Persona Studies 2020, vol. 6, no. 2

UNDERSTANDING MEMETIC MEDIA AND
COLLECTIVE IDENTITY THROUGH

STREAMER PERSONA ON TWITCH.TV

NATHA N J. JA C KS ON U N I V E R S I T Y O F N E W S O U T H W A L E S

ABSTRACT

Video game livestreamers on the leading platform Twitch.tv present a
carefully curated version of themselves - negotiated in part via interactions with
their viewers - resulting in collectively performed personas centred around
individual streamers. These collective personas emerge from a combination of live
performance, platform features including streamer-specific emoticons and
audiovisual overlays, the games that streamers play, and how they play them. In this
paper, I interrogate how these elements culminate in a feedback loop between
individual streamers and non-streamer participants, specifically how platform
features mediate and facilitate interactions between users. I also examine streaming
persona as both a product and expression of this dynamic and the subsequent
emergence of streamer-based social arrangement and collective value systems. I do
this with particular attention to how memes operate uniquely within the
livestreaming mode.

KEY WORDS

Livestreaming; Persona; Mimetic Media; Collective Identity; Performance; Video Games

INTRODUCTION

In this paper, I define memesis1 – not to be confused with mimesis – as the performative process
by which Internet users draw upon existing memes to create new memetic media. By examining
memesis through two contrasting Twitch streamer case studies – RayNarvaezJr and
PaladinAmber – I address the impact of different users on the formation of the streaming
persona and stream collective through questions of user agency. Further, I demonstrate how
memes appear within games and emerge from them during streams, thereby mediating game-
spectator dynamics. Finally, I introduce a temporal layer by linking memesis with the memetic
histories of stream collectives.

This analysis is rooted in streamers’ performances of persona, interactions with their
viewers, and the resulting collective identities that emerge through streams. By understanding
the creation of memes as a cultural process linked to collective identity formulation, a new
framework for examining digital persona arises. This paper demonstrates how multiple
phenomena that occur all over the Internet intersect: memes are built into our online
interactions; the sharing of video game content online has developed and sustained enormous
followings over the past fifteen years; and online identity – both collective and individual – is
constructed and performed online. The arguments I present here generalise well beyond this

Jackson

context to illuminate dynamics within digital culture around identity construction and content
circulation.

TWITCH, PERSONA, AND COLLECTIVE IDENTITY

Online persona can be best understood by recontextualising pre-digital notions of the concept.
Marshall (2010) begins this work by mobilising Jung’s (1928) and Goffman’s (1956)
understandings of persona as a product of interactions with an external collective in everyday
life. On Twitch, the streaming persona emerges and is reflexively defined through the
interactions between the streamer and spectators. Otherwise, also known as impression
management, this constitutes a particular form of micro-celebrity (Senft 2008).

As scholarly interest in Twitch grows, studies emphasise that sociality is one of the
primary motivations for continued use (Ask et al. 2019; Hilvert-Bruce et al. 2018; Sȷöblom and
Hamari 2017; Sȷöblom et al. 2017). Twitch operates as a social space through a combination of
the relationship between active spectators and the streamer-spectator relationship (Bingham
2020; Chen and Lin 2018). As such, participatory communities emerge with the streaming
persona at their core (Hamilton et al. 2014). The values presented by the streaming persona are
thus reflected in the behaviours of members of chat (Consalvo 2018), and I argue that this leads
to a collective variation of the streaming persona that is embodied by the streamer but enacted
by the entire collective.

Twitch’s features enable spectators to perform elements of the streaming persona.
Among the most prominent of these are subscriber emotes. Emotes are emoticons specific to
Twitch. Streamers can design/commission emotes for their subscribers, who in exchange pay a
monthly subscription fee to support the streamer. Emotes form a core component of both the
streaming persona and collective vernacular on Twitch (Consalvo 2017; Ford et al. 2017). When
spectators communicate using subscriber emotes, they mobilise the streaming persona to
create a collective identity. Persona’s collective dimension (Moore et al. 2017) emphasises that
the performer is not necessarily the sole contributor to the performed persona (Marshall et al.
2020). The synchronous nature of Twitch strengthens the perception of intimacy between users
(Johnson and Woodcock 2019), affording streamers a better sense of how content resonates
with their audience and enabling them to adjust their performance accordingly. In this way,
spectator agency is stronger in livestreaming than in asynchronous digital modes (Scully-Blaker
et al. 2017), and so collective identity and persona become negotiations between spectator and
streamer agency.

Moore et al. (2017) name the performative as a key dimension underpinning online
persona, a conceptualisation that foregrounds the balance between truth and fiction as a
performer communicates with their audience. An illusion of authenticity typically accompanies
the performance of micro-celebrity (Marwick 2013), although the liveness and long-form nature
of livestreams present challenges to a sustained and consistent persona. Despite this, Twitch
streamers must still tailor elements of their persona to appeal to their audience (Woodcock and
Johnson 2019).

MEMES, AGENCY, AND IDENTITY

The influence of collective agency on streaming personas and the personas’ subsequent fluidity
bears a striking resemblance to the transformative nature of memes. The transformations that
memes undergo subject to collective agency instil memes with an inherent plurality. Shifman
captures this plurality in her definition of an Internet meme as “(a) a group of digital items

Persona Studies 2020, vol. 6, no. 2

sharing common characteristics of content, form, and/or stance, which (b) were created with
awareness of each other and (c) were circulated, imitated, and/or transformed via the Internet
by many users” (Shifman 2014, p.41). To clarify what constitutes ‘many users’ in this definition,
Milner (2016) notes that media may be considered memetic if it is “consistently shared and
innovatively applied”, even among only a small group (p.38). A number of memes discussed in
this chapter resonate most strongly within a single streamer collective (in fact, may even be
understood completely differently across different groups), so I will rely upon Shifman’s
definition with the caveat that there may be limitations on spread.

Memes carry inherited social and political values that can contradict the intention of
individual manifestations (Glitsos and Hall 2019; Shifman 2019). These inherited values can be
mobilised by streamers to destabilise the existing values of the meme and evoke questions of
agency. Agency is essential to the success of a meme and underscores the distinction between
an Internet meme and Dawkins’ (2006) original “meme” (Milner 2016; Shifman 2013). Wiggins
and Bowers (2015) discuss agency in this context as “characterised by an innate ability to
imagine different outcomes”(p.1894). From this emerges their duality of memetic structures,
whereby an actor enacts their agency to create a new manifestation of a meme whilst
preserving the essence of that meme. This recursive process allows users to encode their
contribution with meaning relative to previous contributions and forms a basis for future
contributions. Phillips and Milner (2018) refer to this uniqueness-connectivity duality as fixity
and novelty, as discussed by Tannen (2007), and via Toelken’s (1996) twin laws of
conservatism and dynamism. This ambivalence is central to my account of memetic behaviours
and memesis. The creation of memetic media is thus a complex practice of layered
intertextuality that contributes to an individual’s identity within their social group (Shifman
2012).

Shifman’s (2014) social logic of participation names the simultaneous active
construction of a unique identity and shaping of social networks through memes. This logic
applies directly to social interactions on Twitch. Accordingly, a streamer can use memes to
define their persona in a way that is unique and relatable, and encourages greater collective
participation. There is a growing body of literature demonstrating how memes can be mobilised
to construct collective identity and perform boundary work that fluidly informs normative
behaviour (Ask and Abidin 2018; Gal et al. 2016; Literat and van den Berg 2019; Phillips and
Milner 2018). Alternatively, memes can act as gatekeepers, permitting membership to - and
hence validating the agency of – only individuals who demonstrate knowledge of unspoken
rules (Nissenbaum and Shifman 2017). Adhering to normative behaviour through memes – that
is, using the ‘correct’ memes in the ‘correct’ ways – lends users credibility. As such, memes
become tools for accruing social capital. Streamers operate slightly differently to non-streamer
participants here, as the latter generates this social capital within the stream collective while
the former sets the standard for the stream collective. However, the streamer also demonstrates
their social capital through memes as signals of broader social and cultural awareness.

METHOD

This paper stems from an ongoing ethnographic project examining the construction and
performance of streaming personas on Twitch. I draw upon two case studies – RayNarvaezJr
(Ray) and PaladinAmber (Amber) – who together demonstrate the diverse ways in which
memes are mobilised to reinforce and redefine aspects of streaming personas and the
associated collective identities. I draw attention to how these streamers curate their streams,
particularly how memes are integrated into streams and emerge from them. The elements of
interest include emotes, the stream screen layout, streamer choices in Twitch alerts and

Jackson

overlays, and the games that they play and how they play them. Taylor refers to much of this as
the “set design” of the stream (2018, p.73) and it culminates in streamers’ presentations of
persona (Sȷöblom et al. 2019). I extend this to more broadly consider memetic decisions that
affect the interactions between members of each stream collective and the subsequent
collective identity.

RayNarvaerJr
January 2019. Texas-based RayNarvaezJr (Ray) is a little way into his stream of the new Kingdom
Hearts III (2019). His facecam sits in the bottom left of the screen, a green screen sets only his
body, his chair, and his microphone visible. On the top left of the screen is an event tracker, listing
the usernames of his most recent four subscribers and donors. Above this sits his ‘sub train’ and
timer, respectively the number of subscriptions that have occurred within five minutes of each
other and a countdown until the train resets lest someone else subscribes. Ray is fighting a boss -
the Rock Titan from Disney’s Hercules (1997).

‘Oh, we up here now, okay’, Ray says as the protagonist Sora, who is dwarfed by the titan
boss, momentarily hovers above the titan’s twin heads before landing between them. After a few
seconds of attacking one of the heads, Ray activates an ability.

‘Wait, what does this one say?’

A bright, multi-coloured train of neons and fairy lights appears along rails made of the
same. As Sora jumps on board, Ray’s jaw drops and his eyebrows rise in surprise.

‘WHAT?! We’re in a train!’ his voice hits some higher notes as he fires projectiles at the
titan and the train chuffs along its rails of light.

‘Oh, hold on!’

He pauses the game. Chat is quickly filling with ‘CHOO CHOO’s and ‘WOO WOO’s as viewers
get in on the locomotive action.

‘Gotta do it for the bit’, Ray makes momentary eye contact with the camera as he pulls out
a wooden train whistle. As he puts it to his mouth, ready to blow, he looks briefly at it before
turning it around, ‘wrong side of the whistle. I’m excited.’

‘We good?’ he says with a nod, left hand holding the whistle to his mouth, right hand on the
controller.

He unpauses the game. Sora continues to fire projectiles at the titan from the train, the
screen filled with bright lights and firework-esque explosions, and Ray blows into the whistle. Chat
fills with Ray’s brownTRAIN emote - a small image of the driver’s car of a brown steam train. He
blows the whistle three times, each for four or five seconds. The fourth blow is cut short as he puts
the whistle down with a curt ‘That’s the end of that.’

It’s not quite the end, however, as the brownTRAINs continue to chug their way through
chat.

* * *

Persona Studies 2020, vol. 6, no. 2

Figure 1. A typical layout of Ray’s stream. He blows his train whistle while chat uses the
brownTRAIN emote (2019).

Ray2 streams a variety of games, and his streaming persona has naturally emerged from
his time with Achievement Hunter3. While working there, he produced gameplay videos for
distribution on YouTube. Marijuana memes have followed Ray, stemming from his insistence
that he would never smoke weed. Five years on from his start as a full-time Twitch streamer,
these memes still circulate on his channel. He engages with relevant memes through Twitch
alerts – streamer-specific audiovisual clips that play over the stream video to signify events like
subscriptions and donations – which allow spectators to enact their agency by directly
influencing the timing of these memes.

PaladinAmber
Eight months later, PaladinAmber (Amber) - streaming from Adelaide, Australia - has titled her
stream “lessons on ‘how not to be an idiot on the internet’ starting now”. Her stream screen
consists of pinks, purples, and soft blues with the appearance of a Windows 98 set up. Her facecam
sits inside a Paint window, with the game in a small Notepad window in the upper right-hand
corner, and her chat in a slightly larger Notepad window stretched across the lower and mid right-
hand side of the screen. Her pink headset with attached cat ears and pink-purple neons lining the
floor behind her reinforce the colour scheme. Her camera has been angled to include a large grey
Totoro plush (from the 1988 film My Neighbor Totoro). Her microphone reaches in from the right,
sitting in frame.

‘If you need rules on the Internet,’ Amber speaks over an audio clip of the Curb Your
Enthusiasm end credits, ‘Chances are you probably shouldn’t be on the Internet at all. Number one
to number ten is “don’t be a fucking idiot”.’

The stream cuts to black on the word ‘fucking’. The Curb Your Enthusiasm credits roll for
nearly fifteen seconds.

‘Holy shit...I’m in a mood. I’m in a mood today.’

In chat, someone expresses the belief that anyone should be able to say anything without
restriction, Amber immediately responds ‘Nope. I think that there are fucking restrictions. I think

Jackson

that you should most definitely, absolutely have restrictions,’ she begins using her fingers to count
the following as entries on a list.

‘If it’s harmful, if it’s hateful, and if it’s not helpful, don’t fucking say it. Absolutely don’t say
it. Absolutely don’t say it.’ As this comes to an end, she triggers another audiovisual clip meant to
emulate the ‘Confused Math Lady’ meme. Though typically an image-based meme used to connote
confusion (sometimes ironically), Amber has added the sound cue of a dial-up tone while
mathematical text and images float in front of her face.

As the conversation moves on a few chat members support her comments. They say that it’s
her stream and she can restrict whatever she wants. But they’ve missed the point.

‘It goes beyond the stream. We’ve gotta stop doing this to people who are in public eyes.
“He was comparing your hair”,’ she reads a comment, ‘Yeah but this is the thing though, out of
everyone he chose, he chose [to compare me to] Weird Al Yankovic, who is not the necessarily the
most attractive looking male I’ve ever seen in my life, and he didn’t say “hey, your hair is giving me
Weird Al Yankovic vibes”, he was most certainly talking about my whole physical appearance, and
if you think I’m wrong...’

The Math Lady meme has been playing for the second half of this. As it continues, she pulls
her microphone close to her mouth.

‘I have some fuckin’ news for you: you’re absolutely wrong. You’re absolutely wrong. Do
you want to know how I know this? I’m a woman, and I have suffered through years and years of
males telling me things and thinking that it’s appropriate, and it most certainly is fucking not.
Alright.’

‘Just don’t do it. Stop comparing people to people, it’s weird,’ after reading a few comments.

A new beat, and Amber is dancing in her chair and speaking quite quickly. The stream feels
upbeat, but Amber doesn’t call upon any memes to make her point.

‘I, honestly, at this point, I just...I love creating content, but here’s the thing,’ she moves the
backs of her hands together and takes a deep breath, interrupts herself reading some comments,
and finally says ‘I’m a very strong individual, and I’m like - my tolerance level - my tolerance level is
like -’

She brings her index finger and thumb together, signifying her dwindling tolerance.

‘Ever so thin at the moment with social medias.’

* * *

Amber4 became a Twitch partner following a series of viral Tweets in 2019. These
Tweets consisted of short stream clips, within which she would confront misogynistic messages
from viewers, including questions like ‘Are you straight? Bi? Single?’ or fetishistic requests to
show her feet for donations. The popularity of these clips stemmed from her use of memes and
generally humorous approach towards calling out the behaviour. Her relationship with memes
extends to producing merchandise proclaiming herself as ‘a meme queen’, and this appears
primarily through carefully selected audiovisual clips that Amber manually triggers in response
to stream events, including in-game occurrence and messages from viewers.

Persona Studies 2020, vol. 6, no. 2

Figure 2. A typical layout of a ‘Just Chatting’ segment of Amber’s stream (2019).

Through these two case studies, it becomes clear how deeply intertwined streaming
personas and memetic media can be. The case studies reveal how shared stream culture and
associated social practices facilitated by memes contribute towards the sociality that
encourages continued stream participation. From this active participation emerges the
collective persona, giving streaming personas themselves a memetic quality. Subsequently,
questions of agency within streams arise. Streaming personas are negotiations of streamer and
spectator agencies, although the balance between the two is not fixed. While the streamer
appears to maintain control, their memetic offers must be accepted by other members of the
collective in order to maintain their audience.

This paper is structured around a discussion of this fluid collective agency, beginning by
separately examining moments when spectator agency is prioritised and then when streamer
agency is prioritised. The final two sections emphasise the truly collective nature of streams by
focusing on instances when agency is more balanced, by explicitly considering the significance
of video games and temporality to memetic practices.

MEMESIS

In the previous section, I described moments from Ray’s and Amber’s streams that may seem
unrelated at first. However, these moments resonated with each streamer’s audience for a
reason deeper than a call to memes. Memes simultaneously become signals of the streamer’s
digital literacy and the values associated with the streaming persona. These memetic moments
are simultaneously serendipitous and planned-for, unique and repeated. This ambivalence
motivates my definition of the term memesis, a term designed to capture the unique aspects of
identity formation and sociality performances arising from the use of memes.

Memes are loaded with meaning and values, and when called upon, these blend with the
context under which the meme appears to create new meanings. In livestreams this process –
memesis – is rendered visible, as the liveness and interactive nature of Twitch streams deny the
time required to create new memetic media. Further, the text-based chat (where links are often

Jackson

prohibited) discourages common meme formats. Thus, memes propagate differently within
livestreams to other settings. Any prepared content, like alerts or emotes, must be versatile
enough to maintain relevance in a variety of scenarios. By focusing on the subsequently visible
process of selecting and employing particular memes, I use memesis to decentre the meme as a
product. This also serves to highlight memes’ social functions as virtual signals of social capital,
and contributors towards collective identity and values.

A portmanteau of ‘meme’ and ‘mimesis’, memesis is the act of creating something
memetic. The conceptual and linguistic relationship here begins with the birth of the word
“meme”, which Dawkins derived from “mimeme” (2006, p. 192). In the context of Internet
memes, Shifman emphasises memes as mimetic through the memetic practice of remixing
(2014, p. 22). Memes are “concrete speech acts” (Nissenbaum and Shifman 2017, p.498), and
therefore their creation is a performative act. As such, I coin memesis here to name the
performativity engaged by drawing upon existing memes in order to create new memetic
media, either through new manifestations of existing memes or entirely new memes.

Figure 3. A collection of Ray’s Twitch emotes.5

Through the remainder of this paper, I demonstrate how memesis operates through
Ray’s and Amber’s streaming practices. These streamers have been chosen as illustrative
examples of practices that occur across the platform, and this paper acts as an introduction to a

Persona Studies 2020, vol. 6, no. 2

wider conceptual project. In this section, I examine memesis as it defines Ray’s and Amber’s
streaming personas and the associated collective identities, as well as how memes function as
generators of social capital within their collectives. In both cases, I refer to memesis within
livestreams as a two-step process – preparation and execution – and compare how it shifts in
nature based on whether either streamer or spectator agency is prioritised. The execution can
be enacted by either the streamer or spectators, and while most streams – including Ray’s and
Amber’s – see a combination of both, the discussion that follows highlights the contrasting
dynamics underpinning memesis in this second stage.

Ray and Memesis Performed by Spectators
One of the core tenets of the production of memetic media is the potential for many people to
engage with and produce their iterations of a meme. Memesis that spectators perform
prioritises spectator agency over streamer agency. This type of memesis is prominent in Ray’s
streams and is most clear when he makes a memetic offer in response to a stream or game
occurrence, which is taken up by chat members. In terms of the two steps of memesis,
preparation is performed by the streamer, but execution is in the hands of the collective
audience. This social acceptance is an essential component of memetic diffusion (Spitzberg
2014), and feeds into memes as social capital.

In the occurrence described earlier in this paper, Ray responds to the in-game
appearance of a train by pulling out and blowing on a wooden train whistle (Figure 3). While he
uses this whistle when trains appear in games, its primary use is when the sub train5 becomes
large. Not only is the sub train a memetic stream feature, being incorporated into many popular
streams, but trains are also associated with the ‘hype train’ meme – a phrase for collective
anticipation and excitement. So, when Ray blows on this whistle, he makes a memetic offer,
which is accepted by the collective when they respond with brownTRAINs (Figure 3) in chat.
This chat reaction becomes an indication of collective hype.

Given that subscriptions require payment (of at least US$5) and brownTRAIN is an
emote accessible only to Ray’s subscribers, the hype is simultaneously a celebration of collective
growth and Ray’s economic success on the platform, as well as the contribution of his viewers to
that success. The sub train becomes a cooperative effort where viewers time their subscriptions
to keep the train going for as long as possible. Consequently, a strengthened sense of community
is facilitated by the streaming persona and expressed through that persona using the
brownTRAIN emote. The emote becomes a signal of collective membership and a memetic
expression of the streaming persona. The ever-present timer on Ray’s stream capitalises
Twitch’s liveness, which is already tied closely to the economics of the platform (Johnson and
Woodcock 2019; Partin 2019).

Going a step further, in 2020 Twitch introduced a feature called the Hype Train -
undoubtedly named memetically. Upon receipt of a variable number of bits6 and subscriptions
within a particular timeframe, the Hype Train begins. Viewers ‘build hype’ by donating and
subscribing while a timer ticks down. Despite demonstrating disdain for Hype Trains (referring
to them as ‘scam trains’), Ray has crafted a specific overlay for Hype Trains. This overlay
includes Vengaboys’ “We like to Party!” on loop and a cartoon graphic of Ray and his wife riding
a train across the screen. Memesis here again links the economics of the platform to
performances of persona as members of the collective collaborate to trigger and maximise the
Hype Train. The choices that Ray makes in how he performs the Hype Train are important,
considering ‘scam train’ discourse. To fully engage in a practice transparently designed to
control users’ spending patterns without acknowledging his own complicity would be in poor
taste. However, the exaggerated over-commitment reads as a memetic parody, which generates

Jackson

social capital. Hype Trains explicitly raise questions around financially gatekeeping
participation and elicit a sense of how users perform in exchange for capital, and how memes
relate to social and economic capital.

Another meme-theme that has become part of Ray’s streaming persona, and that will be
revisited in more detail, involves weed. One of Ray’s emotes, brownBLAZE (Figure 3), depicts
Ray pretending to smoke a joint - the pretence being central to Ray’s engagement with weed
memes, which originated from the claim that it was something he would never do. Memes are
dually concrete speech acts and vernacular (Nissenbaum and Shifman 2017) and emotes echo
this social function. Gestural emotes like brownBLAZE are virtual speech acts, allowing
subscribers to participate by virtually performing the gesture in response to other weed memes
appearing within Ray’s stream.

When streamers bring memes into the live(streamed) setting of Twitch, and they are
taken up by the collective viewership, the memes become part of the streaming persona
expressible by both the streamer and the viewers. The streaming persona operates as a
negotiation between streamer and spectator agencies. This is, in part, inherited from that same
negotiation that occurs through memesis. Thus, the streaming persona becomes a product of
collective agency. In fact, Scully-Blaker et al. (2017) observe this tendency when they identify
Twitch as a site of “tandem play” (p. 2026) between the streamer and spectators. By merely
being, the audience contributes to the stream, and as spectators become more active, this
impact becomes greater. Through memesis, the streamer and the collective work in tandem to
produce a memetic streaming persona, as is the case with Ray and his use of trains and weed.

Amber and Memesis Led by the Streamer
Amber exemplified streamer-led memesis by focusing heavily upon a range of memetic
overlays. Many of these overlays are mobilised to promote desirable values among the stream
collective by addressing and correcting undesirable behaviour. In this way, Amber’s memesis
operates as an instructive tool for moderation that ultimately leads to a self-moderating
collective. She employs new memes and existing memes, some of which occur uniquely in
response to messages from members of chat, while others are employed in response to game
occurrences or things that she says herself. Memes punctuate Amber’s stream, producing a
persona that resists hegemonic masculinity and misogyny on the platform (and within game
culture more generally), whilst also performing femininity in ways that operate both within and
against stereotypical models of gender performance.

The moment presented in this paper’s methodology is one of many instances of Amber’s
integration of memesis into her streaming practice. Amber has used memes to respond to
viewers questioning her sexuality and relationship status and to men warning her that she
ought to start a family before her eggs ‘dry up’. These moments share memesis as moderation
and collective value setting. The general strategy of education as moderation is effective (Cai
and Wohn 2019), with reactive bans being short-term deterrents (Seering et al. 2017). Through
an education-first approach, Amber allows offenders to amend their behaviour before she
removes them, in the process creating entertaining content out of breaches and building social
capital among viewers through a willingness to give second chances.

There is an ambivalence surrounding the boundary work that Amber’s overlays
perform. Namely, the undesirable behaviour must continue to provide stimulus for the memetic
performances that discourage it. In other words, the boundary work itself is visible only because
the undesirable behaviour that it explicitly rejects persists. On the other hand, since Amber
maintains control over the preparation and the execution steps of memesis, her agency is

Persona Studies 2020, vol. 6, no. 2

prioritised, and she can opt to attend to whichever comments she chooses. Looking again at the
introductory example, chat comments both supported Amber and contradicted her. This marks
an important separation between Ray and Amber, as in Amber’s stream the values of the
collective (and streaming persona) are being demarcated both through the roles of ‘insider’ and
‘outsider’ within the collective.

In the introductory example, Amber uses two of her memetic overlays. The first involves
about ten seconds of the Curb Your Enthusiasm theme song, after which the screen cuts from her
facecam to the credits of the show. This meme trades on the show’s awkward tone to emphasise
similarly awkward or cringe-worthy moments. She uses this overlay to address cringe-worthy
comments from viewers, as well as particularly bad failures in games. The second is a
manifestation of the Confused Math Lady Meme (usually a reaction meme of a woman
surrounded by mathematical text used to convey confusion) and Amber has created a live
version of this meme with mathematical text moving over a close-up (Figure 4). These
manifestations are strongly rooted in the referent meme, clearly communicating Amber’s
intended tone when memesis is performed.

What is particularly striking about Amber’s use of these memes is how they have
become part of her stream vernacular. She integrates them seamlessly into the stream, and they
become part of her persona. However, she is also selective about their use. They punctuate the
stream, providing a rhythm for viewers to follow. When she finds a point that she wants to hit
home, Amber speaks without drawing upon these memes. Removing the intertextuality of
memes, as well as their function as punctuation that increases the pace of the stream, creates
the impression of more direct and meaningful streamer-spectator communication. This happens
at the end of my example introducing Amber, where the lack of memesis implies greater
authenticity. Not only does the use of memes inform the streamer-collective relationship, but
the lack thereof also bears significance. In her work on race and digital gaming spaces, Kishonna
L. Gray has identified that the mere presence of people of colour in these spaces is read as an act
of deviance in the face of their hegemonic whiteness (Gray 2014, 2017). This deviant status can
be extended to anyone falling outside of the default (white and male) user category in online
gaming spaces (Chess 2017; Taylor 2018). In this case, Amber’s status as a woman immediately
categorises her as deviant, and she is doubly so when she explicitly mobilises memes to reject
the platform’s hegemonic masculinity.

This use of memes as stream vernacular is also significant as it relates to the distribution
of social capital within the livestreaming format more broadly. Although the preparation stage
of memesis occurs asynchronously, the execution must be synchronous. This separates the
mode from others such as YouTube videos, or posts on platforms like Twitter and Instagram.
There is an improvisational quality to memesis within livestreams, and so a streamer must be
sensitive towards what resonates with their audience and what does not. While most viewers
are unlikely to notice an occasional memetic misfire – accidentally triggering the wrong
audiovisual overlay can receive a positive response – a pattern can damage the social capital
that successful memesis has accrued.

Another example of Amber’s memesis demonstrating her deviance is her use of
‘simping’. Simping, one of many examples of misogyny on the Internet, is a pejorative term
describing a man (simp) going out of his way to accommodate the (emotional) needs of women
purely for the sake of sex or a relationship. Amber has co-opted the term, maintaining the core
meaning but altering the connotations. She uses the term positively as a way of sharing platonic
love and admiration. This represents her work as a streamer, albeit on a smaller scale: she
identifies problematic terms and gestures, and attempts to reframe them more positively. Here,

Jackson

this means changing the diminutive interpretation of simping to an uplifting one. As of
December 2020, Twitch has banned the word “simp”, however Amber’s subversive approach
still bears significance to the culture of the platform. By prioritising her agency in the execution
of memesis like this, Amber can explicitly perform her persona and produce an associated
collective that rejects the dominant (cis-hetero white male) values of the platform.

Figure 4. An example of the Confused Lady (top) and a screenshot of Amber’s variation
(bottom).

MEMESIS AND VIDEO GAMES

With the strong focus on video game livestreaming on Twitch, videogame play is deeply
embedded in the persona of the streamer and the associated collective identity. During play,
memesis occurs both when memes are integrated into game play and when they emerge from
game content, both reflecting and contributing towards the streaming persona. As the game
provides stimulus external to all users, memesis here allows both streamer and spectators to
enact their agency at different times.

W33DGOD (pronounced weedgod) is a recurring avatar of Ray’s that demonstrates how
the integration of weed memes into his persona has sprouted new memes. In January 2016, Ray
played Pokémon Red Version (1996) on stream. As part of this game, the player collects a team of
Pokémon to accompany them along their journey. When Ray caught the plant Pokémon
Vileplume, he nicknamed it ‘WEEDGOD!!!’ as recommended by a member of the chat. Two
months later, when creating his customisable character in Stardew Valley, Ray named him
‘W33DGOD!!!’ and ended up with a template for W33DGOD: a green afro, sunglasses, a tuxedo
top, and red pants. These elements culminate in a unique memetic expression of Ray’s persona.
The visual dissonance signifies Ray’s sense of humour, and the green hair and name allude to

Persona Studies 2020, vol. 6, no. 2

weed memes. The tuxedo is a throwback to an avatar he used in Minecraft videos with
Achievement Hunter, and the red pants are based on a pair that he physically owns (as he states
when first creating the character). This meme is a communication tool intrinsically bound to
Ray and as it spreads, so does his persona. This character has since appeared in many different
games Ray has played (Figure 5) and has spread beyond Twitch. Ray’s fans have produced
W33DGOD fanart, an example of memetic diffusion (Figure 6). Through this fanart, Ray’s
followers signal their identification with Ray’s streaming persona and collective.

Figure 5. Appearances of W33DGOD in Pokémon Red Version (2016), Stardew Valley (2016),
Dream Daddy (2017), and South Park: The Stick of Truth (2016).

Figure 6. Fanmade appearances of W33DGOD on DeviantArt8 and Twitter9

Jackson

While Ray has created a new meme, Amber has integrated existing memes from her
stream into her playthrough of Animal Crossing: New Horizons (2020). In this game, the player is
tasked with inhabiting a deserted island, eventually developing it into a bustling
neighbourhood. It is highly customisable, and Amber has regularly enacted memesis through
the choices that she made. She named her island ‘simptopia’, a haven for simps under her
alternative connotations of the term. Further, when tasked with creating an island tune - an
eight-beat sound clip that plays when speaking to villagers - she recreated the Curb Your
Enthusiasm theme by ear (Figure 7). By incorporating these elements of her streaming persona
into her play, the gameplay becomes a more collective experience, even if just one person is
playing.

Simping made another appearance in Amber’s stream when she played Among Us
(2018) with viewers. The game’s premise is straightforward: players are crewmates on a
(space)ship and must move around the ship performing simple tasks to keep it running. A
designated number of players are randomly chosen to be imposters who kill crewmates. Once a
body is found, the players have a conversation to deduce who is an imposter. A vote then takes
place, and if one person is voted for by the majority, they are ejected into the vacuum of space.
During one round, a viewer followed Amber around the ship. This spooked her, so she called an
emergency meeting and had the rest of the crew vote this person off the ship. It turned out that
this player was not an imposter, and so Amber apologised profusely to them, saying that they
were ‘just simping’. Memesis is naturally incorporated into Amber’s streaming persona and it
strengthens the streamer-spectator bond. However, this instance unveils a potential issue with
this kind of memesis. If a viewer was not familiar with the recontextualisation of the term
within this collective, the reading of this moment would shift substantially. Under this alternate
reading, the joke’s subject goes from Amber’s misreading of the viewer’s play to this viewer’s
‘simp’ status. This indicates that the collective identity is also defined by familiarity with the
streaming persona, and fluency in the stream vernacular.

Figure 7. Amber finalising her island theme in Animal Crossing: New Horizons (2020).

Persona Studies 2020, vol. 6, no. 2

MEMETIC HISTORIES

As seen from the array of examples of memes used by Amber and Ray, memesis is highly
influential in the creation and evolution of collective identity on Twitch. Some of the memes in
this paper have been consistently present for a year or longer, others see fewer references, and
others have disappeared entirely from the consciousness of the stream collective from which
they emerged. This temporality echoes the general behaviour of Internet memes, varying
drastically from the longevity associated with Dawkins’ original conception (2006).

This leads to the concept of memetic history. The memetic history of a stream is the
history of memesis within an individual stream collective’s memory. This became significant to
me when I was present for a reference to a joke that appeared in a couple of streams months
earlier. I understood the joke because not only had I been present for its initial telling in the
stream, but also witnessed its retelling in subsequent streams. My membership to that stream
collective felt solidified by that sense of shared history between myself and the other members
who shared in the joke. The notion of memetic history adds an important temporal layer to
memesis as a method of active participation and collective identity. It suggests that intensities of
viewing and membership duration contribute to a connectedness within the collective. Twitch
viewers that engage more frequently and over a longer period develop a richer sense of the
values at the community’s core. Conversely, this complicates the nature of stream collectives, as
it can often mean that any period of absence creates a disconnect between viewer and
collective.

Examining a streamer’s memetic history can give a sense of the trajectory of the
streaming persona - identifying how it has evolved and how this evolution is a product of
interactions between members of the collective. By looking beyond the memes that appear
within streams to their longevity within the collective, the streaming persona’s characterisation
emerges. For instance, Amber’s memetic history suggests a collective centred on her identity as
a female streamer, as well as an overarching feminist agenda that supports women’s right to
exist on the Internet without harassment or sexualisation. On the other hand, Ray’s memesis
tends to emerge through game content. There is a level of serendipity to memesis through
gameplay, which creates a higher turnover and echoes Ray’s content as a variety streamer. This
contrasts with his more constant memes such as W33DGOD and his alerts. These produce a
fixed core identity, anchoring an otherwise fluid streaming persona.

Memetic histories also elicit a rhythm of change if this fluidity is traced over time. For
example, following a viewing of Bee Movie (2007), Amber latched on to the memetic phrase “do
you like jazz?”, used by the bee protagonist as a hypothetical ice breaker. She integrated this
into stream titles and spoke the line during streams. This memesis became most apparent
during an overhaul of her Twitch alerts in early 2020, replacing several alerts with audio clips
from or related to the movie. The relationship between Amber’s stream collective and memesis
is challenged here, as although she prepares the alerts, members of the collective are required
to initiate them. However, this has significantly contributed to the collective identity of the
stream through its multiplicity and the ability for all members to control its appearance. In
August 2020, Amber changed these alerts for the third time that year, moving from Bee Movie
(2007) to Finding Nemo (2003), and then to Monsters Inc (2001). When she launched the newest
alerts, she emphasised that she was always looking for alerts that everyone could participate in
– usually through a single-word message like ‘BEES’ – but would also continue to rotate them to
keep things fresh. In this way, Amber has initiated a transition from memesis solely initiated by
the streamer to a combination of both streamer- and spectator-initiated memesis tied to the
same source material. It has also created a cycle that demarcates periods in the memetic history

Jackson

of the stream. Ultimately, memetic histories emphasise the experimental and ever-changing
nature of the Internet. Content creators are caught in the paradox of keeping their work fresh
whilst maintaining the core elements that have drawn their audience to them. Some
experiments fail, social capital is lost, and the collective identity is destabilised. Others succeed,
accruing social capital, and solidifying the collective identity.

CONCLUSION

In this paper, I have examined how memes are mobilised on the livestreaming platform Twitch
to facilitate the construction and performance of a curated streaming persona and the
associated collective identity. Through my concept of memesis and a close analysis of two case
studies, I have developed a framework for this examination tied to negotiations of user agencies,
the video game in question, and the notion of memetic history. My analysis draws out how
livestream spectators connect through memetic media, a central facet of contemporary digital
culture. I found that memes provide excellent avenues for decentring the streamer and thus
allowing active participation from spectators. At the same time, memes can also operate as tools
for moderation and expressions of collective values.

Although a novel concept, memesis and the associated digital identity construction
provide insights into livestreaming and Internet culture more broadly. I have demonstrated
how livestreaming’s unique features render memesis visible, however memesis occurs
whenever memes are created. As a cultural process, memesis can be mobilised to understand
how values are purposefully spread through memes online. Further, it provides insights into
how Internet users understand themselves in relation to others that they share virtual space
with. Other forms of digital persona, such as micro-celebrity and influencer culture, can be
understood through this lens. In the broader context of game studies, memesis can be carried
over to asynchronous modes of game-centric sociality. Examinations of memes embedded into
recorded gameplay videos (on for example YouTube), or shared through virtual community
settings like Discord, would benefit from an analysis through memesis. While discussed here
through Twitch, the trifold link between memes, collective identity, and digital persona is ever-
present, and this paper is one step towards better understanding this arrangement.

END NOTES

1. Other definitions do exist, for example https://digicult.it/internet/memesis-
community-and-self-definition-in-the-age-of-memes/#_ftn1, however these have
not gained much traction.

2. https://www.twitch.tv/raynarvaezjr

3. https://www.youtube.com/user/AchievementHunter

4. https://www.twitch.tv/paladinamber

5. Current as of March 2019. Image from
https://twitchemotes.com/channels/85875635

6. The number of consecutive subscriptions within five minutes of each other.

7. Twitch currency used for donating.

8. https://www.deviantart.com/heart0fink/art/W33DGOD-657419352

9. https://twitter.com/royallymad/status/1032441892707946497

Persona Studies 2020, vol. 6, no. 2

ACKNOWLEDGEMENTS

Enormous thanks to my supervisors, Bryoni Trezise and Brigid Costello, who have seen this
work through several iterations, and to Mark R. Johnson for his guidance and feedback. Thanks
also to editors Katja Lee and Chris Moore for their consideration and flexibility during a
challenge year, and to the anonymous reviewers for positive and thoughtful advice.

This research was supported by an Australian Government Research Training Program
Scholarship.

WORKS CITED

Among Us 2018, video game, Innersloth.
Animal Crossing: New Horizons 2020, video game, Nintendo.
Ask, K & Abidin, C 2018, ‘My life is a mess: self-deprecating relatability and collective identities

in the memification of student issues’, Information, Communication & Society, vol.21,
no.6, pp.834–850.

Ask, K, Spilker, HS & Hansen, M 2019, ‘The politics of user-platform relationships: Co-scripting
live-streaming on Twitch.tv’, First Monday, vol.24, no.7.

Bee Movie 2007, film, Paramount Pictures.
Bingham, CM 2020, ‘Talking about Twitch: Dropped Frames and a normative theory of new

media production’, Convergence, vol.26, no.2, pp.269-286.
Cai, J & Wohn, DY 2019, ‘What are Effective Strategies of Handling Harassment on Twitch?

Users’ Perspectives’, in Conference Companion Publication of the 2019 on Computer
Supported Cooperative Work and Social Computing, pp.166-170.

Chen, C & Lin, Y 2018, ‘What drives live-stream usage intention? The perspectives of flow,
entertainment, social interaction, and endorsement’, Telematics and Informatics, vol.35,
no.1, pp.293-303.

Chess, S 2017, Ready Player Two: Women gamers and designed identity, University of Minnesota
Press.

Consalvo, M 2017, ‘When paratexts becomes texts: de-centering the game-as-text’, Critical
Studies in Media Communication, vol.24, no.2, pp.177-183.

—2018, ‘Kaceytron and transgressive play on Twitch.tv’, in K Jørgensen & F Karlsen (eds),
Transgressions in Games and Play, MIT Press, pp.83-98.

Dawkins, R 2006[1976], The Selfish Gene: with a new Introduction by the Author, Oxford
University Press.

Dream Daddy 2017, video game, Game Grumps.
Ford, C, Gardner, D, Horgan, LE, Liu, C, tsaasan, a.m., Nardi, B & Rickman, J 2017 ‘Chat Speed OP

PogChamp: Practice of Coherence in Massive Twitch Chat’, in Proceedings of the 2017
CHI Conference Extended Abstracts on Human Factors in Computing Systems, CHI, Denver,
USA, pp.858-871.

Gal, N, Shifman, L & Kampf, Z 2016, ‘“It gets better”: Internet memes and the construction of
collective identity’, New Media & Society, vol.18, no.8, pp.1698-1714.

Glitsos, L & Hall, J 2019, ‘The Pepe the Frog meme: an examination of social, political, and
cultural implication through the tradition of the Darwinian Absurd’, Journal for Cultural
Research, vol.23, no.4, pp.381-395.

Goffman, E 1956, The Presentation of Self in Everyday Life, Harmondsworth London.
Gray, KL 2014, Race, Gender, and Deviance in Xbox Live: Theoretical Perspectives from the Virtual

Margins, Anderson Publishing.
—2017, ‘“They’re just too urban”: Black gamers streaming on Twitch”, in J Daniels, K Gregory &

T.M. Cottom (eds), Digital Sociologies, pp.355-368.

Jackson

Hamilton, WA, Garretson, O, & Kerne, A 2014, ‘Streaming on Twitch: Fostering Participatory
Communities of Play within Live Mixed Media’, in Proceedings of the 32nd annual ACM
conference on Human factors in computing systems, ACM, ON, Canada, pp.1315-1324.

Hercules 1997, film, Disney.
Hilvert-Bruce, Z, Neill, JT, Sȷöblom, M & Hamari, J 2018, ‘Social motivations of live-streaming

viewer engagement on Twitch’, Computers in Human Behaviour, vol.84, pp.58-67.
Johnson, MR & Woodcock, J 2019, ‘“It’s like the gold rush”: the lives and careers of professional

video game streamers on Twitch.tv’, Information, Communication & Society, vol.22, no.3,
pp.336-351.

Jung, CG 1928, Two Essays on Analytical Psychology, Routledge.
Kingdom Hearts 3 2019, video game, Square Enix.
Literat, I & van den Berg, S 2019, ‘Buy memes low, sell memes high: Vernacular criticism and

collective negotiations of value on Reddit’s Memeeconomy’, Information, Communication
& Society, vol.22, no.2, pp.232-249.

Marshall, PD 2010, ‘The promotion and presentation of the self: Celebrity as marker of
presentational media’, Celebrity Studies, vol.1, no.1, pp.35-48.

Marshall, PD, Moore, C & Barbour, K 2020, Persona Studies: An Introduction, John Wiley & Sons.
Marwick, AE 2013, Status Update: Celebrity, Publicity, and Branding in the Social Media Age, Yale

University Press.
Milner, RM 2016, The World Made Meme: Public Conversations and Participatory Media, MIT

Press.
Moore, C, Barbour, K, & Lee, K 2017, ‘Five dimensions of online persona’, Persona Studies, vol.3,

no.1, pp.1-11.
My Neighbor Totoro 1988, film, Studio Ghibli.
Nissenbaum, A & Shifman, L 2017, ‘Internet memes as contested cultural capital: The case of

4chan’s /b/ board’, New Media & Society, vol.19, no.4, pp.483-501.
Partin, WC 2019, ‘Watch Me Pay: Twitch and the Cultural Economy of Surveillance’, Surveillance

& Society, vol.17, no.1/2, pp.153-160.
Phillips, W & Milner, RM 2018, The Ambivalent Internet: Mischief, Oddity, and Antagonism Online,

John Wiley & Sons.
Pokémon Red Version 1998, video game, Game Freak.
Scully-Blaker, R, Begy, J, Consalvo, M & Ganzon, S 2017, ‘Playing along and playing for on Twitch:

Livestreaming from tandem play to performance’, in Proceedings of the 50th Hawaii
International Conference on System Sciences, AIS, Hawaii, USA, pp.2026-2035.

Seering, J, Kraut, R & Dabbish, L 2017, ‘Shaping Pro and Anti-Social Behavior on Twitch Through
Moderation and Example-Setting’, in Proceedings of the 2017 ACM Conference on
Computer Supported Cooperative Work and Social Computing, ACM, Portland, USA,
pp.2026-2035.

Senft, T 2008, Camgirls: Celebrity & Community in the Age of Social Networks, Peter Lang.
Shifman, L 2012, ‘An anatomy of a YouTube meme’, New Media & Society, vol.14, no.2, pp.187-

203.
—2013, ‘Memes in a digital world: Reconciling with a conceptual troublemaker’, Journal of

Computer-Mediated Communication, vol.18, no.3, pp.362-377.
—2014, MIT Press Essential Knowledge: Memes in Digital Culture, MIT Press.
—2019, ‘Internet Memes and the Twofold Articulation of Values’, in M Graham & W.H. Dutton

(eds), Society and the Internet: How Networks of Information and Communication are
Changing Our Lives, pp.43-57.

Sȷöblom, M & Hamari, J 2017, ‘Why do people watch others play video games? An empirical
study on the motivations of Twitch users’, Computers in Human Behavior, vol.75, pp.985-
996.

Sȷöblom, M, Törhönen, M, Hamari, & Macey, J 2017, ‘Content structure is king: An empirical
study on gratifications, game genres and content type on Twitch’, Computers in Human
Behavior, vol.73, pp.161-171.

Persona Studies 2020, vol. 6, no. 2

—2019, ‘The ingredients of Twitch streaming: Affordances of game streams’, Computers in
Human Behavior, vol.92, pp.20-28.

South Park: The Stick of Truth 2014, video game, Ubisoft.
Spitzberg, BH 2014, ‘Toward a model of meme diffusion (m3d)’, Communication Theory, vol.24,

no.3, pp.311-339.
Stardew Valley 2016, video game, Chucklefish.
Tannen, D 2007, Talking Voices: Repetition, Dialogue, and Imagery in Conversational Discourse,

Cambridge University Press.
Taylor, TL 2018, Watch Me Play: Twitch and the Rise of Game Live Streaming, Princeton

University Press.
Toelken, B 1996, Dynamics of Folklore, University Press of Colorado.
Wiggins, BE & Bowers, GB 2015, ‘Memes as genre: A structurational analysis of the memescape’,

New Media & Society, vol.17, no.11, pp.1886-1906.
Woodcock, J & Johnson, MR 2019, ‘The Affective Labor and Performance of Live Streaming on

Twitch.tv’, Television & New Media, vol.20, no.8, pp.813-823.

Nathan J. Jackson University of New South Wales
Abstract
Key Words
Introduction
Twitch, Persona, and Collective Identity
Memes, Agency, and Identity
Method
RayNarvaerJr
PaladinAmber

Memesis
Ray and Memesis Performed by Spectators
Amber and Memesis Led by the Streamer

Memesis and Video Games
Memetic Histories
Conclusion
End Notes
Acknowledgements
Works Cited