從面具到化身：媒介、展演與文化在VTubing媒介生態演化中的動態交織

From Masks to Avatars: The Dynamic Interweaving of Media, Performance, and Culture in the Evolution of VTubing’s Media Ecosystem

本研究探討VTubing（虛擬網紅直播）如何重拾蒙面展演（masked performance）並使其在主流媒介文化中煥發新生。研究採用綜合了展演理論與技術能供性理論的廣義媒介生態學方法論，結合文獻資料分析與線上田野觀察，從媒介技術、媒介化展演、文化與歷史脈絡三個層次，描述VTubing如何演出並探討其為何能（相對）成功全球化。
以Hololive所屬VTuber直播為例，本研究在分層描述VTubing現象後，首先指出即時動作捕捉所驅動的化身作為創新形式的面具，如何強化VTubing的展演功能，包括放大觀眾沉浸感、避免觀眾假戲真作、降低展演者面對觀眾的心理負擔等。這些功能使VTubing在串流直播生態中具有相對於非蒙面直播的優勢。其次本研究指出，VTubing作為蒙面展演的創新形式，根植於日本技術泛靈論的媒介文化，該文化偏好技術的擬人化，從動畫、虛擬偶像再到VTuber，皆可見其影響。
最後，本研究探討VTubing在當前媒介生態中的演化過程，指出其全球化的成功源自它能適應串流直播生態、搭上日本動畫全球擴張浪潮，以及掌握COVID-19疫情期間全球大封控的機遇。本文建議未來可探討VTubing的全球在地化變體，從而揭示它們在媒介生態中可能扮演的角色。

Research Background and Problem Statement
The COVID-19 pandemic fueled demand for contactless entertainment, spurring the rise of VTubing originating in Japan. By 2024, VTubing had secured a significant share of the global livestreaming market.
This research defines VTubing as a mediated performance where performers employ motion capture technology to control anime-style avatars for livestreaming. While the main distinction between VTubing and traditional livestreaming lies in the use of virtual avatars, this also raises several key questions: Why use avatars for livestreaming? How does the use of avatars affect the performers and audiences of the livestreaming? T
hese questions carry different meanings for different livestreaming participants. For those familiar with Japanese anime culture, VTubing is a natural extension of their previous media practices; for those unfamiliar, it appears difficult to comprehend. This contrast reveals the close relationship between media and media practices and guides this research to reexamine VTubing from a media ecology perspective.
From the historical and cultural perspectives of media practices, an avatar that conceals the performer’s true appearance and identity can be viewed as a new type of mask, and VTubing can be seen as a new form of masked performance. Therefore, VTubing is not an entirely new phenomenon, but rather embodies the retrieval phenomenon proposed in McLuhan and McLuhan’s law of media - masked performance, once obsolesced by mainstream media culture, has been extended through new media into VTubing and gained popularity once again.
This research further poses media ecological research questions: Why has masked performance, long obsolesced by mainstream media culture, been retrieved? In the current media ecosystem, how and why do avatars, streamers, and audiences collectively participate in VTubing as an innovative form of masked
Methodological Framework
To address the aforementioned research questions, this research first explores the broader media ecology, proposing a general methodological framework for analyzing mediated practices. This framework integrates multiple related theories, including Ihde’s technological phenomenology, Hutchins’distributed cognition theory, Latour’s actor-network theory (ANT), and McLuhan’s media theory. These theories examine from different angles how technology shapes human perception, practice, and culture, while pointing out how human usage patterns and cultural contexts in turn influence technological development and innovation. These theories share a non-individualistic methodology, emphasizing that human perception and practice are products of interactions between humans, technology, and environment, rather than solely originating from individual intentions and behaviors.
Second, this research explores Kaptelinin & Nardi’s theory of technological affordances. It focuses on how to concretely analyze the media level of mediated practices.
Finally, this research synthesizes Bateson, Bauman, and Schechner, discussing performance and meta-communication theories. It also concretely evaluates VTubing as a special form of mediated practice - mediated performance.
These theories collectively form this research’s methodological framework, in order to analyze three levels of VTubing: (1) Cultural and historical contexts, describing the existing forms of mediated performance from which VTubing has evolved; (2) Media technology, describing which media technologies VTubing relies on for mediation; and (3) Mediated practice as performance, describing how VTubers utilize these technologies to perform and how audiences appreciate these performances.
Research Methods
As it extends beyond the scope of direct observation, this research’s analysis of VTubing’s cultural and historical contexts necessarily relies on secondary literature, primarily including relevant industry reports and news coverage. For the analysis of VTubing’s media technology and practices themselves, this research employs a non-interventional observation approach in online fieldwork, looking at the livestreams of VTubers under Hololive, a global industry leader. The observation covers 70 Hololive members’streaming content from September 2017 to September 2024. Given the massive volume of streaming content (52,838 hours in 2023 alone), this research adopts an iterative sampling strategy guided by fan-curated highlights, which are then cross-verified with original broadcasts to ensure data reliability.
Research Findings and Discussion
After describing VTubing’s three levels through the case of Hololive livestreaming, this study finds that VTubing differs from unmasked livestreaming primarily in its use of avatars and the resulting aesthetic patterns. To explain why mainstream media culture has retrieved masked performance through VTubing, this research conducts further analysis from a media ecosystem evolution perspective.
First, from a technological affordances perspective, this research explores VTubing’s advantages in the mainstream livestreaming ecosystem. Avatars, as enhanced frame-setting messages, serve dual functions: enhancing audience immersion while clearly signaling fictional performance, thus preventing confusion between performance and reality and potential harassment. By concealing streamers’actual appearance, avatars provide both identity privacy and emotional labor protection. These features give VTubing distinct advantages over unmasked livestreaming.
Second, from a media culture perspective, this research clarifies that VTubers originate from Japan’s techno-animistic culture. This media culture is reflected in preferences for performance objects. Based on this culture, Japan continuously innovates performance objects through new media technologies, from traditional Ningyo Joruri puppets to modern anime characters, virtual idols, and finally evolving into VTubers. This explains why VTubers arose in Japan and why audiences from different cultural backgrounds have varying aesthetic perceptions of VTubers. VTubers not only retrieve the traditional function of masks as media for accessing transcendent worlds, but also expand these possibilities, creating an innovative performance form adapted to globalization - namely, VTubing.
Finally, by integrating technological affordances and media culture perspectives and employing an ecological evolution metaphor, this research traces how VTubing has evolved into a globally successful phenomenon through three key stages: VTubers emerged from Japan’s techno-animistic culture, adapted to YouTube-centered livestreaming ecology through advantages in performance functions, and expanded globally alongside Japanese animation during a period when the pandemic drove demand for contactless entertainment.
In summary, this research reveals that VTubing is not merely a technologically innovative mediated practice or cultural product, but rather a complex phenomenon constructed through equal interaction between human and non-human actors. Its success depends on the dynamic interweaving of media, practices, culture, and the contingent factor of quarantine measures. This research also establishes an analytical framework of mediated performance, providing methodological reference for studying other innovative forms of mediated performance.
Research Limitations and Future Prospects
While focusing on Hololive’s globally successful VTubing aesthetic patterns, this research overlooks other marginal patterns, particularly variants from global localization. Future research can examine these variants to enhance understanding of VTubing and to reveal its potential roles in media ecology.