SBE can activate students both mentally and physically. This activation can trigger academic emotions before, during and after SBE. Academic emotions are those that can arise in different academic settings, and emotions reflect a heterogeneous experience that encompasses a wide variety of psychological phenomena (Barrett, 2004). Emotions are explored and expressed widely, and can be categorized as positive emotions, such as happiness, hope, joy and satisfaction, or negative emotions such as fear, anxiety, sadness, shame and disgust (Pekrun & Linnenbrink-Garcia, 2014). The circumplex model (CM) of affect classifies emotions into two dimensions: valance and arousal (Posner et al., 2005) (Figure 1). Valence is the experience of a situation as either positive and pleasant or negative and unpleasant. Arousal is the human experience of being either physiologically activated or deactivated in a situation (Barrett & Russell, 2014).
In any given learning situation, valence and arousal dimensions occur in combinations (Roth & Walshaw, 2019). High arousal activates F I G U R E 1 A representation of the circumplex model of affect. The horizontal axis representing the valence dimension and the vertical axis representing arousal (Barrett & Russell, 2014) the brain and this improves attention and memory. Valence has a complex impact on learning; that is, positive emotions can broaden attention and cognition, and negative emotions can promote deeper analytic processes and sharpen and detailed memory (LeBlanc et al., 2015).
A range of emotions is essential for engaging and completing an academic task. Emotions can influence the learning process in many ways by impacting cognitive resources, long-term memory, cognitive and metacognitive thinking strategies, and motivational processes (Artino et al., 2012). Emotions are related to the perceptions associated with academic activity (e.g. anxiety, joy or boredom) and to the feeling of success or failure after completion of an academic activity (e.g. pride or shame).
Emotions can be activated when students experience cognitive incongruity, for example, when students discover they are unable to solve a learning task problem and thereby experience surprise, frustration or confusion. Such emotions can have a positive effect on knowledge acquisition (Vogl et al., 2019). Emotions can also be elicited by the content within learning material, such as being sad if the manikin in the simulation dies. Because learning activities take place in a social setting, students can experience social achievement emotions such as shame, for example, if they demonstrate their lack of knowledge in front of their peers and teachers (Pekrun & Linnenbrink-Garcia, 2014).
Negative emotions such as stress and anxiety occur frequently in simulation settings and are thought to interfere with learning (Al- Ghareeb et al., 2017;LeBlanc et al., 2015). Limited research has identified positive emotions in simulation settings. One study explored students' emotional experiences during a simulation in continuing education and found that the students experienced mainly positive emotions such as interest, enjoyment of learning and cheerfulness both before and after the simulation (Keskitalo & Ruokamo, 2017).
Because emotions can be triggered by activity and social interactions, it is assumed that students can experience varied and multiple emotions during SBE. It has been found that emotional experiences can both inhibit and promote learning (Pekrun & Linnenbrink-Garcia, 2014). Previous research in SBE has focused mainly on the presence of student's anxiety (Al-Ghareeb et al., 2017). For educators to develop and deliver comprehensive SBE, further understanding of students' emotions and their effect on learning is needed. This integrative review was performed to summarize the current state of knowledge of emotions and emotions' effects on health professional students' learning SBE.

| ME THODS
Briefly, this review followed the integrative review methodology described by Whittemore and Knafl (2005). A multidimensional approach for searching the research literature by synthesizing quantitative and qualitative findings (Noyes et al., 2019). The approach included four phases to provide a full and comprehensive understanding of current research on students' emotions and learning in SBE.
The study protocol was reported in Prospero (CRD 42018107758).

| Study design
The review was performed following the four stages of Whittemore and Knafl (2005): problem identification, systematic search of the literature, evaluation and analysis of data.

| Search
The main electronic search was conducted in June 2018, using databases included in the Cumulative Index to Nursing and The inclusion and exclusion criteria (Table 1) were compiled after first considering the aim of this review; these criteria guided the final decision about inclusion in the review.

| Data evaluation and results of the search
The search results were imported from EndNote to the screening tool Rayyan. Four researchers were engaged in the screening process. We used the following process for quality assurance of the screening process. The research group had a common understanding of the study aim and the inclusion and exclusion criteria. All four re-

| Data analysis
Three authors (AM, KR and HSS) conducted the analyses individually. All authors met to discuss the data to ensure their credibility and to obtain consensus about the interpretation of the individual themes that emerged during analysis. The analytic process focused on identifying common and unusual patterns, variations, comparisons and relationships. This process allowed us to conduct the analysis to meet the aim of this review regardless of the research methodology or study design (Whittemore & Knafl, 2005).
We obtained comprehensive insights into the data using the following phases. First, reading the studies and especially the results provided us with an overall impression of the data and a sense of the whole. Second, the data were open coded and clustered thematically. We next compared the themes within studies to contrast patterns within each study. Finally, the themes were compared and contrasted between studies in an integrative matrix synthesis (Noyes et al., 2019;Whittemore & Knafl, 2005).

| Critical Appraisal Skills Programme
The Critical Appraisal Skills Programme (CASP, 2013) was used to appraise the methodological quality of the included articles systematically. To minimize bias, three researchers were involved (AM, KR and HSS). To perform an in-depth appraisal and to adhere to the methodological appraisals, two summary matrices were utilized: CASP for quantitative studies and CASP for qualitative studies (Tables 2 and 3 respectively). The studies were evaluated according to their presentation of the design, methods, ethical quality, analyses and discussion. The quality of the studies was evaluated to identify the invalid and highly valid studies. Table 4 describes the purpose, sample, methods and design used to report emotions and learning in addition to the study findings from the included studies.

| RE SULTS
The academic emotions described in the results were evoked by the simulation learning task itself. Emotions vary during the learning activity and were found to unfold from negatively to positively loaded. Three themes emerged from the analysis: (a) simulation as a fearful and stressful situation, (b) variability in emotions experienced during simulation as a rollercoaster of emotions and (c) emotions wide-ranging effects on students' learning in the simulation.
The research findings are presented in detail in the following sections.
Several studies highlighted the reasons for anxiety and stress associated with simulations (Beischel, 2013;Cato, 2013;Groot et al., 2020;Holt, 2017;Ko & Choi, 2020;Najjar et al., 2015;Tremblay et al., 2017;Walton et al., 2011). The students' different roles in a simulation setting trigger negatively loaded emotions. The active leading student role or being a team leader generates more stress, anxiety and cognitive overload than having an observer role (Beischel, 2013;Cato, 2013;Fraser et al., 2012;Schlairet et al., 2015;Tremblay et al., 2017;Walton et al., 2011) because students are put on the spot, must perform in front of teachers and peers, and are anxious about receiving feedback in the debriefing. Al-Ghareeb et al. (2019) Beischel (2013) Cato ( (2) Methods, sufficiently described, location and dates of data collection.

F I G U R E 2 Flowchart
(3) Data analysis, description of methods used for data analysis, verifying data, calculating the response rate.
(5) Research tool development, description of the population.
(6) Administration of tool, who approached potential participants, number of contacts provided.
(8) Results, response rate reported, respondents accounted for, results clearly presented.
(9) Discussion, results summarized, strengths and limitations, the generalizability of results discussed.
= High, all/almost all criteria met.
= Medium, some of the criteria are not met or not satisfactorily describe.
= Low, few or no criteria are met or satisfactorily described. (2) Was a qualitative methodology appropriate?

TA B L E 2 CASP Quantitaive
(3) Is it worth continuing?
(4) Was the design appropriate to address the aims of the research?
(5) Was the recruitment strategy appropriate?
(6) Were the data collected in a way that addressed the research issue?
(7) Has the relationship between the researcher and participants been addressed?
(8) Have ethical issues been taken into consideration?
(9) Was the analysis sufficient?
(10) Is there a clear statement of findings?
(11) How valuable is the research?
= High, all/almost all criteria met.
= Medium, some of the criteria are not met or not satisfactorily describe.
= Low, few or no criteria are met or satisfactorily described.

TA B L E 4 Summary of included studies
Authors

| Theme 2: Variability in emotions experienced during simulation as a rollercoaster of emotions
Emotions vary before, during and after the simulation (Tremblay et al., 2017;Walton et al., 2011). Anxiety, being alert, feeling insecure or disorganized and joking to mask fear are predominant at the beginning of a simulation. Stress, embarrassment, confusion, enjoyment and pride are evident during the unfolding of the scenario. The various reported emotional outcomes after the end of the simulation range from fear, anxiety, shame, frustration, confidence, calm, nervousness, disappointment, surprise and worry, to enjoyment, excitement, happiness, pride and feeling challenged (Behrens et al., 2019;Fraser et al., 2012;Fraser & McLaughlin, 2018;Groot et al., 2020;Holt, 2017;Ko & Choi, 2020;Schlairet et al., 2015;Walton et al., 2011). Students also report feeling a "sigh of relief" after the scenario has ended (Holt, 2017;Walton et al., 2011).
Invigoration was predominant and suggests that students can experience simulation as a pleasant and activating learning activity. Fraser and McLaughlin (2018) found that tranquillity seems to increase during the debriefing, which suggest that students feel comfortable and relaxed during the debriefing.

| Theme 3: Emotions wide-ranging effects on students' learning in the simulation
Results were ambiguous concerning how emotions affected students' learning. Studies that focused on the learning process show how emotion can impact students' cognitive load and that stress and a high cognitive load can influence students' ability to learn (Fraser et al., 2012(Fraser et al., , 2014Fraser & McLaughlin, 2018;Schlairet et al., 2015;Tremblay et al., 2017). Students who experience high cognitive load may miss important clinical observations (Schlairet et al., 2015).
Stress limits their ability to focus and make it difficult to think and do at the same time; for example, students report forgetting new knowledge and memory loss as a result of stress during simulations.
The literature show that stress and anxiety can negatively influence learning outcomes (Cato, 2013;Fraser et al., 2012;Fraser & McLaughlin, 2018;Tremblay et al., 2017;Walton et al., 2011); for example, by impairing performance and the development of competency. However, stress and anxiety can also benefit learning outcomes. For example, Demaria et al. (2010Demaria et al. ( , 2016 found that students exposed to emotionally challenging scenarios (e.g. by letting the manikin die) achieved greater practical competence than those who were not exposed to this scenario. Several studies found that moderate anxiety is experienced by students as a driver of learning (Al-Ghareeb et al., 2019;Behrens et al., 2019;Groot et al., 2020;Ko & Choi, 2020). Five studies identified compound learning outcomes such as confidence and feeling more secure, despite the students being emotionally aroused. Students have reported that they gain experience by practising clinical reasoning and critical thinking, and by learning to identify errors, and that the simulation experience seems to motivate them to improve and to direct their own learning (Groot et al., 2020;Holt, 2017;Ko & Choi, 2020;Najjar et al., 2015;Tremblay et al., 2017;Walton et al., 2011). For example, during a simulation, students learn to analyse the patient's situation and to prioritize and implement interventions (Walton et al., 2011). Holt (2017) described complex affective learning outcomes, such as empathy and confronting ethical issues.
Ambiguous results were found in the mixed-methods study by Beischel (2013). The quantitative aspect of their study found that anxiety does not influence the learning process, whereas the qualitative aspect found that anxiety prevents learning. Experience of anxiety causes slower thinking and has a negative effect on students' ability to think clearly. Walton et al. (2011) found that stress and anxiety can hinder the learning process even though the students reported complex learning outcomes.

| The emotional rollercoaster during SBE
Learning situations foster intertwined mixed emotions (Shuman et al., 2013) and this also occurs in the simulation context. Our analysis showed that during simulation students express emotions ranging from fear to excitement, enjoyment and pride. Students have reported feeling confident, calm, surprised, shamed, frustrated, proud and happy at the end of a simulation (Behrens et al., 2019;Tremblay et al., 2017;Walton et al., 2011). These findings support the idea that emotions are dynamic and that students can experience both positive and negative emotions in the same educational context Shuman et al., 2013).
In the debriefing phase where learning mainly occurs, negative emotions can shift to positive if students are given the opportunity to reflect on the scenario situation -what worked, what did not work -and to obtain supportive feedback (Dufrene & Young, 2014).
Such an emotional change can play an important role in students developing confidence about the skills needed to become a professional health-care worker.
Our analysis also show that students can experience emotional overload during SBE. In our analysis overload was defined as a high cognitive load. A high cognitive load has been found to affect working memory capacity by limiting the processing of novel information and working memory (Van Merrienboer & Sweller, 2005). Studies have identified that high cognitive load during simulation can reduce performance and memory months after the simulation (Fraser et al., 2012;Fraser & McLaughlin, 2018;Schlairet et al., 2015) which suggest that a higher cognitive load can also impact learning outcomes.

| Negative emotions are more than a barrier
Studies of emotions in education are dominated by studies of anxiety and stress (Pekrun & Linnenbrink-Garcia, 2014). This is also the case in the literature examined in this review because several of the included studies described a one-sided focus on simulation as a nerve-wracking experience that is loaded with stress, fear, discomfort and anxiety. Such emotions are categorized as negative and unpleasant according to the CM.
Most of the studies in this review reported anxiety and stress as hampering learning (Cato, 2013;Fraser et al., 2012;Fraser & McLaughlin, 2018;Tremblay et al., 2017;Walton et al., 2011). In academic situations, anxiety is perceived as the most problematic emotion for academic success. Anxiety impacts the learner's ability to focus, which negatively impacts working memory.
Research on anxiety in test situations supports the idea that negatively loaded experiences adversely influence cognitive processes by requiring use of the available cognitive resources to cope with anxiety, leaving few resources for memorizing, interpreting and solving complex tasks (Pekrun & Linnenbrink-Garcia, 2014;Valiente et al., 2012). This is supported by the findings of our review, in which stress and anxiety were found to increase memory loss, decreased the ability to focus on the task, and hinder thinking (Cato, 2013;Walton et al., 2011). However, our results also found that moderate stress and anxiety have a positive effect and can facilitate students' learning because the anxiety associated with SBE improves students' preparedness (Behrens et al., 2019;Ko & Choi, 2020). Demaria et al. (2010) and Al-Ghareeb et al. (2019) also found that moderate stress and low-level anxiety were associated with greater practical competency and optimal performance.
Incorporating some tension before a simulation, for example by requiring students to prepare in advanced, is supported by the simulation framework (Jeffries, 2020). Walton et al. (2011), Behrens et al. (2019 and Holt (2017) found that students can feel confused and frustrated during the simulation. Students face confusion and frustration when experiencing incongruity between cognitive capacity and performance and when their existing knowledge is inconsistent with new information (Vogl et al., 2019). However, such unpleasant emotions can benefit learning. Most learners want to overcome the feeling of confusion, and to do so, they engage and seek knowledge to resolve the feeling of incongruity. Therefore, confusion can increase students' deeper level of understanding (Lehman et al., 2012). In particular the opportunity to resolve their confusion during a simulation activity broadens the potential for learning (Fredrickson, 2001). Through the debriefing phase, learners are given the opportunity to explore and discuss problems experienced in the simulation scenario.
Our analysis also showed that, despite the students' anxiety, stress and discomfort, complex learning outcomes do occur (Al-Ghareeb et al., 2019;Demaria et al., 2010;Groot et al., 2020;Holt, 2017;Najjar et al., 2015;Walton et al., 2011). Students gain confidence and understanding, learn to analyse the patient's situation and increase their practical competency. According to the CM, both negative and positive emotions can activate students, and this activation can benefit the learning process by increasing attention, alertness, concentration and engagement in the learning task (Ainley et al., 2002). The results of our analysis show that students are highly emotionally activated during a simulation. Arousal can play a crucial role in students' learning and learning outcomes beyond the potential for pleasant or unpleasant emotions to interrupt the learning process.

| Enjoyment and relaxation foster learning in a simulation
Students' positive emotions include excitement, pride and enjoyment (Behrens et al., 2019;Groot et al., 2020;Holt, 2017;Schlairet et al., 2015). These are defined as activated emotions in the CM (Posner et al., 2005). Positive activated emotions can foster interest, and the motivation to study, and can contribute to the desire for students to make the effort to explore new knowledge. Interest helps students to maintain attention in the academic setting. Sharpened attention, increased effort and interest are beneficial in the learning process (Fredrickson, 2001). By being in a positive state, students can feel free to think critically and solve problems in flexible and creative ways (Pekrun et al., 2002). Solving problems and seeking understanding are important in a simulation. These actions are related to metacognition and deep learning approaches, which are associated with higher achievement scores (Pekrun & Linnenbrink-Garcia, 2014;Trigwell et al., 2012).
Students also described positive deactivated emotions, such as a "sigh of relief" and calmness after the scenario has ended (Fraser et al., 2012;Groot et al., 2020;Najjar et al., 2015;Schlairet et al., 2015), and these emotions are most prominent during debriefing (Fraser & McLaughlin, 2018). Students can benefit from being in a low-arousal state before entering the debriefing phase. Being relieved and calm can make the space needed for cognition and reflection, which are important for understanding and learning during a simulation (Neill & Wotton, 2011).

| Strengths and limitations
To ensure validity of our analysis, the research process involved four researchers, including a senior librarian for quality assurance of the database search. Our analysis was strengthened by the broad search conducted and the updated search to include the latest research.
Despite the common aims used in many studies, research on emotions and learning encompasses a complex field, and emotions have been investigated in various dimensions. Learning has been defined as an outcome or process, which meant that comparison between studies was problematic and complicated our conclusions.
Many of the studies included were conducted with a small number of participants, thus, it may be premature to draw conclusions about the topic.

| Relevance to educational practice
The usual recommendation for simulations is to reduce stress to increase the learning experiences (Al-Ghareeb et al., 2017;Cantrell et al., 2017). Our findings indicate a more complex and nuanced picture; that is, there is a wider spectrum of emotions occurring in simulation than simply students' stress and anxiety. The black-and-white perspective that students' stress and anxiety can have a negative effect on learning may be misleading. Our analysis shows that simulation can trigger multiple and changing emotions. Educators should consider the full range of students' emotions when planning and performing SBE. The need to establish a psychologically safe learning environment is relevant to SBE theory and is important for reducing student's anxiety (Kolbe et al., 2020). However, educators should also consider using SBE to trigger comprehensive emotions that can benefit students' learning such as curiosity, interest and confusion.
The emotional rollercoaster that students experience during simulations highlights the importance of time for reflection and clarification during debriefing. Not including adequate debriefing can mean that students leave the simulation with unresolved frustration and confusion. Therefore, debriefing should be considered as an essential part of SBE.

| CON CLUS ION
The main findings of our analysis were the strong focus on stress and anxiety in the literature and the broader spectrum of students' emotions uncovered. This review did not identify unilaterally empirical proof for promoting positively or negatively loaded emotions limiting students' learning during SBE. The role of arousal in learning needs further clarification. The benefits of experiencing a range of emotions during a simulation should not be overlooked because the change in students' emotions from fear and stress to enjoyment and less arousal can help to facilitate their sense of amazement and investigation. Students should be allowed the opportunity to express frustration and to resolve their confusion. This opportunity is best given during debriefing, which supports the idea that debriefing is essential for learning during SBE.
Our results disclosed that little attention has focused on the effects of positive emotions on learning. This coincides with research about academic emotions overall (Pekrun & Linnenbrink-Garcia, 2014). The reasons for the lack of research may be that students often are unaware of their enjoyment of learning and the difficulties observing and measuring positive emotions.
Because emotions are a highly subjective experience, future research should include more qualitative research to understand more fully students' learning during simulation. Research on the impact of emotions on learning during simulation should focus on how emotions change throughout the simulation, how activated emotions affect learning and how to scaffold and implement simulation scenarios to trigger academic emotions such as curiosity, frustration and surprise.

ACK N OWLED G EM ENT
The authors thank the anonymous peer reviewers for their feedback.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

E TH I C A L A PPROVA L
All authors provided substantial contributions to study design, analysis and synthesis of data. All authors have agreed on the final version to be published.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author.