“Case By Case”: Investigating the Use of a VR-Based Allegorical Serious Game for Consent Education

Abstract

The topic of consent within interpersonal relationships is sensitive and complex. A serious game can provide a safe medium for the exploration of the topic of consent. In this paper, we aim to alleviate the challenges of designing a serious game artefact with the implicit goal of exploring the topic of consent. The resulting artefact, “Case By Case”, is a VR-based serious game targeting university students, which uses an allegory-based approach to achieve its goal. The participants play the role of a detective who is tasked with determining if individuals have committed theft, which serves as an allegory for breach of consent. “Case By Case” provides the users an opportunity to reflect on their decisions within the game and apply them to the complex situations of consent such as victim-blaming and bystander awareness. To evaluate the effectiveness of the game in achieving its implicit goal, we ran a user study (n = 24). The results show that “Case By Case” provided a safe environment for the users to reflect on the concept of consent and increase their understanding about the topic further.

1. Introduction

The concept of consent in interpersonal relationships is a complex topic with modern day ramifications. Consent in the context of interactions of sexual nature presents its own unique challenge in modern society. The academic or legal definition of consent is straightforward, relying on clear, communicated and revocable permission to engage in sexual activity. The practical aspects of consent mediated through cultural norms, communication skills, and social settings complicates the establishment and breach of boundaries [1]. A direct approach to educating individuals, eliminating taboo around open discussion of the topic and creating supportive infrastructure for victims has been undertaken by organizations like RAINN [2], JAG [3], CARA [4], and even through art [5] and games [6].

However, the statistics on non-consensual incidents [7] indicate that this problem requires more concentrated effort to resolve. For example, in the UK, data shows that more than 1 in 4 women have been raped or sexually assaulted, 1 in 6 children have been sexually abused, and 1 in 18 men have been raped or sexually assaulted [4]. The ubiquity of internet connectivity has created new spaces in which consent takes a virtual form, with the very first incidents reported in 1993 [8]. The rise of social media has given a platform to problematic individuals who create more misinformation about consent within the minds of the young male audience [9]. This has further detrimental impact due to the absence of statutory relationships and sex education even in developed nations like the UK [10].

In this paper, we aim to address the objective of educating young adults about setting sexual boundaries, asking for consent, and interpreting body language and words associated with consent. The main challenge is the design of tools that can adequately handle the sensitive topic and at the same time avoid triggering survivors of sexual violence. Our design process results in the creation of an allegory-based serious game artefact “Case By Case”, which can be used to educate young university-going adults. Our artefact focuses on exploring the importance of consent, developing empathy for the reality of victim-blaming/shaming, as well as learning to identify when someone has not consented. A user-study based on “Case By Case” showed that the game produces the desired impact. Our contribution is therefore two-fold; an artefact aimed at consent education and a serious game design process that can address sensitive topics through allegorical means. We hope that this encourages future research into creative use of serious games for tackling delicate subject matters while alleviating potential for harm.

2. Background

2.1. Consent

To design an artefact which lets young adults explore the concepts of consent safely, we first explored aspects of consent that have been shown to work. For example, enthusiastic consent [10] has emerged as an useful concept in the last decade. Enthusiastic consent encourages individuals to correctly interpret and react to verbal responses, bodily cues, revocation of consent and even renewal of consent [11], while ensuring involuntary bodily reactions are not misinterpreted as enthusiastic consent [12]. Individuals, both perpetrators and bystanders, have been known to normalise and excuse sexual violence. This creates an environment of rape culture and victim blaming/shaming. Victim-blaming marginalises the victim, which makes it more difficult for them to seek support and justice [13]. While the topic of consent is extensively researched and there are many different directions to take, we focus on three aspects of enthusiastic content, viz., lack of enthusiastic consent, renewal or assumption of continued consent; and revocation or withdrawal of consent. We also focus on the bystander dimension, specifically the emergence of victim-blaming and rape culture. Research shows that education programmes across USA tend to avoid discussing the topic of sexual consent within their sexual education courses [14]. When the topic is addressed, it lacks sufficient or comprehensive detail to be useful to young people [15]. In many cases, it is limited to a “risk-reduction” approach [16]. Unwillingness to discuss the topic can be attributed to lack of creative methods that allow students to safely engage with the sensitive context of the topic.

2.2. Serious Games and Virtual Reality

Serious games are defined by Abt [17] as games with a carefully thought-out educational purpose, but not lacking entertainment value. They have served as learning tools as well as tools for changing perception through reflection [18,19]. Serious games can be designed to scaffold reflection leading to social change through the “specific frame of gaming experiences” [20]. A player can find it easier to relate to the game content without fear of consequences for their in-game actions, essentially creating a safe space to explore and learn about subject matter in a way that is not necessarily afforded in the real world. Since the emergence of affordable head-mounted displays (HMDs) like Oculus Quest, their value in training and education has been investigated repeatedly. Freina & Ott [21] identify that virtual reality (VR) HMDs allow users to have an immersive but safe virtual interaction with situations that can be dangerous. Evidence suggests that the improvement in learning is due to the immersion that VR HMDs enable [22]. Participants can still experience distress or anxiety similar to that in a real world situation [23]. With a topic like consent, we need to carefully balance between the immersive advantages versus the emotional response to use of the serious game artefact.

2.3. Serious Games Tackling Consent Education

Serious games tend to tackle a wide variety of subjects, including consent education. We found several examples of serious games that tackle themes of relationships and consent. This includes research-based artefacts that rely on formal design decisions: “Survive the Party” [24], “PR:EPARe” [25] and “YMCA” [26]. We also found additional artefacts which can be categorized as games for change (see gamesforchange.org). Examples include “Stuck in a Dark Place” [27], “How to Blorrble Blobble” [28] and “Vantage Point” [29]. These artefacts are presented as end-products but do not discuss a formal grounding for the design decisions. Our analysis of these games shows a common theme of direct exploration. The player is placed into simulated situations of non-consent [24,26,27] and takes on an active or bystander role in the incident. Alternatively, the player engages with the concept through post-incident interactions but with direct reference to scenarios involving non-consent [25]. These examples rely on experiential learning [30] which encourages reflection on actions and consequences. However, these games (except Blorrble Blobble [28]) present narratives, settings and story-arcs with explicit reference to sexual situations. The sensitive nature of the topic of consent and sexual assault as the material content of the game presents potential risks and ethical implications that game designers need to address especially when a research artefact is involved.

3. Design

3.1. Ethical Considerations and Allegorical Abstraction

The topic of consent withing interpersonal relationships is a sensitive topic. Any intervention has to explicitly consider the emergence of harm or triggering events in its implementation. Designers of the intervention need to ensure that harm and triggering events are alleviated or entirely avoided through proper design. If the designers fail to do this within the game design, it can result in actual incidents [8] or produce trauma [23] and harm [31]. Exposure-based experiences have been explored in depth through the medium of VR in the form of VR-based Exposure Therapy (VRET) [32,33]. However, the designers of VR medium-based serious games need to recognize that there is a greater risk of psychological harm [34]. This risk has to be considered carefully due to immersive nature of this medium [22].

From an ethical perspective, if an alternative approach can be less distressing while attaining the objective of the intervention, it should be preferred over direct exposure. The game-design can allow the player to skip potentially triggering scenes. “Stuck in a Dark Place” [27] includes a “classroom” mode that enables some separation from the triggering experiences. However, this does not limit the contact with the topic itself, rather avoids parts of the game only.

Keeping the ethical consideration in mind, we tailor our approach to explore the use of allegory through games the as a“presentation of the world in a reflective, playful mode” [35], while aiming for simplicity rather than complexity. Allegories can scaffold the communication of a complex concept through the simplicity of presentation [36,37]. This is in line with recent attempts to redefine consent education via allegorical devices while retaining the entertainment factor. “How to Blorrble Blobble” [28] uses an allegorical approach replacing consent with dance. The “Tea and Consent” video produced by the Thames Valley police in 2015 [38] is another good example of the use of allegory in the context of consent. While the “Tea and Consent” concept has seen criticism [39], we see it as an opportunity to explore how an allegorical representation can result in useful experiential learning through exploration in a safe environment which is devoid of direct and explicit reference to sexual situations.

Allegorical devices rely on simplified alternative experiences that can be used to reflect upon the subject of the allegory [35]. During the design of “Case By Case”, we identified that theft could serve as a useful allegory of consent. Theft as a concept is well-understood by most individuals and doesn’t require any background explanation in its simplest form. It also shares multiple overlapping themes with consent, such as violation of trust, boundaries and impacted by power dynamics. Thus theft serves as a good allegorical replacement for consent and sets up the context for reflection about consent once the player has experienced the game.

3.2. Case By Case’s Elemental Tetrad

We present the game design of “Case By Case”, through the lens of Schell’s Elemental Tetrad [40] consisting of Story, Mechanics, Aesthetics and Technology.

3.2.1. Story and Scenario Design

The player plays a detective in “Case By Case” and is presented with three different cases involving theft. The end condition of each case requires them to make a determination if theft occurred or not. Once the player solves the case by arriving at a determination, they are presented with the next one (see Figure 1). The player has to play through all the three scenes to complete the game. The setting for each “mini-game” is an office-like space. At the start of a new case, three non-player characters (NPCs) appear in the office. One of these NPCs acts as the accused while another one acts as an accuser. The third NPC, Detective Jay plays a supporting role to the player’s detective character. Each of the suspect and accuser NPCs will reveal details of the incident related to the theft as they are interrogated by the player’s detective character. In each situation, Detective Jay will offer their opinion supporting the accuser or the suspect. The dialogue is set up to reveal all the information gradually, and each piece of information requires the player to probe the NPCs through pre-formulated queries. This is intended to mimic how individuals, victim or suspect, are unlikely to offer up all the information in a structured and coherent manner during an interrogation. At each step, the player has the option to make a determination of theft or innocence. This approach helps simulate situations where individuals are likely to make judgements or form opinions without engaging with all the facts. Once the player makes their choice, either choosing that the crime of theft has occurred or not, Detective Jay reveals the real answer and summarises the situation and the implied relevance to consent.

Each case is designed to reflect the targeted aspect of consent.

• Assumption of continued consent: In the first case, the accused assumes that they are allowed to use the victim’s bank account as they were previously allowed to do so without consulting them first.
• Revocation of consent: In the second case, the accused continues to use the accuser’s phone even after they are forbidden by the accuser from using it.
• Inability to consent: In the third case, the accuser is inebriated in a social setting and unable to stop the accused from taking their bracelet
• Victim-blaming and rape culture: Detective Jay will take on the role of an individual engaging in victim-blaming occasionally. When this happens, their opinion is supportive of the accused. In the third case, they will opine that the theft was the accuser’s fault due to their inebriated state (i.e., victim-blaming) and mention that the bracelet was bound to be taken since the accuser did not say ‘no’ (i.e., lack of explicit withdrawal of consent).

3.2.2. Mechanics

We choose to limit distracting factors in the game-play by creating a seated experience, and trade-off some realism of a police investigation (e.g., statements would be recorded separately) for the purpose of focusing on the reflection aspect of the game. The entire flow of a single case occurs with the same scene with the player’s detective character and the NPCs seated around a table. The dialogue is implemented as interactive speech bubbles that appear above the NPCs’ heads (see Figure 2). The player interacts with the NPCs to trigger the interactive speech bubbles and other interactive objects within the scene to move the game forward. The dialogue tree is non-linear and consists of a branching structure based on the player’s previous responses. This encourages active learning-type behaviour rather than passive consumption of information. The summary of the scenario at the end provides an opportunity to reflect upon the immediate feedback and rethink preconceived notions.

3.2.3. Aesthetics

In selecting the aesthetic representation of the game, we choose relatively simple presentations that help situate the player into the narrative, but not overwhelming them with visual complexity and clutter. The detective’s office, in which the game takes place, has minimal furnishing, just adequate to establish the concept of an office space but without any distracting features (see Figure 2). The character design takes into account two factors. The first factor is to abstract out any confounds that can emerge from preconceived notions about gender-roles. The second factor is to provide a level of approachability and relatability to the characters. We avoid selecting human models as this is likely to create issues with the first factor and also complicate development effort due to need for more realistic behaviours from the models. Rather, we chose an anthropomorphic representation offered by Kawaii Slimes [41]. These cartoon-like characters make the game more approachable for a diverse audience and also shift the focus from character identities to the story.

3.2.4. Technology

The hardware choice was primarily rooted in the use of VR HMDs, specifically the Meta Quest 2. This helps us leverage the advantages of immersion offered by VR HMDs. Our implementation ran on the Meta Quest 2 because we had one readily available. However, the game does not rely on hardware-exclusive features and can be easily ported to a variety of HMDs and VR-ready PCs. The player uses the controllers in a ray-casting mode to interact with the interactive elements within the scene. This can easily be swapped out with any other input modalities, such as touch or mouse-based inputs, depending on the device being used.

4. Implementation

“Case By Case” was developed using Unity VR and deployed on the Meta Quest 2. Assets and character models used to populate the office-space scene were obtained from the Unity asset store or other free online sources. The core mechanic of the game is interaction with the NPCs through a branching dialogue narrative. This branching narrative was implemented using the Dialogue Editor [42]. The Dialogue Editor asset sped up the narrative design process and allowed us to test the narrative flow prior to deployment. To mitigate any unconscious bias from users based on chosen names of the NPCs and to promote inclusivity, we implemented a dynamic naming system on top of the dialogue narrative. This system randomised the names and pronouns of the NPCs in the game every time it ran, allowing us to present a different set of named individuals for each player.

5. Study Design

The aim of the study is to investigate if the allegorical device used by “Case By Case” is adequate enough to cause the participants to reflect on the topic of consent in a safe, indirect, but immersive, environment and increase their knowledge about the topic. This aim drives our research question, “Does the allegorical device used by a serious game produce quantifiable positive impact on individuals learning about a sensitive topic?” We use self-reporting to measure changes in individuals’ knowledge and perception about the sensitive topic of consent after exposure to (i.e., having played) the serious game “Case By Case”.

5.1. Participants

The target audience for “Case By Case” is university students, specifically those aged between 18 and 24 years. At Lancaster University, students are currently required to undertake consent training as a part of their “Ready for University”preparation. The current training is a formal course which tackles the subject directly with explicit references to potentially traumatic situations. We position the game as an alternative that complements the formal course as: a. a refresher on the topic; b. an alternative format to engage disinterested individuals; c. a less intense alternative for the interested but apprehensive audience.

We used convenience sampling to recruit participants for our study through email and social media promotions within the University circles. 24 participants were recruited for the study (15 male, 7 female, 1 non-binary and 1 preferred not to say). Nine participants had prior consent training either through the University or a Student Society. Before participants began the study, they were made fully aware of what the study entailed. We did this by ensuring they understood that the game is heavily abstracted, and the questions asked during the questionnaires were knowledge-based and generic to ensure that the participants are distanced from anything that could cause harm. These precautions were necessary since the topic of consent can bring up traumatic memories for survivors. The study was carried out after following standard procedure for ethical approval within the university.

5.2. Measures

The change in the participant’s knowledge and level of comfort with the topic of consent was measured through self-reporting based questionnaires. Each participant was asked to fill up the questionnaire (pre-hoc) consisting of 4 Likert-style questions and two open-ended questions. We avoided bringing up specific scenarios involving consent. In addition to the pre-hoc questionnaire, we also captured general demographic data like gender, field of study, and prior consent training if any. After the player finished playing the game, they were asked to complete the post-hoc questionnaire. The post-hoc questionnaire repeated the Likert-style questions from the pre-hoc questionnaires, scenario related questions, game related questions and open-ended questions that allowed participants to articulate what they learned through the game.

5.3. Procedure

The study was carried out on the Meta Quest 2 and a tablet was used to administer the questionnaires. The participant was provided with the experiment consent form and participant information sheet. Then they were allowed to familiarize themselves with the controls of the device and the game. Next, they filled up the pre-study questionnaire. After this, the participant played the game. Once they finished playing the game, the filled up the post-game questionnaire and answered the open-ended questions. The participants were not compensated for their time in any form and participation was voluntary with informed consent. Each session lasted a maximum of 30 min.

6. Results

6.1. Pre/Post-Hoc Questions

The pre-hoc questionnaire contained two Likert style questions asking about the level of comfort with consent and the participant’s ability to clearly define victim-blaming. The remaining two Likert style questions were aimed at identifying if the individual would be able to identify and intervene in a non-consent situation. The change in the participants’ responses is shown in Figure 3 and tabulated in

Table 1. Participant responses representing changes between pre-game and post-game play. Key: Comfort levels are represented as Comf−−: extremely uncomfortable; Comf−: somewhat uncomfortable; Comf+: somewhat comfortable; Comf++: extremely comfortable. Genders as M: Male; F: Female; NB: Non-binary; Unk: Not provided/withheld. Y: Yes (Y++ higher than Y+); N: No (N−− lower than N−).

Question	Response Category	Pre				Post
		Total	M	F	NB/X	Total	M	F	NB/X
Comfortable with topic of consent	Comf++	11	6	4	1	17	11	5	1
	Comf+	11	8	3	0	5	4	0	1
	Neutral	1	0	0	1	0	0	0	0
	Comf−	0	0	0	0	1	0	1	0
	Comf−−	1	1	0	0	1	0	1	0
Can define victim-blaming	Yes++	13	6	6	1	19	11	7	1
	Yes+	9	8	1	0	4	4	0	0
	Neutral	2	1	0	1	0	0	0	0
	No−	0	0	0	0	1	0	0	1
	No−−	0	0	0	0	0	0	0	0
Can identify non-consent situation	Comf++	7	3	3	1	13	7	5	1
	Comf+	14	10	3	1	11	8	2	1
	Neutral	3	2	1	0	0	0	0	0
	Comf−	0	0	0	0	0	0	0	0
	Comf−−	0	0	0	0	0	0	0	0
Comfortable intervening in non-consent situation	Comf++	3	3	0	0	9	7	2	0
	Comf+	11	6	4	1	11	6	4	1
	Neutral	4	2	2	0	2	2	0	0
	Comf−	6	4	1	1	2	0	1	1
	Comf−−	0	0	0	0	0	0	0	0

Counts in the Total columns equal sum of M, F and NB/X.

.

Overall, participants indicated a high-level of awareness, with minimal instances of low levels of comfort. Males showed improvement in their post-hoc responses across all four questions except for two individuals who reported decrease in confidence of being able to define victim blaming. The person who withheld their gender information also exhibited a similar response for the question. For females, there was overall improvement except for two individuals who reported decrease in the level of comfort with consent. To determine the statistical significance of our results for all demographics, we applied the Wilcoxon signed-rank test [43] and report our results indicating the median (Mdn) value for responses. For the first question, no significant difference was found between the pre-questionnaire (Mdn = Comf+) and post-questionnaire (Mdn = Comf++), z = −0.89, W = 15, r = −0.30. For the second question, participants’ post-questionnaire scores (Mdn = Comf++) were significantly higher than the pre-questionnaire scores (Mdn = Comf++), z = −1.89, W = 4.5, r = −0.67, p < 0.05. For the third question, participants’ post-questionnaire scores (Mdn = Comf++) were significantly higher than the pre-questionnaire scores (Mdn = Comf+), z = −2.67, W = 0, r = −0.89, p < 0.05. For the fourth question, participants’ post-questionnaire scores (Mdn = Comf+) were significantly higher than the pre-questionnaire scores (Mdn = Comf+), z = −3.30, W = 0, r = −0.88, p < 0.05. These results are tabulated in

Table 2. Questionnaire analysis using Wilcoxon signed-rank test. Sig.: p < 0.05; N.S.: p > 0.05.

Question	z	W	r	Significance
Comfortable with topic of consent	−0.89	15	−0.3	N.S.
Can define victim-blaming	−1.89	4.5	−0.67	Sig.
Can identify non-consent situation	−2.67	0	−0.89	Sig.
Comfortable intervening in non-consent situation	−3.29	0	−0.88	Sig.

.

The average of all definitions was 1.7 points with men averaging 1.6 points and women averaging 2.1 points. For the defining victim-blaming points were given if the definition included: holding victim partially/fully at fault, held at fault for harm/injury/misfortune/ crime that has happened to them, shifts blame from the perpetrator, prevents someone from coming forward, blaming the victim to feel safer, blaming the victim because the world is just so they deserved it, blaming the victim because they should have foreseen it, and invalidating the victims feelings. The average of all definitions was 1.8 points with men averaging 1.75 points and women averaging 1.71 points. For the defining bystander awareness, points were given if the definition included: being aware of your surroundings and noticing when something is not right, feeling responsibility for over people’s well-being, not falling victim to bystander effect, helping even when social pressure to not act is present, choosing to actively step in when someone’s behaviour is inappropriate, and understanding signs of distress or danger. The average of all definitions was 1.7 points with men averaging 1.4 points and women averaging 2.4 points.

6.2. Game Feedback

In the post-hoc questionnaire, we asked two Likert style questions on immersion and enjoyment of playing the game. The overall response was once again positive and in each case, only 5 participants rating the immersion below neutral and similarly another 5 rated the enjoyment as neutral. We allowed the participants to provide open-ended feedback on improvements. The repeating themes were inclusion of sound, better animation and more scenarios. However, participants agreed that they liked the concept, found the game informative and interesting.

To gauge the impact of the game as a learning tool, we also asked two additional questions (see Figure 4). The first one asked if the game would make the individual more interested in learning about consent. Once again, only five participants responded saying that they would not be motivated to explore the topic further due to the game. Two of the questions were designed to compare if the game was perceived to be better than other existing formal training. Figure 4b shows the responses of the nine participants who indicated that they had prior formal training. The trends are in favour of the game. Five participants indicated that the game is better and three indicated the game may be better and only one participant thought the formal training was better while finding the formal training itself to be moderately useful.

6.3. Questions about Determination of Theft

Overall, the participants were able to successfully complete the game answering the questions correctly with most participants getting every question correct. For the test scenario included in the pretest, all participants managed to select the correct answer and explain their rationale. Within the game, 2 participants selected the wrong character when making their final verdict about who was to blame in the first scenario, and 1 participant made this mistake within the second scenario. However, the questions were largely answered correctly with some participants noting that they had accidentally selected the wrong character when asked, but they knew who they should have selected.

7. Discussion

7.1. Study Results

Overall, the participants considered “Case By Case” positively. All the participants improved their understanding and confidence surrounding consent barring those with already high levels of understanding and confidence. There was a negative impact in only a handful of participants, which could be attributed to overall noise (the reduction was not drastic, rather maximum of two levels lower) or in an extreme case, these participants realised that their understanding of the concept of consent was either flawed or incomplete.

Majority of the female participants already felt confident in their knowledge of these elements of consent and thus there was limited scope for improving their knowledge. A potential inference is that women were already better informed about consent while men were not. The only topic that men already felt confident more than women was ‘intervening in a non consensual situation’. This could be due to men feeling more inclined to intervene due to feeling they need to be a ‘man’ [44].

For the definition questions, the averages were lower than expected. One explanation is that the participants didn’t attempt the question under ‘exam conditions’ and were relaxed in their response. Furthermore, we compared the participants’ responses to official definitions. It would be unrealistic to expect exact matches in wording. Women performed better at defining consent and bystander awareness than men did. While men only did slightly better than women did at defining victim-blaming. Men tended to over-explain their answers which added no value to their definitions while women were more consistent in their answers providing more information with less padding.

The main take-away from the study is that all participants felt that the game was safe to play and that it did not threaten their well-being or bring up any trauma while engaging their attention towards the topic of consent. This suggests that the allegorical device used by the game worked as desired.

7.2. Extensibility

The study answers our research question that allegorical devices can work. We address the conversation about more complex scenarios in future work. However, the interesting observation was related to how limited reach the existing consent training has. Out of the entire sample (n = 24), only nine participants declared that they had any form of consent training. This highlights a critical gap in the educational framework that the game can fill. The game was also considered to be a better alternative to the actual consent training by a majority of these participants with formal training. This is in stark contrast to only one participant reporting that the formal training was very useful. This also reveals that even formal training (including detailed and complex scenarios) still sees limited enthusiastic uptake. We contend that our solution is not a replacement to other formal training. Rather, it is best suited to be included within a suite of artefacts that deliver consent training. Finally, while we tackled consent as our subject-area, we consider that the allegorical device approach is well suited for creating opportunities for reflection for other sensitive and contentious topics. Serious game designers should consider it as an alternative when handling topics that have high sensitivity or when prior opinion is likely to be deeply ingrained.

8. Limitations and Future Work

8.1. Study Limitations

The study was run with twenty-four participants selected from a university environment and the age range of 18 to 24 years which we identify as our target audience. The limitation is that the results may not generalize fully to older audiences due to differences in exposure to consent education and technology. We also do not claim long term persistence in behaviour of the participants towards consent. This is similar to limitations noted in prior studies that investigate effects of serious games [18,45]. A detailed study with a large sample size is also out of scope with the current scale of the project due to practical limitations. The current results are encouraging enough to consider this for future work.

8.2. Character Choices

As part of our study we gathered a significant amount of user feedback on ways to improve the game. One of the requests was to include more human-like characters to make the game more immersive, but several participants commented that the characters were cute and enjoyable. With more wide-spread prevalence of VR HMDs, participants actually expect the applications or games to have a polished look and feel, reflecting outcomes of a commercial product development process.

8.3. Scenarios

The most requested improvement was to include a wider range of scenarios with more complex structures and topics, such as a scenario where consent is given along with some scenarios which are less black and white. While this can be achieved by tailoring the story-based scenarios, designers should be careful not to trade-off the benefits of the simplicity of an allegorical device in lieu of complexity. Perhaps a staggered approach that builds up the complexity as a part of an integrated meta-game reflection based in real-world discussions may be more prudent. In the future, we aim to work with experts tasked with developing consent training materials to identify more scenarios. With improvements to the stability and portability of the game, this approach of an allegory-based serious game has potential to be part of the consent training materials used by the University for educating students about consent.

9. Conclusions

In this paper, we explored how an allegorical device can be applied to the design of a VR-based serious game. The aim was to provide an alternative indirect but safe environment where players can explore the concept of consent without being traumatized or harmed. We showed how this could be applied through the design and implementation of the game “Case By Case”. We further evaluated the efficacy of the game through a study involving 24 participants. Results of the study showed mostly positive engagement with the game and improvement in understanding of the concept of consent. We also identified that while there is a good level of awareness of the concept in the target population of University-going young adults, the serious game is best utilised as a part of a portfolio of resources aimed at delivering training related to consent.