Wild,stressful,or stupid: Que es Bandersnatch ? Exploring user outcomes of Netflix’s interactive Black Mirror episode

Abstract

Prior scholarship related to interactive digital narratives (IDNs) typically has analyzed this emerging genre of storytelling on web-based platforms. The Bandersnatch episode of the Black Mirror series on Netflix offers a unique opportunity to explore the affordances and audience effects of IDNs appearing within a popular streamed TV program. This case study combines updated models of the uses and gratifications theory of media use with human–computer interaction frameworks to identify cognitive themes and emotional valences among Bandersnatch users, as expressed through tweets related to their experience. Findings show positive emotions were associated with perceived degree of control, as well as immersion or sense of flow. Gratifications associated with binge-type behavior on streaming services were also present. However, perceived realism, parasocial identification with the character, a need for a sense of completion, and the burden of choice created stress among many users. Social media served as a valuable ancillary vehicle for users to participate with others around the program, thus elevating the interactive narrative elements of the experience, which were otherwise limited and controlled.

Keywords

Bandersnatch Black Mirror digital TV interactive digital narrative social media Uses and Gratifications video games

‘Lmao I was NOT ready for the first decision in #Bandersnatch but now the remote is glued to my hands’.

Twitter user

Introduction

With little promotional fanfare, Netflix released its first interactive video in December 2018 as a segment of the popular dystopian Black Mirror series. The Bandersnatch episode presents as a hybridized video game within a TV show or movie. Roth and Koenitz (2019: 249) define this category of entertainment as an interactive digital narrative (IDN), which is an ‘emerging expressive form in the digital medium’ that allows users to participate in the experience they are viewing and influence the narrative by choosing from potential options within a computational system.

Early in the history of television, researchers and the popular press associated TV watching with lazy, uneducated, lower-class people who did not want to work at being entertained (Glick and Levy, 1961). This portrayal led to stereotyping audiences as ‘couch potatoes’ and TV as the ‘boob tube’. Thus, as Butsch (2001) noted, TV audiences became painted as lowbrow types who watched programs passively, meaning without effort or intention. Subsequent scholars added to the description of TV viewing as passive (Finn, 1992; Levy and Windahl, 1984; Rubin, 1993). Audiences were considered to be active only if they intentionally watched programs with high educational or informational value (Butsch, 2001).

Scholars have since argued that the so-called active versus passive audience framework is flawed (Jenkins, 1992/2020; Seaman, 1992) and all TV viewing is performative in nature (Abercrombie and Longhurst, 1998). The addition of interactive participatory components of human–computer interaction (HCI; Heeter, 1989; Lee et al., 2010) to television shows has further evolved the way scholars analyze media uses (LaRose and Eastin, 2004; Ruggiero, 2002) and changed audience expectations. The incorporation of games into media is even being explored by legacy news organizations, to increase interactivity, learning, and engagement with audiences (Ferrer-Conill et al., 2020). Audiences of all ages have come to expect to participate with televised media in some form, often using existing social media platforms to share their experiences and knowledge about a program (Jenkins, 1992/2020; Murray, 1997/2017). As fictional entertainment, the Bandersnatch episode subsequently won two Emmys for outstanding television movie and outstanding creative achievement in interactive media within a scripted program (Television Academy, 2019). Still, categorizing this emerging genre of entertainment is challenging for industry producers, scholars, and audiences alike. User outcomes are also of interest to researchers as this type of newer media programming combines digital elements with more traditional features of storytelling.

Background: Black Mirror and Bandersnatch

Charlie Brooker, a well-known British broadcaster, created the futuristic Black Mirror series, which aired solely in the United Kingdom until Netflix purchased it in 2015. Brooker has continued to write the series with executive producer Annabel Jones (Strauss, 2019a). In 2017, Netflix executives approached Brooker and Jones with the idea of incorporating interactive choices into an episode of Black Mirror. At the time, the technology did not exist for users to choose branching narratives within a feature-length show. During the next 2 years, Brooker and Jones worked with the Netflix team to write the narrative and develop the branching technology, an expensive process Brooker described as difficult and challenging, requiring twice the effort and cost of a typical Black Mirror episode (Rubin, 2018; Strauss, 2019a). Although Netflix has not released the cost of production for Bandersnatch, it was part of more than US$10 billion the company spent on creating original content in 2018. In producing Bandersnatch, Netflix developers also successfully created a software tool called Branch Manager that will allow future producers to incorporate this type of interactivity into other Netflix programs (Rubin, 2018).

The main character in Bandersnatch is a young amateur video game designer, Stefan Butler, who is attempting to enter the gaming industry in the 1980s. Set in the United Kingdom, the video offers the viewer/user choices that range from low-level significance (which cereal to eat) to high stakes (kill dad or back off). Choices are given on screen in a textual binary form. Once users make a decision, the video seamlessly incorporates that choice into the narrative. Some decisions dramatically alter the outcome of the story; others are less significant. Ironically, control is a subtheme of the episode as Butler repeatedly questions whether he is in charge of his actions as the video progresses. The episode has been described by media columnists as a ‘choose-your-own-adventure’ story (e.g. Sims, 2018; Strauss, 2019b), a genre made popular in the 1980s and 1990s by a children’s book series published by Chooseco LLC, which has trademarked the phrase.¹

After its release, the Black Mirror episode quickly triggered thousands of social media impressions globally with #Bandersnatch trending on both Twitter and Instagram. Reddit users attempted to crowdsource solutions to the game by posting flowcharts tracking the various endings of the episode based on all possible outcomes. The incorporation of interactive elements into the episode was a first for Netflix and could be a prelude of more IDNs to come across a variety of genres now that the technology exists (Strauss, 2019a). In fact, Netflix later released an interactive episode of the comedy Unbreakable Kimmy Schmidt in May 2020, although it offered fewer ending choices than Bandersnatch and the choices were even less consequential in determining the story’s outcome (Kang, 2020).

Research scope

As audiences move away from traditional forms of television viewing, producers of streamed content are attempting to attract and engage viewers through these types of interactive experiences that offer varying degrees of audience control. Although passive viewing of Bandersnatch was possible, because the screen defaulted to an option if a choice was not made in 10 s, Netflix reported that 94% of viewers were actively making decisions throughout the experience – an outcome that Netflix executives deemed successful (Strauss, 2019b).

To further explore social and emotional outcomes of interacting with IDNs, the purpose of this case study is to categorize the cognitive themes users expressed on social media in response to their Bandersnatch experiences. Emotional valences (positive, negative, neutral) associated with these themes are also examined through answering the research questions. Framed by newer models of the Uses and Gratifications (U&G) theory of media audiences that incorporate digital affordances (e.g. LaRose and Eastin, 2004; Ruggiero, 2002; Steiner and Xu, 2020; Sundar and Limperos, 2013) and concepts of human–computer interactivity in digital narratives, this study contributes to the academic literature on this genre by examining audience effects, based on social media posts, of engaging with this newer form of storytelling on a mass media entertainment platform. As more IDNs appear on streaming platforms, this study provides a framework with which to analyze users’ reactions and experiences on a variety of platforms.

Literature review

Originally, the U&G theory of mass media use (Blumler, 1979) was developed around the premise that audiences make media choices based on needs they are seeking to gratify. Thus, motivations range from active (cognitive, emotional, personal, and social) to more passive (entertainment/diversionary). Television viewing has long been categorized as a ritualized passive activity, particularly when audiences are watching entertainment programming, rather than reading newspapers, which requires more active participation and is therefore characterized as instrumental or purposive behavior (Finn, 1992; Levy and Windahl, 1984; Rubin, 1993). Numerous scholars have criticized U&G theory, however, as well as the concept of active audiences, particularly since the digital age (LaRose and Eastin, 2004; Ruggiero, 2002; Sundar and Limperos, 2013). Some argued that the early research was too reliant on self-reports and largely ignored the social origins of audience needs (Ruggiero, 2002). LaRose and Eastin (2004) took issue with the premise that audiences specifically seek types of programming to fill certain gratifications and recommend incorporating more social cognitive variables. Yet, as Ruggiero (2002) noted, the theory is still relevant for media use studies as long as it is situated in a more modern context of new media ecologies. Similarly, Sundar and Limperos (2013) endorsed adding interactivity to audience studies, as well as user agency to generate content, and navigability. Because Bandersnatch offers agency to the user, resulting in possible gratifications, an updated U&G model is presented below to guide the present study.

Active audience framework

In a series of studies in the late 80s and early 90s, Rubin and his coauthors explored the relationship between media activity levels among audiences based on their social contexts. For example, loneliness was associated with passive TV viewing, which served as a substitute for interpersonal interaction (Perse and Rubin, 1990; Rubin and Rubin, 1985). In addition to activity levels, Rubin (1993) recommended incorporating other variables when studying cognitive outcomes of media uses: media orientation (instrumental or ritualized), media attitudes (perceptions of realism and affinity), social and psychological factors, and involvement.

Television viewers who were heavily involved in the action and closely identified with characters were the most likely to form parasocial relationships with the actors, leading to stronger cognitive outcomes (Rubin et al., 1985). The concept of parasocial relationships with mass media figures, which describes an illusion of intimacy that audience members may feel toward certain characters on the screen, was first described by Horton and Wahl (1956) and has since been established through subsequent research within both mass media and virtual contexts (e.g. Bond, 2020; Kim and Kim, 2020; Levy, 1979). Jenkins (1992/2020) referred to the blurring of fact and fiction within fan communities as ‘emotional realism’, which can result in vocal fan critics who feel powerless over a narrative that was designed by mass media producers for commercial reasons. Instrumentalized media use and viewers’ perceptions of realism also correlated with more salient cognitive effects (Rubin et al., 1988).

Television and fan scholars, however, have taken issue with the active audience theory as being too focused on the comfortable middle class while ignoring the constraints put on viewers by corporate-owned media, which are not inclined to give audiences true interactivity or power (Jenkins, 1992/2020; Seaman, 1992). Abercrombie and Longhurst (1998) also criticized the passive versus active categorization of audiences, arguing that merely being an audience member constitutes a performance that is reflective of one’s identity. Indeed, as more viewers openly share their thoughts and experiences about a program online, media scholars are presented with further opportunities to examine the performance aspect of audiences and fans (Gray and Lotz, 2012; Jenkins, 2006).

In presenting his updated model of U&G theory, Ruggiero (2002) argued that additional affordances of digital media must be taken into account, particularly interactivity, demassification, and asynchroneity. He characterized interactivity associated with the Internet as strengthening the U&G notion of active user while posing some limitations because the user must decide from an array of choices online (pp. 15–16). Demassification represents the control the individual has over newer forms of media through deciding how to engage with digital content and in what form (p. 16). Asynchroneity signifies the lack of time constraints on viewing digitized media; users are no longer required to watch a program live, for example, as it airs on one network. Streaming content providers, such as Netflix, have time-shifted TV watching even further. Ruggiero noted, however, that the traditional academic questions are still relevant, even as newer forms of mediation communication are explored: ‘Why do people become involved in one particular type of medium or another and what gratifications do they receive from it?’ (Ruggiero, 2002: 29). Costello and Moore (2007) further defined active viewing in the digital age as a continuum of behaviors that could range from seeking information online related to the television show (lower levels) to socially conversing with others regarding the program (higher levels).

Interactive media

Prior to the digital era, scholars defined interactive media as a two-way platform that blurred the line between the sender and receiver of mediated messages (Singer, 1998). Interactivity has several dimensions, however, some requiring more effort from the audience than others. Ha and James (1998) described the dimensions of playfulness and choice as a lower level of interactivity, fulfilling entertainment and self-communication needs, while information collection and reciprocal communication were classified on the higher scale of interactivity.

Television traditionally has been a one-way form of mass communication, but producers have experimented with interactive technologies as early as the 1970s. The Warner-owned QUBE, for example, was an interactive cable television system that allowed viewers to purchase movies, play games, shop and respond to opinion polls by pressing one of five buttons on a remote (Arceneaux, 2018). Arguing that QUBE was evolutionary, not revolutionary, Arceneaux (2018) described QUBE as having originated in the pay-per-view cable space without ever fully embracing true interactive capabilities. For example, he noted that QUBE did not offer interactive narrative programming that would allow viewers to actively choose the storyline. Financial challenges and logistics prevented the company from further innovation. Although QUBE offered a glimpse of interactive electronic media, the company’s ultimate failure was partially a result of its more linear, old-media mindset, which was not conducive to meaningful audience participation (Arceneaux, 2018).

Digital media affordances of storytelling

In her seminal work, Hamlet on the Holodeck (1997/2017), Murray asserted that the hypertextual affordances of digital media offered unique opportunities for interactive storytelling on a screen, which may result in increased levels of audience interest and engagement. Since the 21st century ushered in the digital era, scholars and developers have been studying interactive storytelling largely on web-based platforms. Two European conference series, for example, merged in 2008 to form an annual International Conference on Interactive Digital Storytelling. In these proceedings, academics have presented findings and theoretical frameworks related to various types of interactive narratives, including video games, documentaries, journalistic, and artistic designs, some incorporating virtual and augmented reality (e.g. Koenitz, 2010; Roth and Koenitz, 2019). Interactions offered by IDNs vary in the degree of agency given to users, which Murray (1992/2017) defines as meaningful and satisfying actions. Thus, interactivity can range from allowing audience questions (low), to inviting user-generated content, and ultimately, giving actual control of the narrative to users (high).

Storytelling in the news and documentary genres, sometimes called the i-doc (interactive documentary), has been presented in combinations of online and live performances (Aston and Odorico, 2018). For example, the UK-based Choose Your Own Documentary combined narrative storytelling with film clips and invited live audience participation through remote controls. The BBC offers digital interactive documentaries as well, showcasing the works of young creators in an online feature called Arts. Much of the past research at the time of this study, however, has focused on IDNs that did not originate from mass media producers. Bandersnatch is unique in that Netflix presented the episode as part of a series commonly viewed on a television or computer screen, thus combining interactive elements with streamed TV watching.

Interaction with televised content

In the early 21st century, digital technologies began providing new affordances for audiences wishing to interact with television programming via digital platforms and devices, which in some cases has forced producers to relinquish control over outcomes of their shows. The exponential growth in social media usage since 2006 has allowed television viewers to share ideas and comment about shows in real time with friends and strangers through what has become known as the second screen (i.e. mobile phones, laptops, or tablets). Contemporary audience studies have shown that second screen activity related to a television show of shared interest can promote a sense of community, increase engagement, and facilitate formal and informal learning (see e.g. Gil de Zúñiga et al., 2015; Nee 2013; Nee and Dozier 2017; Park, 2015; Selva, 2016; Wilson, 2016). Additionally, Pires de Sá (2018) found that viewers of a Brazilian telenova had a heightened perception of media realism after engaging in Facebook posts about the show.

Jenkins (1992/2020, 2006) envisioned this convergence between old and new types of media as not taking place on one sole platform, rather he saw a ‘participatory culture’ emerging in which fans would have a reciprocal relationship with producers by spreading their own media and ‘poaching’ mass media content to create derivatives of popular works. Murray (1997/2017) described how digital interaction could take place with television shows through a companion website in which users explore virtual artifacts related to the program, such as divorce papers belonging to the main characters or photo albums and diaries. Murray also envisioned the merger of the TV and computer as one that could invite participation by allowing users to choose different branches of a narrative. Audiences also might have the opportunity to chat with each other in virtual rooms configured within the program, forming a ‘Greek chorus’ of viewers (Murray, 1997/2017: 324). Skeptics, however, argued that digital environments would never be intellectually or creatively superior to printed works (e.g. Birkerts, 1994; Carr, 2010). These arguments were similar to early criticism of television as a passive, dumbed-down medium that creates ‘couch potatoes’ (e.g. Postman, 1985)

Jenkins, however, differentiated between interactivity as a property of technology and participation, as a property of a culture wherein people contribute knowledge regarding shared experiences (Jenkins et al., 2016). Producers can integrate interactivity into digital storytelling by offering preprogrammed choices for the user while participation could take several forms, including user-generated content circulated through social media (Jenkins, 2006; Jenkins et al., 2017). In her updated edition of Hamlet (2017), Murray outlined how her predictions of a TV/computer merger have been validated by academics, practitioners and the marketplace, although she noted that complex digital storytelling is still in the early stages. She described viewer interaction with television shows as taking place sequentially through other frameworks, such as social media, after watching the show, or simultaneously while watching it.

Reality TV singing competition shows were among the first to incorporate the opinions of viewers. In its early years, American Idol opened phone lines at the end of the show, asking viewers to vote for their favorite singer and later via text asynchronously. Online fan communities formed around some contestants while other communities emerged around less popular singers encouraging people to ‘vote for the worst’ in an attempt to disrupt the outcome of the show (Jenkins, 2006). Increasingly, fans began interacting with other genres of TV shows, including fiction, into their own hands, creating fan sites, ‘poaching content’, and using social media to discuss plotlines and content (Jenkins, 1992/2000, 2006). For Jenkins (2006; Jenkins et al., 2017), the key to participatory culture has been the ability for amateurs and non-elites to create, produce, and spread media to the masses, which is a direct affordance of information communication technologies (ICTs), and can lead to a greater perception of interactivity that consumers have come to expect. In his updated 2020 edition of Textual Poachers, Jenkins further explicated his notion of what he had called transmedia storytelling, using multiple platforms to expand on one narrative. He described fan communities as producers of collective intelligence who engage in problem-solving around mass media programs through networked communities (1992/2020, xxv).

Murray (1992/2017) described a more advanced stage of interactivity that would provide a unified environment for watching and interacting with a TV program. HBO GO, for example, combines social apps alongside streamed content. Fan interaction, however, has caused a tension between audiences and media producers, which van Es (2016) called a ‘participation dilemma’ for corporate media owners. When viewers are given too much influence over an outcome of a show, compelling television may not be the result. Using NBC’s The Voice as a case study, van Es demonstrated how producers had to pull back on including social media comments synchronously into the show and instead moved engagement to asynchronous online platforms. Thus, audience participation became reactive (responding to polls and questions) rather than true interaction that would have influenced the outcomes. As of 2017, Murray noted that simultaneous actions presented as choices to the viewer within a TV program still needed further exploration. Clearly, with Bandersnatch in 2018, Netflix became one of the first to attempt this branching narrative on a mass scale.

In addition to traditional TV shows, researchers have found that even asynchronous viewing of streamed and digital content produces similar social outcomes on the second screen as synchronous or appointment viewing (Nee and Barker, 2020; Pittman and Tefertiller, 2015; Steiner and Xu, 2020). Steiner and Xu (2020) suggested that newer forms of technologies expand the gratifications sought beyond those typically studied in U&G research. Their study of viewers who binge watched streamed content, for example, identified numerous motivations for watching hours of programming at a time, including improved viewing experience, cultural inclusion, to catch up, relax and gain a sense of completion.

Interactivity and HCI

While traditional mass media more recently have invited interaction through digital platforms (or in most cases reaction), the concept of interactivity is uniquely central to the discipline of computer-based communication. Scholars have positioned the computer as inherently theatrical, offering storytelling opportunities with rich user experiences, especially those involving virtual worlds and artificial intelligence (Laurel, 2013; Murray, 2011; Ryan, 1991). A foundational premise of HCI scholarship is that users are able to act upon and change something through communication mediated by computer technology (Heeter, 1989). In this context, researchers use the term interaction to describe human/machine communication (Quiring and Schweiger, 2008). Heeter’s (1989) framework for analyzing interactivity includes the amount of choice afforded to the user, the effort required, the responsiveness of the medium to the user, as well as the user’s ability to add new information to the system and communicate with others interpersonally. Other scholars have conceptualized interactivity by assessing system features of the technological tools (Bucy, 2004) and focusing on whether the users perceive their experience to be engaging and mutually understood by others (Burgoon et al., 2000). Social media platforms, therefore, could be an ancillary system feature or affordances, allowing users to communicate with each other while interacting through ICTs.

Interacting with a program or game could lead to a sense of immersion, which in the context of this study is defined as being physically transported to a simulated place (Roth and Koenitz, 2016). In a participatory environment, immersion also involves doing the activities afforded by the medium, a type of digital swimming, as Murray (1997/2017) put it. From a psychological perspective, immersion could lead to being in a core flow state, which involves intense engagement and high arousal typically occurring as a result of taking part in an activity that is both challenging and enjoyable (Csikszentmihalyi, 1975). Flow theory also has framed studies that explore cognitive outcomes resulting from the use of participatory websites. In studying Facebook use, for example, Mauri and coauthors (2011) found that users experienced high positive valence and high arousal while on the site, which are both indicative of a core flow state. Roth and Koenitz (2016) noted that IDN interactors who experience flow could also feel a sense of presence, although designing IDNs to maintain consistent engagement among users is challenging.

Interactivity in video games

Many virtual video games afford players the ability to converse socially with each other, personalize their environment, and control their activities. Kaplan and Haenlein (2010) categorized various forms of social media applications based on degree of social presence/media richness and self-presentation/self-disclosure. They found that virtual game worlds offered the highest degree of social presence/media richness. The authors noted that although self-disclosure is limited in these environments, some players spend so much time with the game that they start to identify with their avatar characters in real life. The ability to interact with others may be a factor in cognitive outcomes as well. In a study of playing online games with human opponents versus computer-controlled, Weibel et al. (2008) found that participants who played with humans reported stronger effects of presence flow and enjoyment.

Other studies of cognitive outcomes of video games have correlated perceived realism of the game with heightened attention identification, enjoyment, and even physical aggression (Krcmar et al., 2011; Shafer et al., 2011). Violence and aggression are indeed characteristic themes of many popular video games. Lin (2013) also found that the interactivity of video games increased the aggressive affect among players whereas those who watched recorded events did not have the same response. Greitemeyer et al. (2019) demonstrated that people with dark personalities (‘everyday sadists’) enjoyed playing violent video games more than nonviolent games and showed an improved mood after playing them.

Interactive digital narratives

Building upon the principles of HCI and video games, IDNs have emerged as a type of entertainment video that enables users, generations of whom grew up playing video games since the 1980s, to influence the storyline through either a computer keyboard or television remote (Roth and Koenitz, 2019). Thus, IDNs provide users with the agency to make choices that alter their experience of the story. IDNs differ from video games, however, in that the choices available are more limited to maintain a cohesive story flow.

Although IDNs are fairly new, prior research has shown positive emotional outcomes when viewers are given the opportunity to interact with digitized content. Lee et al. (2010) conducted an experiment involving a 7-min interactive narrative called Modern Cinderella. Participants were assigned to one of four conditions involving solitary versus co-viewing with another person and interactive versus linear viewing. Those who had the interactive experience reported higher enjoyment. Solitary viewers paid more attention to the story and thought more carefully about their choices than those viewing with others in the same room.

Although Lee et al. (2010) concluded that IDNs were a promising development that could potentially be used for a number of genres, designing the technology to provide options, as well as a keeping cohesive storyline, is challenging. A producer of the Bandersnatch episode, Charlie Brooker, confirmed this dilemma, telling The Hollywood Reporter that the biggest storytelling obstacles he and his coproducers faced was creating cohesion among the various endings of the story while remaining consistent with the main characters (Strause, 2019b).

Roth and Koenitz (2019) studied reaction to the Bandersnatch episode among 32 media students in the Netherlands. Participants reported positive results on scales measuring usability, local effectance (direct input on a scene), and global effectance (impact on the overall narrative), indicating they felt they had a role in creating the narrative. However, elements of an immersive experience (presence and flow) were low and neutral, respectively. The authors suspected this was a result of low identification with the main character. Yet, participants rated character believability curiosity suspense and enjoyment positively. Participants also reported that they were in a more positive than negative mood after interacting with Bandersnatch.

In that study, Roth and Koenitz (2019) noted that the level of confusion was lower than the researchers had expected while the level of interest was higher. They explained that this finding could be due to the fact that participants were already interested in interactive narratives. Therefore, they recommend exploring these findings with a larger, more diverse sample. They also questioned whether Bandersnatch audiences framed the episode as a TV show or game.

Guided by the literature presented here on updated U&G models and interactivity, and research conducted thus far on IDNs, this study addresses the following research questions (RQ) through a content analysis of social media posts related to the episode:

RQ1: How did users classify the episode (e.g. interactive video game TV show movie)?

RQ2: What cognitive outcome themes did Bandersnatch users express on social media?

RQ3: Which emotional valences were associated with the cognitive outcomes?

Methods

To gauge user reaction, the author conducted a qualitative content analysis on public Twitter posts related to the Bandersnatch episode. Many media scholars have studied Twitter posts in relation to second screen activity because the platform tends to be public in nature and posts are displayed chronologically in real time (e.g. Moe et al., 2016).

Data collection

Tweets were collected for 4 days (December 28–31, 2018) immediately following the release of Bandersnatch on Netflix until the conversation subsided. A software program called Twitter Archivist was used to gather tweets using the hashtags #Bandersnatch and #BlackMirror, which collectively resulted in 4500 tweets and more than 41.5 million impressions (total number of tweets with the search term or hashtag appearing in the timeline). Location of tweets represented a wide range of countries, including England, France, the United States, South Africa, Italy, Costa Rica, Mexico, and Brazil.

Sample and procedure

The researcher exported the data set from Tweet Archivist to a Microsoft Excel file. The content of the tweets was sorted alphabetically, allowing the researcher to separate retweets (tweets that had been shared) from original tweets. A total of 3156 retweets were eliminated, leaving 1344 original tweets for data analysis. Because tweets were posted in a variety of languages (e.g. Spanish, Portuguese, Dutch, French, Turkish, Greek), the researcher used Google translator to read the content in English. To answer the research questions, the author reviewed the content of the original tweets looking for variables and themes that reflected the social and cognitive outcomes of media use as identified in the literature. Emotions associated with those outcomes were also observed when present. The author categorized the posts into themes through an iterative process as recommended by Lindloff and Taylor (2017), noting both commonalities and outliers in the sample.

Findings

With regard to the first RQ, how users classified the episode, the tweets contained a wide variety of categorization terminology. Users classified Bandersnatch as a movie, choose-your-own-adventure story, TV show, and video game. One post referred to it as a ‘TV turned video game’. Perhaps the best description was a tweet in Spanish asking: ‘Que es #Bandersnatch?’ Confusion was a subtheme even for those who enjoyed it. One user posted: ‘Poo it’s finished and I’m totally…erm…WTF DID I JUST WATCH? #Bandersnatch’.

The researcher recorded the adjectives used to describe the experience and uploaded them to a word cloud generator online Cloud Word (see Figure 1). Descriptors included wild, great, fun, good, and stupid.

Figure 1.

Words most commonly used to describe Bandersnatch in the sample of tweets.

RQs 2 and 3 explore cognitive outcomes themes and the emotions associated with those themes, as expressed through the sample of tweets. The following categories represent the media-use related themes that appeared in the data set: 1. Degree of control, 2. Perceived realism, 3. Social conversation, 4. Immersion, 5. Sense of completion/confusion (see Table 1).

Table 1.

Cognitive themes expressed by Twitter users in relation to the Bandersnatch episode and associated emotional valences.

Cognitive theme	Emotional valence (most common)	Example tweet (sic)
Degree of control	Positive/enjoyment	Holy crapola started black mirror bandersnatch this is v. exciting!
Perceived realism	Negative/stress	Watching Bandersnatch and I’m already a anxious mess. What if I choose the wrong cereal??? What’s going to happen!?
Social conversation	Positive/neutral	What are the best choices to make in bandersnatch movie?
Immersion/flow	Positive/enjoyment	I’m on the third ending of #Bandersnatch and I’m HOOKED
Sense of completion/confusion	Negative/neutral	I didn’t understand anything and I didn’t get to the end #Bandersnatch (translated from Portuguese)

Degree of control

Despite the rather limited binary choices in the episode, users expressed excitement about the usability and affordances of the game: ‘Holy crapola started black mirror bandersnatch this is v. exciting!’

Users referred to the creator Charlie Brooker as a ‘genius’ and described the experience as a precursor of 21st-century TV. Users also indicated a sense of agency in influencing the narrative: ‘My black family just made this white man yell at his father. Ooops. #Bandersnatch’. Others, however, questioned who was actually in control of the experience:

So is someone controlling me into making terrible decisions that will ruin my life like in Bandersnatch? Cuz I kinda knew it.

Who is in control? Netflix obviously they controlled me gave me the illusion of controlling Stephan while in fact they were in control and now I’m lost in the storyline and can’t get out.

They’re controlling MY decisions for god’s sake!!!

Yo, Bandersnatch clever af I was wondering what the technology aspect of it was whole time it’s us the users

Perceived realism

Users in this category expressed strong identification with the character. These tweets also contained a high degree of anxiety and stress because of the dark and violent consequences of their choices. Users questioned whether they were making the ‘right’ decisions and expressed concern that their indecisiveness would negatively impact the main character. These tweets were associated with negative emotional valences:

NO MATTER HOW HARD I TRY I ALWAYS END UP MESSING UP YOUR LIFE STEFAN IM SORRY #bandersnatch

Why do I get so nervous every time I have to choose? Even choosing between cornflakes is too much??

Watching Bandersnatch and I’m already a anxious mess. What if I choose the wrong cereal??? What’s going to happen!?

I was less anxious to watch black mirror bandersnatch because the thing I dislike most in life is indecision (translated from Turkish)

Bandersnatch really had my indecisiveness and depersonalization issues through the roof.

everytime he mentions someone else being in charge of his life during therapy, i just feel so guilty help

i literally get so stressed about these decisions even if it’s just what fricking song he’s gonna listen to #bandersnatch

this bandersnatch black mirror movie has me STRESSED I am literally the most indecisive person in the world

I had to put Stefan in prison and end it man. Fucking hell. #BlackMirror #Bandersnatch

should someone who is very bad at making decisions (me) even attempt Bandersnatch????? should I really ruin this poor boy’s life or what?

Social conversation

Although the episode itself offered no opportunity for users to interact with each other, as a video game might, users did so through social media. This conversation took the form of sharing flow charts with all possible outcomes and crowdsourcing for help and suggestions. The tweets that attempted to share knowledge or engage with others around the show reflected positive or neutral valences:

Wow. Bandersnatch was intense. Here are all the choices I made. No spoilers really.

what ending did y’all get on #Bandersnatch I need to know the other endings.

WHAT ENDING DID Y’ALL GET FOR #BANDERSNATCH? MINE WAS HE KILLED HIS DAD AND WENT TO JAIL

What are the best choices to make in bandersnatch movie?

Immersion/flow

Time spent on Bandersnatch was another dominant theme in the data set of tweets. Many users posted that they were well beyond the 1 h and 30 min that Netflix estimated the episode to last. These tweets were generally positive:

I wrote down all my answers for #bandersnatch so I’m gonna watch it again and choose differently. I’m excited lol it’s so fun

brb going to spend the rest of the day obsessing over Bandersnatch

You could probably lose a whole day watching bandersnatch

I’m on the third ending of #Bandersnatch and I’m HOOKED

Okay I’m halfway (I think) through bandersnatch and I already can’t wait to choose everything I didn’t already.

Woowee there went another two hours playing Bandersnatch.

Sense of completion

A common frustration expressed by users was that they were not able to find a sense of completion in the narrative. Users in this category were searching for a happy ending, which was elusive because of the dark nature of the show. Negative emotions were associated with this theme:

Going back to choose Frosties in #bandersnatch tomorrow – hoping for a jollier ending

I think I spent 2-3 hours on #Bandersnatch why is there no happy ending Wdf

Get me out of this fucking loop. #Bandersnatch (translated from Spanish)

BANDERSNATCH WAS SO STRESSFUL I AM GOING TO HAVE TO DO IT A MILLION TIMES TO GET ALL THE ENDINGS

Need to finish bandersnatch wish me luck

I think we finished bandersnatch but honestly idk

Although negative posts were far less dominant than the positive themes in the sample overall, users expressed confusion and frustration over the experience:

I didn’t understand anything and I didn’t get to the end #Bandersnatch (translated from Portuguese)

Too many choices drives me crazy #bandersnatch

oh my god there are so many alternative endings of bandersnatch my brain aches.

the only bandersnatch endings i didn’t got was stefan dying but this became so stupid to me huh

Just watched bandersnatch and disappointed tbh goes back on itself way too much

Discussion

Bandersnatch is a mass commercial television production that offered some limited interactive choices to the user. Findings of RQ 1 showed that users who posted about the experience on Twitter were confused about how to classify the program. Movie, TV show, and game were used to describe Bandersnatch. From an academic perspective, the episode is classified as an IDN, but users were only able to participate with each other and describe their experiences externally through social media. Thus, the experience was interactive only through the affordances of social media. The episode did incorporate the characteristics of digital media technologies that Ruggiero (2002) defined: interactivity, demassification, and asynchroneity. As prior research on streamed content has shown, social outcomes of asynchronous streamed content, can be similar to those of live or appointment TV (Nee and Barker, 2020; Pittman and Tefertiller, 2015; Steiner and Xu, 2020). Using Heeter’s (1989) framework, the degree of interactivity was low, however, because choices were limited, minimal effort was required, and users could not add new information to the system or construct their own meaning and endings within the narrative. The responsiveness of the medium was high, though, and users were able to communicate with each other, not as a system feature, but through social media, thus further representing demassification and asynchroneity. Therefore, this episode may fit within the evolutionary, not revolutionary, category of media technologies as Arceneaux (2018) defined QUBE – an early precursor to interactive television.

Indeed, social media offered the true affordance to users by giving them an ancillary platform where they could converse about the episode with each other, in a way that both Jenkins (1992/2020) and Murray (1997–2017) foresaw media convergence occurring. This process is representative of active viewing because it incorporates interpersonal conversations around the program (Costello and Moore, 2007). Some were seeking help from others in a manner representative of crowdsourcing and co-viewing, activities associated with the affordances of digital technologies (Lee et al., 2010). In a sense, social media became the platform through which users interacted with other players, leading to a stronger sense of social presence and enjoyment. This finding is consistent with prior research (Weibel et al., 2008) where participants reported stronger effects when playing online games with humans rather than computers.

In answering RQ2 and RQ3, regarding cognitive outcomes and emotional valences of users, a majority of the tweets analyzed in this study addressed the novelty of the episode in a positive way, although many users described making decisions as anxiety producing and stressful. Because of the requirements placed on the viewers, the Bandersnatch experience differed from traditional TV watching, which has been historically associated with passive viewing (Finn, 1992; Levy and Windahl, 1984; Rubin, 1993) and loneliness (Perse and Rubin, 1990; Rubin and Rubin, 1985). Additionally, those who chose to post their views of Bandersnatch on social media were intentionally seeking social connections with other users as well as actively exhibiting performative audience behaviors, earlier described by Abercrombie and Longhurst (1998). Ha and James (1998) classified the highest level of interactivity as including reciprocal communication while lower level dimensions were play and choice. Thus, those who were engaging on social media about the episode were demonstrating the highest degrees of interaction even if their posts were negative.

Because of the novelty aspect of the episode, the media orientation of Bandersnatch was instrumental, not ritualized. As noted in the literature review, instrumentalized media use combined with perceptions of reality and strong identification with the characters produce more salient cognitive outcomes (Rubin, 1986; Rubin et al., 1988). The posts included in the perceived realism theme demonstrated heightened emotional reactions as users worried about making the ‘right’ choice for Stefan and related their indecisiveness to their own lives. Even though the storyline was dark and violent, users in this category seemed to have formed a parasocial relationship with Stefan. This theme was inconsistent with Roth and Koenitz’s (2019) study of Bandersnatch users who did not tend to identify with the main character. As those authors noted, however, the participants in that study were a limited number of media students who may not be representative of a larger population of users. In the present study, users wanting a happy ending for Stefan further demonstrated their identification with the character and their cognitive need for a sense completion, which has been associated with binge watching entire seasons of streamed content in one sitting (Steiner and Xu, 2020). Others who might be described as ‘everyday sadists’ (Greitemeyer et al., 2019) were more eager to make violent choices and did not seek out a happier ending.

Gratifications of binge-type behavior (Steiner and Xu, 2020) were evident in the tweets that addressed immersive themes. These tweets tended to be playful and positive, although some guilt was expressed because of the amount of time spent playing the episode. Although causation cannot be determined, some user descriptions could be evident of an immersive experience that might have led to a core flow state (Csikszentmihalyi, 1975; Murray, 1997/2017; Roth and Koenitz, 2016) with users expressing high arousal by taking part in the episode, which they described as both challenging and enjoyable – similar to the findings of the Mauri et al. (2011) study of Facebook use. The control theme was consistent with Roth and Koenitz’s (2019) findings in which participants reported a high degree of usability within Bandersnatch and perceived local as well as global effectance. The posts in the control category reflected this perception, although a smaller number of users expressed that the producers of the episode were controlling them, not vice versa. The subtheme of control within the episode itself likely was designed to make users question their control of their own lives, resulting in some confusion and questioning among viewers, as well as praise for the ‘existential’ experience.

In the final analysis, Bandersnatch was not designed for sociability or true participation, however. Netflix did not provide a chat room or opportunities to further explore Stefan’s universe. For example, a web-based virtual world might have been constructed in which users could play clips from Stefan’s video game or search his bedroom. These elements would have elevated Bandersnatch to a more truly IDN, similar to what Murray (1997/2017) envisioned. Instead, the ‘Greek chorus’ she described took place among fans on third-party social media sites.

Media producers no longer have sole control over communication channels and audiences (Jenkins et al., 2017). In fact, Netflix as a streaming service and content creator has been a key disrupter of traditional televised media. While Netflix and Brooker’s team maintained ultimate control over the narrative in Bandersnatch, user participation came in the form of posting reactions and conversation starter about the show and even spoilers, such as flowcharts of possible endings, on social media. Some users referred to Charlie Brooker, whom already was widely trusted by fans, as a genius for creating the Bandersnatch episode. In fact, his genius might have been in giving users the illusion of an interactive experience, which in reality was a closed, authored environment with few affordances for true participation in constructing the narrative. Now that Netflix has developed this branching technology, future endeavors could be more justifiable and less expensive, but still time-consuming to produce and questionable in terms of affordances.

Limitations and conclusion

Although this study does contain a larger and more diverse sample than the only prior published research on Bandersnatch (Roth and Koenitz, 2019) at the time of this writing, the content analysis was performed on a purposive sample of Twitter posts during the first 4 days of its release, and therefore may not be representative of all Bandersnatch user experiences. Twitter users are a small subset of the population globally and the use per capita varies widely among countries (Statistica, 2020). In the United States, for example, 22% of American adults use Twitter, but they tend to be younger, more highly educated, and have higher incomes than US adults overall, and only a small portion of users tweet actively (Wojcik and Hughes, 2019). By posting publicly on social media, these users already were exhibiting strong cognitive outcomes and a desire or need to engage with others regarding the unique activity. Additionally, because of the dark nature of Black Mirror, users’ reaction to this episode may not be generalized to other types of genres, such as comedies, that attempt to incorporate interactive storytelling elements.

Still, these unfiltered asynchronous comments provide some insight into the cognitive and emotional effects of participating with IDNs. Categorizing Bandersnatch is still enigmatic as Roth and Koenitz (2019) also noted. In answering the research questions regarding user reactions on social media, users classified the episode as both a game and movie, combining elements of entertainment media with HCI. Although the episode itself may not meet earlier academic definitions of high interactivity (Ha and James, 1998; Heeter, 1989; Singer, 1998), users expressed positive emotions and enjoyment related to their perception of control and global effectance on the narrative. Those who identified with the main character reported the highest amounts of stress because they did not want to make the ‘wrong’ choices for him. Those who needed a sense of completion or who were confused by the experience expressed more negative or neutral views. Finally, social media served as an ancillary vehicle for added interactivity, allowing users to fulfill a need for social interaction which was missing in the episode itself. Since this study was conducted, Netflix has released a variety of IDNs around genres ranging from children’s shows to games and reality TV. To gain a better understanding of the effects of these programs, researchers might expand upon the findings of this study through experimental designs and qualitative interviews with users.

Footnotes

ORCID iD

Rebecca C Nee

Note

Author biography

Rebecca Coates Nee is an associate professor in the School of Journalism & Media Studies at San Diego State University. She has authored more than a dozen peer-reviewed journal articles addressing the changes in journalism practices and audience behaviors due to the growth of digital technologies and social media. A former TV news anchor/reporter, Nee earned her doctorate in educational technology from Pepperdine University and a master’s in journalism from Northwestern’s Medill School of Journalism.

References

Abercrombie

Longhurst

(1998) Audiences: A Sociological Theory of Performance and Imagination. Thousand Oaks, CA: SAGE.

Arceneaux

(2018) The many sides of QUBE: Interactive television and innovation in electronic media 1977-1983. Journal of Broadcasting and Electronic Media 62(3): 531–546.

Aston

Odorico

(2018). The poetics and politics of polyphony: Towards a research method for interactive documentary. Alphaville: Journal of Film and Screen Media 15: 63–93.

Birkerts

(1994) The book as emblem: The besieged stronghold? Journal of Scholarly Publishing 26(1): 3.

Blumler

(1979) The role of theory in uses and gratifications studies. Communication Research 6(1): 9–36.

Bond

(2020) The development and influence of parasocial relationships with television characters: A longitudinal experimental test of prejudice reduction through parasocial contact. Communication Research 0093650219900632.

Bucy

(2004) Interactivity in society: Locating an elusive concept. The Information Society 20(5): 373–383.

Burgoon

Bonito

Bengtsson

. et al. (2000) Testing the interactivity model: Communication processes partner assessments and the quality of collaborative work. Journal of Management Information Systems 16(3): 33–56.

Butsch

(2001) Class and audience effects: A history of research on movies, radio, and television. Journal of Popular Film and Television 16(3): 112–120.

10.

Carr

(2010) The Shallows: How the Internet is Changing the Way we Think, Read and Remember. Ocala, FL: Atlantic Books Ltd.

11.

Costello

Moore

(2007) Cultural outlaws: An examination of audience activity and online television fandom. Television & New Media 8(2): 124–143.

12.

Csikszentmihalyi

(1975) Flow: The Psychology of Optimal Experience. New York, NY: Harper and Row Publishers.

13.

Ferrer-Conill

Foxman

Jones

et al. (2020) Playful approaches to news engagement. Convergence. DOI: 10.1177/1354856520923964.

14.

Finn

(1992) “Television “addiction?” An evaluation of four competing media-use models.” Journalism Quarterly 69(2): 422–435.

15.

Gray and Lotz (2012) Television Studies. Cambridge, UK: Polity Press.

16.

Gil de Zúñiga

Garcia-Perdomo

McGregor

(2015) What is second screening? Exploring motivations of second screen use and its effect on online political participation. Journal of Communication 65(5): 793–815.

17.

Glick

Levy

(1961) Living with Television. Chicago, IL: Aldine.

18.

Greitemeyer

Weiß

Heuberger

(2019) Are everyday sadists specifically attracted to violent video games and do they emotionally benefit from playing those games? Aggressive Behavior 45(2): 206–213.

19.

ames

(1998) Interactivity reexamined: A baseline analysis of early business web sites. Journal of broadcasting & electronic media 42(4): 457–474.

20.

Heeter

(1989) Implications of new interactive technologies for conceptualizing communication. In: Salvaggio

Bryant

(eds) Media Use in the Information Age: Emerging Patterns of Adoption and Consumer Use. New York, NY: Routledge Communication Publishers, pp. 217–235.

21.

Horton

Wohl

(1956) Mass communication and para-social interaction: Observations on intimacy at a distance. Psychiatry 19(3): 215–229.

22.

Jenkins

(2006) Convergence Culture: Where old and New Media Collide. New York, NY: New York University Press Publishers.

23.

Jenkins

(1992/2020) Textual Poachers: Television Fans and Participatory Culture. Updated 20th anniversary edition. New York/London: Routledge.

24.

Jenkins

Ito

boyd

(2016) Participatory Culture in a Networked Era. Cambridge, Mass: Polity Press.

25.

Jenkins

Lashley

Creech

(2017) A forum on digital storytelling| Interview with Henry Jenkins. International Journal of Communication 11(8): 1061–1068.

26.

Kang

(2020). ‘Unbreakable Kimmy Schmidt: Kimmy vs. the Reverend’: TV review. The Hollywood Reporter. Available at: https://www.hollywoodreporter.com/review/unbreakable-kimmy-schmidt-kimmy-reverend-review-1294248 (accessed 7 March 2020).

27.

Kaplan

Haenlein

(2010) Users of the world unite! The challenges and opportunities of Social Media. Business Horizons 53(1): 59–68.

28.

Kim

(2020) How does a celebrity make fans happy? Interaction between celebrities and fans in the social media context. Computers in Human Behavior 111: 106419.

29.

Koenitz

(2010) Towards a theoretical framework for interactive digital narrative. In: Aylett

Lim

Louchart

(eds) Interactive Storytelling: Third Joint Conference on Interactive Digital Storytelling. Berlin: Springer, pp. 176–185.

30.

Krcmar

Farrar

McGloin

(2011) The effects of video game realism on attention retention and aggressive outcomes. Computers in Human Behavior 27(1): 432–439.

31.

LaRose

Eastin

(2004) A social cognitive theory of Internet uses and gratifications: Toward a new model of media attendance. Journal of Broadcasting and Electronic Media 48(3): 358–377.

32.

Laurel

(2013) Computers as Theatre. Reading, MA: Addison-Wesley.

33.

Lee

Heeter

LaRose

(2010) A modern Cinderella story: A comparison of viewer responses to interactive vs linear narrative in solitary and co-viewing settings. New Media and Society 12(5): 779–795.

34.

Levy

(1979) Watching TV news as para-social interaction. Journal of Broadcasting and Electronic Media 23(1): 69–80.

35.

Levy

Windahl

(1984) Audience activity and gratifications: A conceptual clarification and exploration. Communication Research 11(1): 51–78.

36.

Lin

(2013) Do video games exert stronger effects on aggression than film? The role of media interactivity and identification on the association of violent content and aggressive outcomes. Computers in Human Behavior 29(3): 535–543.

37.

Lindlof

Taylor

(2017) Qualitative Communication Research Methods. Thousand Oaks, CA: SAGE publications.

38.

Mauri

Cipresso

Balgera

. et al. (2011) Why is Facebook so successful? Psychophysiological measures describe a core flow state while using Facebook. Cyberpsychology Behavior and Social Networking 14(12): 723–731.

39.

Moe

Poell

van Dijck

(2016) Rearticulating audience engagement: Social media and television. Television and New Media 17(2): 99–107.

40.

Murray

(1997/2017) Hamlet on the Holodeck: The Future of Narrative in Cyberspace, Updated Edition. New York, NY: Simon and Schuster Inc.

41.

Murray

(2011) Inventing the Medium: Principles of Interaction Design as a Cultural Practice. Boston, MA: MIT Press.

42.

Nee

(2013) Social TV and the 2012 election: Exploring political outcomes of multiscreen media usages. Electronic News 7(4): 171–188.

43.

Nee

Barker

(2020) Co-viewing virtually: Social outcomes of second screening with televised and streamed content. Television & New Media 21(7): 712–729.

44.

Nee

Dozier

(2017) Second screen effects: Linking multiscreen media use to television engagement and incidental learning. Convergence. 23(2): 214–226.

45.

Park

(2015) Pathways to expressive and collective participation: Usage patterns political efficacy and political participation in social networking sites. Journal of Broadcasting and Electronic Media 59(4): 698–716.

46.

Perse

Rubin

(1990) Chronic loneliness and television use. Journal of Broadcasting and Electronic Media 34(1): 37–53.

47.

Pires de Sá

(2018) Connected co-viewing on Facebook: A Brazilian telenovela and the perception of media realism. Television and New Media 19(7): 646–659.

48.

Pittman

Tefertiller

(2015) With or without you: Connected viewing and co-viewing Twitter activity for traditional appointment and asynchronous broadcast television. First Monday 20(7). Available at: http://journals.uic.edu/ojs/index.php/fm/article/view/5935/4663 (accessed 7 March 2020).

49.

Postman

(1985) Amusing Ourselves to Death: Public Discourse in the Age of Show Business. New York, NY: Penguin.

50.

Quiring

Schweiger

(2008) Interactivity: A review of the concept and a framework for analysis. Communications 33(2): 147–167.

51.

Ryan

(1991) Possible Worlds, Artificial Intelligence, and Narrative Theory. Bloomington, IN: Indiana University Press.

52.

Roth

Koenitz

(2016) Evaluating the user experience of interactive digital narrative. In: Chambel

(ed) Proceedings of the 1st International Workshop on Multimedia Alternate Realities, Amsterdam, 16 October 2016, pp. 31–36.

53.

Roth

Koenitz

(2019) Bandersnatch yea or nay? Reception and user experience of an interactive digital narrative video. In: Proceedings of the 2019 ACM International Conference on Interactive Experiences for TV and Online Video. New York, NY: ACM Press, pp. 247–254.

54.

Rubin

(1993) Audience activity and media use. Communication Monographs 60(1): 98.

55.

Rubin

Perse

Powell

(1985) Loneliness parasocial interaction and local television news viewing. Human Communication Research 12(2): 155–180.

56.

Rubin

Perse

Taylor

(1988) A methodological examination of cultivation. Communication Research 15(2): 107–134.

57.

Rubin

(1985) Interface of personal and mediated communication: A research agenda. Critical Studies in Media Communication 2(1): 36–53.

58.

Rubin

(2018) How the surprise new interactive Black Mirror came together. Wired. Available at: https://www.wired.com/story/black-mirror-bandersnatch-interactive-episode/ (accessed 2 February 2020).

59.

Ruggiero

(2002) Uses and gratifications theory in the 21st century. Mass Communication and Society 3(1): 3–37.

60.

Seaman

(1992) Active audience theory: Pointless populism. Media, Culture and Society 14(2): 301–311.

61.

Selva

(2016) Social television: Audience and political engagement. Television and New Media 17(2): 159–173.

62.

Shafer

Carbonara

Popova

(2011) Spatial presence and perceived reality as predictors of motion-based video game enjoyment. Presence: Teleoperators and Virtual Environments 20(6): 591–619.

63.

Sims

(2018) The branching horrors of Black Mirror’s ‘Bandersnatch.’ The Atlantic. Available at: https://www.theatlantic.com/entertainment/archive/2018/12/black-mirror-chooses-your-own-adventure-bandersnatch/579118/ (accessed 2 February 2020).

64.

Singer

(1998) Online journalists: Foundations for research into their changing roles. Journal of Computer-Mediated Communication 4(1): JCMC412.

65.

Statistica (2020) Leading countries based on Twitter users as of July 2020. Available at: https://www.statista.com/statistics/242606/number-of-active-twitter-users-in-selected-countries/ (accessed 15 August 2020).

66.

Steiner

(2020) Binge-watching motivates change: Uses and gratifications of streaming video viewers challenge traditional TV research. Convergence: The International Journal of Research into New Media Technologies 26(1): 82–101.

67.

Strauss

(2019a) ‘Black Mirror’ reveals cast, premiere date for season 5. The Hollywood Reporter, 15 May, 2019. Available at: https://www.hollywoodreporter.com/live-feed/black-mirror-season-5-cast-premiere-date-trailer-1210653 (accessed 2 February 2020).

68.

Strauss

(2019b) ‘Black Mirror duo on the challenges of Netflix’s first interactive movie – And why they would do it again. The Hollywood Reporter, 24 May, 19. Available at: https://www.hollywoodreporter.com/live-feed/black-mirror-team-complexity-bandersnatch-netflix-data-1213270 (accessed 2 February 2020).

69.

Sundar

Limperos

(2013) Uses and grats 2.0: New gratifications for new media. Journal of Broadcasting and Electronic Media 57(4): 504–525.

70.

Television Academy (2019) Bandersnatch (Black Mirror) awards and nominations. Available at: https://www.emmys.com/shows/bandersnatch-black-mirror (accessed 7 March 2020).

71.

van Es

(2016) Social TV and the participation dilemma in NBC’s The Voice. Television and New Media 17(2): 108–123.

72.

Weibel

Wissmath

Habegger

. et al. (2008) Playing online games against computer-vs. human-controlled opponents: effects on presence flow and enjoyment. Computers in Human Behavior 24(5): 2274–2291.

73.

Wilson

(2016) In the living room: Second screens and TV audiences. Television and New Media 17(2): 174–191.

74.

Wojcik

Hughes

(2019) Sizing up Twitter Users. Pew Research Center. Available at: https://www.pewresearch.org/internet/2019/04/24/sizing-up-twitter-users/ (accessed 7 March 2020).