Perspective Chapter: Identity in Metaverse

1. Introduction: Welcome to the world of “Social VR”

Social VR refers to online 3D virtual spaces where communication occurs through avatars, with immersion via VR headsets as a key feature. Approached as a new living space for humanity within virtual reality (VR), it is currently attracting significant attention. Notable services include VRChat (USA), NeosVR (Czech Republic), Resonite (Czech Republic), Cluster (Japan), and Virtual Cast (Japan).

Already, many users live their lives in this virtual world, where entirely new norms and cultures – unimaginable in the physical world – are emerging [1] (see Figure 1).

According to our survey [2], more than 90% of Social VR users are on “VRChat.” As of New Year’s Day 2025, VRChat reached approximately 130,000 concurrent users [3, 4]. Based on this and various other data, Nem estimates that there are millions of active Social VR users worldwide [2]. The global COVID-19 pandemic and falling VR headset prices have caused user numbers to surge in recent years, with a 30% increase in 2025 compared to the previous year [2–4].

However, the high level of anonymity results in the lifestyles and cultures of its residents remaining little known to the general public.

2. Over 5,000 responses from metaverse residents: The “Social VR Lifestyle Survey”

To shed light on the reality of Social VR life, Virtual Girl Nem, a heavy Social VR user, launched the “Social VR Lifestyle Survey” in 2021 together with anthropologist Liudmila Bredikhina (“Mila” in the VR world). Conducted annually and published online [2, 4–6], this survey is likely the first large-scale, platform-agnostic statistical study of Social VR residents worldwide. All survey procedures, full results (including those not cited in this chapter), and limitations have been made publicly available in the form of online reports. However, we present some essential information about the 2023 survey (see Figure 2) to facilitate understanding of the subsequent explanation of interesting findings.

• Survey protocol:

  Purpose: To clarify the actual lifestyles of Social VR users.

  Participants: English or Japanese speakers who have used Social VR with a VR headset at least five times in the past year. N = 2007 (Platform: VRChat 1,645, Cluster 146, NeosVR 104, Virtual Cast 94, ChilloutVR 3, Rec Room 3, Roblox 3, Horizon Worlds 3, Others 3; Region: Japan 1547, North America 221, Europe 115, Asia (except Japan) 106, Oceania 11, South America 6, Middle East 1).

  Method: Online public questionnaire.

  Promotion: Through media, communities, and snowball sampling.

• Limitations:

  Survey was conducted voluntarily by Nem and Mila, with respondents recruited anonymously in both Japanese and English through public calls. As such, several limitations should be noted. First, there is a possibility of non-probability sampling bias and heavy-user bias due to the open survey method. To minimize these risks, responses were collected from a wide range of users via various media sites and related communities, and the results were compared with those from two previous surveys (1,197 samples in 2021 and 876 samples in 2022) to check for extreme biases. In addition, language bias remains, as the survey was conducted only in English and Japanese. Sample size also affects sampling error; therefore, data analysis was conducted only for categories that met the pre-established minimum sample size (typically at least 70 samples, corresponding to a maximum error of approximately 12%). Other limitations include cross-sectional self-reporting, platform culture effects, and survivorship bias. As of 2025, many people worldwide still do not use the metaverse, and thus, as with any such study, it remains difficult to capture global diversity. Nevertheless, the significance of empirically investigating the lived realities in the Metaverse is substantial.

The survey goes beyond platform usage statistics to explore topics such as romantic relationships in the metaverse and “phantom sense” (sensations felt in VR despite not existing physically) [4]. The results reveal surprising aspects of metaverse life.

By 2024, the total number of responses had exceeded 5,000, with findings presented at academic conferences and at the United Nations’ Internet Governance Forum Kyoto 2023 (IGF Kyoto 2023), generating considerable attention.

This chapter presents the latest identity-related data from the 2023 survey and the 2024 survey. The 2024 study, themed “Identity in the Metaverse,” was conducted with Professor Ohyama’s collaboration as supervisor and included many open-ended qualitative questions [6]. For the qualitative component of the survey, we conducted systematic coding of the open-ended responses. First, a subset of responses was carefully reviewed by the research team to identify major themes, from which a preliminary category framework was developed. Two coders then independently applied this framework to the full dataset, after which their results were compared and reconciled through discussion. Additionally, from the perspectives of categorizations and diversity of responses, we have incorporated representative quotations. Responses were originally written in either Japanese or English; the Japanese responses were translated into English by the authors. The original Japanese texts are available in the source document of the survey.

In a VR world where you can freely choose your avatar’s appearance, how do users’ identities change? Let us take a look at the diversity of new lifestyles and self-concepts emerging in virtual spaces.

3. “Cosplay of identity” in the metaverse

Psychologist Erik Erikson described identity as a fundamental organizing principle developing throughout life [7]. He stated that feeling comfortable in one’s own body is one of the most important conditions for achieving optimal identity. Traditionally, identity was tied closely to one’s physical self – name, appearance, voice – and was something to “accept” rather than design by oneself.

In the metaverse, however, users can freely change all three – name, appearance (via avatars), and voice (via voice modulation technology). Unlike the physical world, where individuals cannot change identities with a couple of clicks, the metaverse allows for designing identity and even switching between multiple identities. Nem calls this concept “cosplay of identity” [1].

Just as cosplay enables people to transform into fictional characters in the physical world, metaverse avatars allow people to “wear” an identity as their “soul,” not just as clothing.

4. Avatars: Identity in the visual world

Among the three crucial elements of identity – name, appearance, and voice – appearance and visual outlook play a particularly central role in the metaverse, where they take the form of avatars. This section explores metaverse-era identity through avatars-as-bodies from various perspectives. For example, in the Japanese sci-fi anime Ghost in the Shell: S.A.C. 2nd GIG, episode 13, “Face” (2004), a woman, obsessed with possessing her beloved, steals the shape of his cyborg body and ultimately kills him to replace him entirely. Such stories resonate because appearance is not merely a “tool for tasks” but a marker of identity that proves one’s existence.

In the physical world, it is difficult for people to change their bodies, making life heavily influenced by body shape, physical sex, and other aspects of physical appearance – factors that underlie persistent prejudice and discrimination. In contrast, the metaverse will enable complete freedom of bodily design.

Who do you want to be in the metaverse? Face, height, body shape – all can be customized in the metaverse. The appearance need not match one’s physical body; one can live as male, female, or even nonhuman – robot, monster, or supernatural being. Through avatars, people can design their identity and live as their “desired self.” In metaverse societies, where communication occurs without reference to physical traits, prejudice based on bodily attributes may increasingly be seen as obsolete. Let us now examine what kinds of avatars people actually use in Social VR.

5. Avatar appearance: Both men and women choose feminine avatars

Among the visual traits of avatars, feminine or masculine avatar appearance is especially important since it is often regarded as central to identity in the physical world. In avatars, however, users can adopt any appearance, regardless of their sex in the physical world.

To distinguish between avatar appearance (masculine or feminine) and sex in the physical world, survey participants were asked: “Please state your sex in the physical world.” This was labeled as “physical sex.” This expression was chosen to avoid discrimination, respect diversity, and examine how avatar appearance is expressed in the metaverse, where diverse self-identifications may not be fully recognized by society. In the 2023 survey, when asked, “Does your avatar (the one you use the most) look masculine or feminine?” results revealed a striking imbalance: Regardless of physical sex, about 80% of respondents used female avatars. By contrast, only 8% of physical females used male avatars. Overall, about 60% of Social VR users used avatars of the opposite appearance (see Figure 3). “avatar appearance” implies an avatar’s visual characteristics, such as feminine- and/or masculine-looking avatars. Since we did not explicitly ask respondents about their gender identity, nonbinary persons may have answered as “others” or may have responded in terms of physical characteristics or sex assigned at birth. Therefore, in this survey, it is not possible to clearly distinguish these categories in the analysis.

By region, female avatars predominated across all areas – exceeding 60% in North America and Europe and reaching a remarkable 78% in Japan. By platform, VRChat and Virtual Cast exhibited particularly high proportions of female avatars (approximately 80%), whereas NeosVR and Cluster were closer to 60% (see Figure 4). These findings suggest pronounced cultural differences among platforms.

6. Living in the desired appearance

One of the first reasons that come to mind for why someone might use an avatar of the opposite appearance from their physical sex is incongruence between their mind and body. Figure 5 shows the results of the question: “If you use an avatar with an ‘opposite’ appearance from your physical sex (ex., you are male and use a feminine-looking avatar), please select the primary reason.”

On VRChat, 8% said they did so because they felt an incongruence between body and mind. Broken down further, this was cited by 25% of North American users and 13% of physical females. This finding is consistent with the presence of transgender individuals in Social VR at a certain proportion.

Preliminary laboratory experiments conducted by Ohyama and Hanashima provided additional indicative evidence supporting these survey results. Participants were asked about their spectrum of gender identity using the implicit gender identity test, and to use an avatar congruent or incongruent with their sex assigned at birth, and then rate their sense of body ownership (Does this avatar feel like your own body?). If gender identity is similar to their sex assigned at birth, incongruent gender avatars reduce the sense of ownership. However, when participants’ gender identity leaned toward the opposite sex, they reported a stronger sense of ownership with incongruent avatars. These findings are reported here only as supplementary context and should be regarded as preliminary until formally published.

These results suggest that transgender individuals can live more authentically when using avatars aligned with their gender identity. Thus, although “gender incongruence” is not the primary reason for adopting incongruent gender avatars, it represents a significant and effective use of the metaverse for fostering self-authenticity.

7. Reasons for the preference of feminine avatars

Across Social VR as a whole, 75% of the users who chose feminine avatars were physically male. Why do men want to become women in the metaverse? Our survey revealed three main reasons:

1. Want to be cute

   The first and most common reason can be described as a fashion-like desire to appear cute. The top reason given by physically male users for choosing female avatars was simply, “because I like the avatar’s appearance” (63%). Given that cute female characters are overwhelmingly popular in anime and other media, users could be interested in adopting popular cute appearances.

   In the physical world, it is not easy for men to cosplay as female characters, but in the metaverse, such limitations do not exist. Additionally, women generally have a broader range of fashion options than men, with more available outfits and accessories, which makes “becoming female to enjoy fashion” a frequently heard opinion. In other words, feminine avatars in the metaverse seem to embody a significant “fashion” dimension.

   Figure 6 presents responses from the 2024 survey, themed “Identity in the Metaverse,” highlighting the most frequent open-ended answers from physically male respondents about why they use feminine avatars. In the 2024 survey, as well, “want to be cute” emerged as a common response.

2. Needs for expressing emotions

   The second reason (27%) was that female avatars make it easier to express oneself and communicate. In cultures such as Japan’s, men are often expected to be stoic and unemotional. By using feminine avatars, men can more freely show their feelings without intimidating others. Feminine avatars, therefore, function as tools for emotional expression. From the open-ended responses in the 2024 survey, some male respondents mentioned a desire to escape from the pressures of conforming to socially stereotypical notions of masculinity.

3. Needs for getting closer to others

   Feminine avatars facilitate physical and psychological closeness in communication. Psychological studies show that women generally maintain smaller personal spaces than men, and women interacting with other women are the closest of all [8, 9]. In VRChat and Virtual Cast, “cuddly” female-to-female interactions (rare among men in reality) are common. This closeness makes female avatars effective tools for reducing interpersonal distance. Although no empirical data could be identified to substantiate this claim, the author Nem – who has been a heavy user of Social VR since its earliest stages worldwide – observes from personal experience that the use of cute feminine avatars for close-range communication and social touch has become common, even in everyday interactions (e.g., see Figure 7).

Thus, while gender incongruence explains some cases, most men use feminine avatars for self-expression and communication reasons.

8. Diversity of avatar species

Beyond gender, another striking trait is avatar type or “species.” In Social VR, users can be anything – schoolgirl, angel, giant robot, dog, or even an avatar close to one’s physical world appearance (see Figure 8). To study this diversity, researchers categorized avatars into seven species groups after field research and user interviews:

Humanoid type

• Humanoid – close to humans in the physical world (e.g., anime-style girls, photorealistic avatars).

• Semi-humanoid – human-like with added fantasy elements, such as cat ears, tails, etc. (e.g., cat girls, fox spirits, angels).

• Humanoid robots or cyborgs – partly or fully mechanical, human-like bodies.

Nonhumanoid type

• Animals – nonfantasy animals, nonhuman form.

• Plants – nonfantasy plants (e.g., leek, tree, radish).

• Monsters – Fantasy animal or plant, nonhumanoid (e.g., tentacled aliens, dragons, skeletons).

• Others – anything not fitting above (e.g., oscilloscopes, cars, nonhumanoid robots).

Users’ self-perceptions determined classification; for example, a cat-ear avatar could count as “humanoid” (if the ears are accessories) or “semi-humanoid” (if the ears are part of the body).

To obtain a statistical understanding of the distribution of actual avatar types used, we asked users, “What type of avatar do you use the most often?” in the survey conducted in 2023 (see Figure 9).

As a result, humanoid-type avatars dominated overall, accounting for 94%. Within this category, semi-humanoids led at 47%, surpassing humanoids at 41%. Humanoid robots or cyborgs made up 4%, while nonhumanoid types were rare.

Why are semi-humanoids preferred? Figure 10 shows open-ended responses regarding the reasons for choosing semi-humanoids. The results indicated desires to be “cuter” or to experience “specialness different from real life.” Fantasy traits such as animal ears or wings accentuate individuality.

Interestingly, although all the people are human-shaped in the physical world, the distribution of choosing humanoid-type avatars dropped below half in the metaverse. This suggests a tension: Avatars must be human-like enough for communication, yet users prefer not to be bound by real-life human form when expressing identity.

As shown in Figure 11, platform differences were also evident. NeosVR, for instance, had only 34% plain humans, while animal and monster avatars were more common, reflecting its “furry culture.” By contrast, VRChat favored anime-style humans.

9. Cognitive and behavioral changes induced by avatar use

Yee and Bailenson (2007) defined the Proteus Effect as the process whereby “an individual’s behavior conforms to their digital self‑representation independent of how others perceive them [10].” As with the Proteus Effect, it is known that the appearance of an avatar can influence the behavior and cognition of its user. For example, participants using muscular avatars in immersive VR reported approximately 16% lower pain on the Pain Assessment Scale compared to those using average-body avatars. The effect was strongest when avatar and participant gender matched, particularly for male users – illustrating how a muscular avatar can enhance perceived physical resilience [11]. In the case of cognitive change, the use of a dark-skinned avatar reduced implicit racial bias in light-skinned participants [12].

In the 2024 survey, respondents were asked, “Does your behavior in Social VR (such as the way you speak, react, and act) differ from how you behave in the physical real world?,” and 62% answered, “somewhat different.”

Open-ended responses revealed that people became more social and outgoing, adjusted their behaviors to match their avatars’ appearance, and felt liberated from physical-world roles. Some even changed their movement styles or speech patterns. Avatars act as mirrors of the mind, reshaping personality itself.

10. Influence of VR Life: 92% report positive effects

In the 2024 survey, we also asked respondents, “How does your life in Social VR generally influence your life in the physical world?,” and 92% answered, “positive.”

The reasons for these positive effects include the expansion of social connections, improvements in mental health, and the broadening of values and learning.

Negative effects cited included spending excessive time in VR, which reduced focus on physical life, sleep deprivation, and physical or mental strain caused by overuse or interpersonal conflicts. Thus, while VR life is overwhelmingly beneficial, it requires balance to avoid harmful side effects.

11. Do users hide their VR identity? 55% said “no”

When asked, “Do you keep your identity in Social VR hidden from your friends or family in the physical world?” in the 2024 survey, 55% said, “no” Results were similar across regions.

Reasons for hiding included fear of prejudice, generational gaps, and the desire to separate physical and virtual lives. Reasons for not hiding included viewing physical and virtual as connected and wanting to promote the metaverse.

This suggests that disclosure depends largely on personal values and circumstances, though many still feel that physical society is unready to accept VR identities.

12. Main identity in life: Which is the “true self”?

Respondents were asked: “If you have different identities in Social VR and the physical world, in the future, would you want your Social VR identity to become your main identity in your life rather than your real world identity?” in the 2024 survey.

Surprisingly, 13% answered that identity in Social VR is already their main identity, and 26% want identity in VR to be their main identity in the future (see Figure 12). Reasons for wanting VR as their main identity include being their desired self, better self-expression, and living more positively. Reasons against include preferring separation, valuing physical reality, or seeing VR as “just a hobby/game.”

13. Social and self-augmentation through the metaverse

Up to this point, we have examined how identity and communication are affected based on surveys of people who incorporate the metaverse into their daily lives. As of 2025, people cannot live solely within virtual space. As such, the lives of metaverse users will inevitably unfold within a cyber-physical society – a world in which cyber space and physical space are interconnected, or in which physical space is extended and augmented by cyber space.

At such a time, one might feel that the physical body – difficult to alter in accordance with identity – constrains authentic ways of living. However, we anticipate that society will not fall into a negative, dichotomous situation of “cyber space versus real space.”

Multiple academic studies have shown that using an avatar resembling one’s physical body increases the sense of body ownership – that is, the feeling that the avatar is one’s own body – more than when using avatars that do not resemble one’s body [13, 14]. On the other hand, as with the Proteus Effect discussed earlier, prior research reports that embodying a body that looks different from one’s physical self can reduce prejudice toward people with different physical characteristics [12].

Based on these academic findings, the relationship between the self in the physical world and the self in the cyber world can be diagrammed as shown in Figure 13.

Phase 1 represents the stage of beginning to use the metaverse, showing the influence of physical world identity and body on experiences with avatars in the cyber world. Physical world identity does not always align with the physical body. Using an avatar different from one’s physical body may initially feel more awkward than using one resembling the physical body. At the same time, in the cyber world – the metaverse – users tend to behave, to some extent, in line with the personality traits implied by the appearance of their avatars. See Section 15 for discussion on how people estimate an avatar’s personality from its appearance (body image).

Phase 2 represents the impact of using an avatar different from one’s physical body in the cyber world on physical world identity. Experiencing interactions in the metaverse through a body different from one’s preexisting physical world self can bring new perspectives and insights to one’s physical world identity as well.

Continued use of avatars in the cyber world reduces the initial discomfort felt toward them, and avatar-based experiences are perceived with greater reality. In this way, experiences in the physical and cyber worlds influence one another cyclically.

As phases progress further, users may begin to use different avatars across different cyber worlds or communities. In such cases, the self at that time is shaped within each world or community, while simultaneously influencing the physical-world self. Identity is inherently multifaceted, and avatars that allow easier expression of such facets may, in extreme cases, lead to transformations that could be called identity changes. However, in most cases, using a body different from one’s physical body allows people to observe themselves more objectively or to take the perspective of others, resulting in new insights about the self and others. This can be compared to entering a new school or workplace, or traveling to an unfamiliar culture: Encounters with new people and environments are stimulating and anxiety-provoking, but they provide opportunities for self-discovery and a stronger sense of connection to society.

The metaverse, with its potential to construct worlds far broader than the physical one, expands not only society but also the self. Thus, the expansion from a purely physical-world society to a cyber-physical society, including the metaverse, can be expected to augment the self and help realize a society in which people understand one another more deeply.

As mentioned at the beginning of this chapter, we argue that it is not appropriate to draw a sharp distinction between the existing physical world and the cyber world of the metaverse as “real” versus “virtual.” The reason is that the metaverse has already become part of people’s real lives.

Moreover, for those whose physical bodies do not align with their identities, life in the metaverse – using an avatar consistent with their self-identity – may feel more real than life in the physical world.

Humans receive information about the world through the five senses and the body as physical stimuli. Yet, the vast multimodal information we perceive is subject to selection, integration, and transformation by the sensory organs, cognitive processing in the brain, and top-down influences from past memory and learning. The world is recognized only after this internal cognitive processing. Thus, we do not observe the physical world objectively as it is.

For example, when several people travel together, each person’s “real” travel experience may differ. Conversely, when reading a novel, one may vividly feel the protagonist’s experiences as if they were their own. In this sense, the “real world” exists individually within each person’s brain or mind.

In the metaverse, both body and environment can be tailored to the world one imagines. Therefore, experiences in the metaverse can become as real – or even more real – than those in the physical world because they resonate directly with the reality perceived in one’s brain and mind.

14. The era of designing identity: How to create the best “self”?

What does VR identity mean to each individual? As we have seen, there are as many earnest answers as there are people. In this world, where a completely new lifestyle – switching between multiple identities – has only just begun to materialize, countless users are now struggling and experimenting in their own ways.

How should one design the self in VR? What avatar appearance should it be? What species should it be? How should one navigate between multiple selves and multiple relationships? How should one combine and balance VR with physical reality?

Challenges certainly remain, but the fact is that many users feel they can be more authentically themselves in Social VR than in the physical world. By analyzing the realities of these identities, we may have glimpsed a future where people can live truer to themselves – a future that is easier for everyone to inhabit.

Who do you want to be in the metaverse? The doors to the metaverse era are already open to you.

15. For the future of metaverse: Ethical and social perspectives

While there are various concerns regarding the use of the metaverse, many of these have not yet been sufficiently examined with scientific evidence. As authors, we wish to avoid unnecessarily amplifying such anxieties. Many of the ethical issues in the metaverse overlap with those already encountered in social networking services (SNSs). Although SNSs have generated new social challenges, they have spread widely due to their overwhelmingly positive value. Likewise, discussions about the metaverse should not focus merely on whether to use technology but rather on how individuals and society can experiment with positive ways of using it.

On this basis, we discuss one particularly important feature that differentiates the metaverse from SNS: embodiment. Ohyama, together with members of Japan’s metaverse industry, established a domestic committee for the international standardization of avatars. This committee has been discussing the minimum ethical and social considerations that should form a global common foundation for avatar use. Virtual Girl Nem, one of the authors of this chapter, participates as a representative of metaverse users. Key items identified as desirable considerations for metaverse systems, applications, content, or services are summarized in Table 1.

Item 1 may seem self-evident, but, as noted above, avatar “species” are highly diverse, and the first image users have of an avatar can vary depending on national culture or the worldview of a given platform. For example, a user who expects a photorealistic avatar may instead encounter an anime-style avatar. A user who expects a full-body avatar may instead find only the face and hands displayed. Since avatars are experienced as one’s body, it is recommended that users be given at least a general idea of what kind of avatar they will inhabit before the experience begins. However, it is not realistic to list every possible detail, but ensuring that accessible information is available at the outset regarding the type of avatar to be used is not difficult. At the same time, in cases where the purpose of the experience requires deliberately withholding information about the avatar to be used, prior disclosure is not necessary. Ohyama conducted an online survey about avatar’s appearance and personality image from their appearance in Japan. In the experiment, 300 women and 300 men in their twenties (Mean = 25.88; SD = 2.53) were randomly presented with figures of 10 different avatars as shown in Figure 14.

For each avatar, participants were asked about the personality impression they perceived from their appearance. Personality was assessed using the Japanese version of the Ten Item Personality Inventory developed by Oshio et al. [15]. Table 2 shows the means and variances of the personality ratings for each avatar.

The values in the table represent the mean, and the values in parentheses indicate the standard deviation (SD).

Although individual differences exist, on average, some avatars were perceived as more “agreeableness” but less “extraversion,” while others were perceived as more “openness.” This suggests that personality impressions derived from an avatar’s appearance are, to some extent, commonly shared among observers.

In this survey, we also asked participants to rate the degree to which they felt their own personality resembled the avatar’s perceived personality, and the degree to which their own appearance resembled the avatar’s appearance using seven-point Likert scale. Figure 15 shows a histogram of the differences between similarity evaluations of personality and appearance (personality similarity – appearance similarity).

In each graph, the bin between greater than –1 and up to 0 indicates the number of respondents who evaluated personality similarity as equal to or slightly lower than appearance similarity. For all avatars, this bin contained the largest number of respondents. This indicates that, in general, perceived similarity in appearance and perceived similarity in personality to avatars are felt to be roughly equivalent.

On the other hand, for about 15% to 30% of participants, the difference between appearance similarity and personality similarity was larger. In other words, some participants felt that although the avatar did not resemble their appearance, its perceived personality was similar to their own (the positive side of the bins in each graph of Figure 15). This suggests that some individuals may not feel their physical-world appearance matches their personality and, therefore, may feel more comfortable living in a body in the metaverse that better reflects their personality. Conversely, some participants felt that although an avatar’s appearance resembled their own, its perceived personality was dissimilar (the negative side of the bins in each graph of Figure 15). In such cases, using such avatars might lead individuals to behave more in line with the personality impressions conveyed by the avatar rather than with their own physical-world self.

Item 2 in Table 1 concerns the protection of personal information. In SNS and e-commerce, the secondary use of personal data has already been identified as a major issue. In avatar-based experiences, information such as body movement, gaze, facial expression, and voice is often collected in order to operate the avatar as one’s body. It is essential that users be able to confirm, whenever they wish, what kinds of information are collected and how they are being used, so that such information is not gathered without consent or repurposed in ways different from those agreed upon.

Item 3 is important from two perspectives: appearance and behavior. Regarding appearance, care should be taken to ensure that users are not forced to inhabit avatars with highly revealing clothing against their will. Regarding behavior, safeguards are needed not only against violent actions directed toward the user’s avatar but also in cases where a user’s avatar is made to perform violent actions against others. Recent research has reported that when an avatar is covertly manipulated from outside to display movements different from those the user actually performs, the user may mistakenly believe they themselves carried out those actions.

Item 4 extends to the creation and operation of avatars. The ability to become one’s desired self in the metaverse, without relying on one’s physical-world name, appearance, or voice, is a remarkable feature. Yet, this also makes malicious impersonation possible. Although not involving avatars, lawsuits have already occurred concerning digital characters. A well-known case is the character Pepe from the 2005 comic Boy’s Club, which was repurposed as discriminatory and violent memes far removed from the creator’s original intent, raising issues of image infringement [16]. With avatars too, problematic uses that contradict the creator’s intended character are emerging, and instances of avatars being sold that closely resemble specific existing avatars are often observed. In Japan’s international standardization committee, countermeasures under discussion include embedding usage conditions and information about the legal rights holder directly into avatar data formats. The crucial point, however, is to avoid imposing rules that would force users to adopt avatars resembling their physical selves in order to curb malicious uses. Importing such physical-world restrictions into the metaverse would undermine its potential to expand both self and society.

Furthermore, generative AI has made it possible to create CG models of public figures, such as prime ministers, and have them act in ways that defame their reputation without their consent. AI is increasingly capable of producing human-like behaviors that are indistinguishable in appearance from real people. While it is desirable for AI to support people as AI, when AI replaces real individuals in daily contexts and convincingly impersonates them as if they were real, unforeseen problems may arise. The functionality of AI has advanced to the point where, depending on the domain, it can already respond with greater speed and accuracy than humans. Yet, if AI interacts socially as if it were a real person, deeper discussion will be required regarding AI’s user interface and its role within human society.

Item 5 is one of the most important perspectives for the wider adoption of the metaverse. Until now, we have focused mainly on the metaverse as Social VR, but it is also valuable as a digital twin for education and training. In such applications, less realism may sometimes be more effective. As noted earlier, the cyber world can feel real when it approximates the mental image that users perceive as reality. Therefore, even with current VR technology, appropriately designed experiences can produce realism comparable to that of the physical world. For example, prior studies have shown that simulated virtual experiences can be effective for PTSD exposure therapy [17]. However, in educational content such as horror or accident simulations (e.g., amputation or falls) or other traumatic scenarios, excessive continuity with reality may pose psychological risks. A notable (though anecdotal) case reported strong stress responses from a user whose avatar was decapitated in VR [18]. As content realism increases, such risks may become more serious. To address this, Ohyama has proposed, as shown in Table 3, four levels of body reality tailored to the purpose of use, along with questionnaire items for evaluating them [19].

Level 4 corresponds to the question widely used in evaluating body ownership and indicates a state in which one feels that the experience is occurring in their own physical body in the physical world.

Level 3 indicates a state in which, while understanding that the experience is taking place in the cyber world different from the physical world, it is nonetheless felt as one’s own experience.

Level 2, by contrast, refers to the experience of playing a character in cyberspace that is not oneself. In Social VR, the sensation of communicating with others while using an avatar that does not resemble one’s real self – either as oneself or role-playing as a character imagined from the avatar’s appearance – corresponds to Level 3 or Level 2 body reality.

Level 1 indicates a state in which one does not act by their own will but instead feels as if they are reliving another person’s experience through that person’s body. Even in this state, as with watching a film, people can empathize with the characters and feel their experiences as real.

For safety-training content that recreates accidents or disasters, where psychological risk is a concern, it may be effective to intentionally design for Level 1 body reality rather than Level 2 or higher. In other words, the Level 4 body reality – long regarded as essential in VR research on body ownership – is not necessarily required in all VR content or metaverse experiences. Going forward, it will be important to deliberately design experiences to achieve the necessary and sufficient level of embodiment for the intended purpose. This variation in levels of reality, which cannot be designed in the physical world, will further enhance the experiential value of the metaverse.

Similar to the issues in SNS, guidelines will be needed to protect privacy and identities of minors or prevent impersonation; however, there is not sufficient global knowledge to clarify the situations, facts, and countermeasures in the use of the metaverse. In other words, the metaverse holds many possibilities, including uses that have not yet been conceived, and it is important to avoid constraining such possibilities without clear evidence. There remains room for discussion on what kinds of rules should be established and at what levels. At present, we have begun discussions with international experts within ISO/IEC.

Conflict of Interest

The authors declare no conflict of interest with respect to the research presented in this chapter.

Data and materials’ availability

The anonymized datasets from the 2023 and 2024 surveys, together with the full questionnaires in Japanese and English, are available at [insert repository name/URL]. Access to the data shown in this chapter is provided by contacting the authors directly. The raw survey data are not publicly available, as respondents did not provide informed consent for data sharing beyond the purposes of this study. Interested readers may contact the authors for further information about the survey design and analysis procedures.