Principal Findings
Using serious games for CPR training was a novel experience for all participants. Several benefits of this training tool proposed in the literature (eg, the focus of the cognitive part of the training and the added value of group practice) have also reoccurred among our four categories. However, our results also draw attention to other important properties of scenario-based MVW training when used for medical education.
The most important finding was the close relationship between the level of activation and degree of difficulty, and the level of engagement reported by the subjects. One of the commonly assumed strengths using serious games is the capacity to engage the player [11,19]. Our data support this, as all participants were positive about the experience and gave examples of their engagement in the virtual world training. This engagement could be an effect of the training tool itself, but might also be related to the seriousness and importance of the subject. Conversely, the inherent characteristic of this technology to give rise to engagement cannot be taken for granted, as pointed out by Choi and Baek [20]. Conceptual similarities exist between the construct of engagement and that of flow, as described by Csikszentmihalyi [21]. In his model of optimal experience, Csikszentmihalyi predicts a flow experience when the level of challenge and the level of personal skills are matched [21]. Based on our results, this model seems appropriate. The CPR scenarios were quite repetitive and the level of difficulty did not increase, so the subjects (presumably getting more skilled) tended to experience less challenge, and hence moved from a sense of anxiety and arousal towards one of control, relaxation, and boredom. This finding underscores the concept of leveling in gaming practice (ie, a progressive increase in difficulty to maintain challenge and motivation) that should also be considered in serious games.
Focusing on our psychometric data, mean concentration (a conceptually important component of flow) displayed a tendency to increase after the first training scenario. Triangulating these data with the students’ experiences could imply that part of the increase is to be ascribed to the increased proficiency of maneuvering in the virtual environment and overcoming technical difficulties.
The seemingly close relationship between activation and challenge with engagement might be more pronounced than that experienced in real-life learning. According to constructivist learning theories, as the learner gets more accustomed and proficient, he or she moves into more complex reasoning and deeper understanding [22]. Accordingly, this development might not be as pronounced in a virtual world, and engagement in a virtual world might be more dependent on novelty and challenge. The assumption that a virtual world offers a less rich environment, in which the altered perception in the learning space affects our ways of learning, should be further elucidated. Furthermore, the use of serious games for tasks that may be monotonous can be questioned, since one of the key features of this educational tool (engagement) may be lost.
During training, feedback on performance was mainly given by peers in the virtual world, and by an instructor in the real-world immediately after each scenario. Overall, this feedback was appreciated, but the participants asked for more direct feedback within the virtual world. The virtual environment provides possibilities that tantalize the user to test and vary (eg, actions, appearances, and surroundings). In particular, with such characteristics incorporated into the virtual world, it is not surprising that the participants, whether previously accustomed to computer games or not, seek immediate feedback. Serious games lacking such features might render less active experimentation, and possibly less engagement [20].
The training was characterized as being mainly cognitive. Participants were accustomed to traditional CPR training, and several trainees noted the lack of psychomotor skills training as an important insufficiency. During training, the level of mental strain was low to moderate; this was confirmed in the discussions, and lack of stress was seen as another unrealistic feature of the virtual world training. The low level of stress has the potential to create a more structured and optimal learning situation during the initial phase of training. However, without training in a stressful and complex environment, transfer to real-world CPR situations may be hampered.
Discussions with the participants demonstrated an attitude of high capability in bystander-CPR. This theme was reflected by an attitude of mastery and a demand for more complex and difficult scenarios; there were many creative suggestions for variation and increased difficulty. In a previous study, we found that medical students reported that a weaknesses of MVW-CPR was in the area of, “tasks too easy, more options wanted” [16]. These finding agree with the increase in self-efficacy beliefs. It can be argued that shifting from a clinical high-stakes environment to a simulated (at least in this case) low stakes training environment might alter self-efficacy beliefs. Such an effect would be predicted by Bandura’s ideas about cost of failure, in which increased social cost tends to lower self-efficacy beliefs more if performance does not meet the needs of the situation [23]. If situations shift from partly decontextualized learning and training to high-stakes clinical reality, problems could arise in terms of misconceived self-beliefs and attitudes. Based on this issue, some positive aspects of using virtual worlds for learning and training (eg, engagement and experimentation) might have a downside of inadequate attitudes about situational demands and individual capabilities, as discussed by Wang et al [6]. Taking such risks into consideration, using virtual worlds for training in medicine (and probably other high risk professions) should be addressed with caution, not only by the trainees, but also by teachers, instructors, managers, and developers.
Serious games have been mentioned as an attractive alternative for digital natives [19], although this argument has also been disputed [24,25]. In this study, all participants were between 20 and 30 years of age, and the level of previous videogame and computer game experience varied. There was no clear correlation between age and experiences in the virtual training environment. The participants that were most clear about the lack of realism in the MVW were among the youngest. Therefore, the common belief that younger people would ask for more (and more easily accept) learning in virtual environments, could be questioned. These attitudes are likely attributable to more factors than simply age alone.
One category highlighted in this study was related to the influences of the interface. Most participants held strong opinions about its effects. Sweden is a highly computer technology-developed country, and students are accustomed to using computers; the Internet is regularly accessed for private and educational purposes [26]. Despite this general trend, students that were less experienced with computer games often mentioned difficulties with navigation in the MVW, along with other technical problems, whereas students with more experience tended to be forgiving towards software deficiencies. Representatives of the latter group tended to compare the MVW-CPR training with previous gaming experiences, and provided suggestions about how to make the interface and software more compelling, immersive, and enjoyable. To some extent, the opinions of the less-experienced and more-experienced participants are contradictory. Making use of high-end technological solutions that create more options for interactions, with the goal of making the experience more realistic, complex, and unpredictable, could render a virtual world that takes longer to get familiarized with. Interest in technology and computer games will vary among different users in medical fields, so a balance between technical solutions and usability must always be considered [27]. Furthermore, in the partly-experimental setting of this study, technical support was readily accessible, and interest from the instructors was high. Using this kind of training tool in an average educational setting also emphasizes the need to keep such technologies manageable for the users.
General discussions about videogames and computer games indicate that sex differences is a common issue; females and males seem to be attracted to different game genres [28]. Males and females also spend different amounts of time in different computer-related activities. It has also been reported that videogames and computer games have different meanings to men and women [29,30], and that men and women are attracted by different features in virtual worlds [31]. Kron et al discovered sex differences concerning attitudes towards video games among medical students [1], although a simplistic gender view is questioned by others [32]. Addressing this issue was not an aim of this study, making it hard to draw any further conclusions from present work on this issue. However, there seem to be some differences between males and females in our findings. Among the subjects, more men had a history of using videogames or computer games: this likely explains the tendency for female participants to focus more on technical difficulties in the MVW. Conversely, this trend does not explain why the importance of the training (versus importance of the interface itself) was raised more often by the females. Male participants appeared to display a more relaxed attitude towards the seriousness of the topic, and showed an interest in discussing how the tasks could be made more challenging, and the technical solutions more appealing. This difference in the importance of CPR competence among future physicians is mirrored by our previous results that demonstrated differences in self-efficacy beliefs before virtual world training among males and females [16].
In medicine, insufficient teamwork skills have been identified as a common cause for suboptimal performance and harm [33-35]. This problem also seems to be true during CPR [36,37]. Our findings indicate that the participants appreciated the team focus, and the creation of an atmosphere of shared tasks and responsibilities seemed to be of importance for engagement in the scenarios. Not only was there a common belief in the strength of practicing a team endeavor in a team setting, but the students also considered it to be more fun, inspiring, and rewarding to train together. Inferring theories of situated learning [38], it is likely that positive experiences might occur due to the students learning with their peers. This matter, investigated by deNoyelles and Seo, might be of great importance for virtual world learning [39].
The participants trained in four short virtual world scenarios on two occasions. Several participants were aware of the rapid deterioration of CPR skills, and suggested MVW training as a good way to repeat and retrain. Distributed training using self-directed methods, such as a serious game technology, seems like an attractive option for retraining CPR.
Our group of informants consisted of a very homogenous group of early Swedish medical students, making it difficult to generalize our results. However, we believe that our findings can be applicable to other virtual world scenario-based training programs in medicine, in which clinically inexperienced (but at least moderately computer-experienced) users can train while reflective feedback is given. The strengths of this study lie in its novelty and ability to highlight certain characteristics of medical team training in virtual worlds, and how these MVWs are experienced. Methodologically we have also triangulated our results with psychometric process variables and previous results to reach better credibility [16].
Further investigations should involve studies of more heterogeneous groups, and include outcome measures focusing on behaviors and performance after training (transfer). Studies focusing on when and where virtual world training is particularly effective would also be of interest. The perceptions and experiences of instructors, teachers, and other stakeholders involved in this educational technology are also warranted. In our data there were many suggestions about how to create better and more challenging scenarios, and make use of more exciting technical solutions. Adding stressful elements, such as a real-time timing and scoring that reflect chances for successful resuscitation, unforeseen interruptions, or multiple emergency medical conditions might increase the experienced levels of stress, and could be included when the learner has completed the basic procedural and team-oriented steps. Technology is advancing quickly in this area, and it would also be interesting to understand which features would facilitate learning. There is a definite risk of being captured by the possibilities of this learning tool, and how it is used and varied among game developers. Instead, it is important to utilize experience-based and scientifically grounded knowledge in the field of learning psychology.
There are several limitations of this study. Our participants were 12 first year Swedish medical students that actively answered a call for participants. Adding more informants could have added more experiences and unraveled additional categories. The scenarios were all CPR-related and the virtual training environment was partly developed for this study, and lacked many features of real-world scenarios (eg, noises, bystanders). Although our study was designed for a rich input of data, the aim of understanding how virtual world team CPR training is experienced cannot be fully reached. Experiences depend on individuals’ subjectivity, how the teams are arranged, and the learning situation in which the study is performed.