ORIGINAL RESEARCH ARTICLE Using games to disrupt the conference Twittersphere

Social media tools are changing practices in many industries, including academia, and the Twitter platform is widely recognised as the ‘tool of choice’ for microblogging. Academic conferences often use social media to provide conference ‘back-channels’. This article describes a conference game using toys as alter egos, driven through Twitter. We found that the soft toy game format was participated in by a majority of the attendees, with early posts in advance of the conference a good signal of engagement. We look at what the organisers learnt from the game and how such games, including Twitter elements, could support wider networks beyond the conference itself.


Conferences and Twitter
Microblogging has changed the way academic communities interact and collaborate over the past decade, although there is a paucity of knowledge about the academic use of social media for work and professional learning in the day-to-day environment (Li and Greenhow 2015).Twitter is widely recognised as the 'tool of choice' for microblogging for knowledge transfer and socialisation (Cleveland, Jackson and Dawson 2016) and there is a growing body of work focussed on the use of this platform at academic conferences (Aramo-Immonen, Jussila and Huhtamaki 2015; Chung and Woo 2016;Ross et al. 2011).While much of this research focussed on the positive aspects, negative aspects of the use of Twitter have also been identified.Privacy-leaking and other nefarious uses are highlighted by Cleveland, Jackson and Dawson (2016), and Quan-Haase, Martin and McCay-Peet (2015, p. 10) propose that the concept of an 'invisible college' of Twitter users is messy, consisting of 'overlapping social contexts (professional, personal, and public), scholars with different habits of engagement, and both formal and informal ties'.
Any potential challenges posed by such negative usage of Twitter has not always been off-putting to conference organisers and it is now common for academic conferences to promote the use of Twitter to provide conference backchannels due to the many 2 Citation: Research in Learning Technology 2018, 26: 2036 -http://dx.doi.org/10.25304/rlt.v26.2036(page number not for citation purpose) potential positives (Reinhardt et al. 2009).However, the 'norm' is often within particular safe parameters.For instance, the organisers will use it as a broadcasting tool, making announcements and sending reminders; delegates will use it as a back channel-enhancing information and for sharing questions and thoughts (Kimmons 2016;Reinhardt 2009).A Twitter hashtag or account has now become almost a standard conference offering allowing for some to engage in more interesting and experimental usage.

Twitter as a conference game
Since 2015, members of the Playful Learning Special Interest Group (PLSIG) for the Association of Learning Technology (ALT) have used Twitter in various iterations of games at a number of different conferences including ALT's annual conference (ALTC), Librarians' Information Literacy Annual Conference (LILAC), and Playful Learning (PL) Conference.Members of this group had previously used Twitter traditionally as organisers of a number of events, and they were motivated to explore more innovative and expansive uses of the platform.
While a variety of conferences, as mentioned above, have offered complete conference games running before, throughout, and after the event, the initial game planned for ALTC in 2015 was the first that was planned by this particular group.The idea behind the game was initially just to add playful elements to the conference, but while planning it was soon realised that a game could also be used to encourage positive behaviours and help with issues conference organisers often run into such as aiding networking, gaining feedback, and promoting the event.With this in the mind, the objectives were widened out to include: • get participants to talk to one another; • demonstrate a range of game-based approaches; • encourage positive behaviours such as sharing feedback and ideas; • provide a playful element to the conference.
It should be noted that even this early conference game experimented with the 'traditional' conference use of Twitter.Twitter was used to drive the game's overarching narrative through a fictional evil character DarkBot (@BotofEvil) who planned to take over the world through poor-quality learning technology.This early game did not rely solely on Twitter as a mechanism for its delivery and the majority of the game could occur with minimal Twitter interaction.However, the successful and numerous interactions and engagement through Twitter with DarkBot encouraged the organisers to experiment further with this mechanism in later conferences.The hashtag for the game #altcgame, which on this occasion ran separately to the conference hashtag, saw 252 tweets @BotofEvil 548 and #bots 255.
Further developments and new iterations of this conference game were run at LILAC conferences and further ALTC events where Twitter was used in much the same manner and caused similar levels of conversation and engagement.However, these games were not without their challenges and failures, and it was reflection upon these that led to the development of a more immersive, avatar-based game.For example, Twitter had been previously used as an added extra, primarily pre-conference to encourage networking and a sense of community around the conference, when delegates arrived at the conference and other games took over.At Payful Learning, it was decided to make Twitter the focus, with the avatars encouraged to challenge, befriend, 3 (page number not for citation purpose) and support each other through the platform.A further challenge had been that when using Twitter, primarily pre-conference, feedback from day delegates arriving on the second or final day had been that they felt left out; as the tweets diminished, those joining 'late' were excluded.Keeping Twitter as a focus, with face-to-face elements the 'added extras', meant that all delegates could join in regardless of when they were physically joining the conference.

Use of toys as alter egos
In 2017, for the PL Conference, Twitter was adopted in a more radical and embedded way.After several versions of conference games had been played by various committee members, feedback and challenges were reflected upon.This resulted in a Twitter-focussed game, rather than the platform being an addition to the face-toface games offered at the conference.In order to give a level of anonymity to those playing, avatars were decided upon, in this case toys, who took on lives of their own throughout the game.
It was an easy decision to introduce this game at the PL Conference as it provides a great opportunity to be more experimental as delegates who attend are already in a more playful, lusory attitude.This article describes a conference game using toys as alter egos, driven through Twitter.It investigates the importance of Twitter to develop and support communities of practice, describes the format of the game, and evaluates its effectiveness for enhancing participant socialisation.

Social media and communities of practice
A positive dialogue that the conference wanted to exploit with its use of Toy Twitter was that of the community of practice (CoP) that is building around PL.A CoP can be a comforting place to be, peopled with like-minded souls and familiar territory.However, as Wenger et al. (2002, online) state, they also need 'interesting and varied events' in order to keep the ideas flowing and to bring new people into the community.Social media platforms are a valuable tool which 'function as knowledge sharing arenas for a community of practice' (Aramo-Immonen et al. 2016, p. 586).Using Twitter more strategically pre-, during and post-conference meant that the organisers could help to foster several of the seven principles of Wenger et al. (2002) for cultivating CoPs, in particular: 'Invite different levels of participation'; the expanded use of Twitter to include alter egos allowed delegates the opportunity to move more easily between the three identified degrees of participation: core, active, and peripheral, which in turn would help each member feel involved.
'Combine familiarity and excitement'; bringing in an unfamiliar and playful element to the CoP by allowing delegates to interact with each other as alter egos was a way to 'provide a sense of common adventure' (Wenger at al. 2002, online).

Game design
The idea of toy Twitter alter egos is not new; they were used in an emergent game played from April to July 2008 (Pugh 2008).As part of the sign up to this game, players chose to be a member of one of the two groups and this determined whether they accepted missions or drove the narrative of the game.The group accepting missions was sent instructions to set up a toy alter ego through Twitter.While this was not to support a conference, the freedom players experienced through their alter egos reflects the experience delegates had at the PL Conference.Plant (n.d.) commented '…the fact that the avatars were cuddly toys … ensured that a certain kind of playful atmosphere emerged'.This freedom was certainly something the game organisers thought would be important to explore for PL.
Two weeks prior to the conference, all delegates were sent instructions via email to bring a toy and to register that toy on Twitter in advance of the conference.The toys were encouraged to start tweeting at this point, with the first toy accounts appearing and engaging with each other the week before the conference started.A week before the conference, they were asked to complete a toy top trump card and they were sent Twitter travel games (Figure 1) to play on their journey.The pre-conference Twitter interactions were already encouraging playful interactions where 217 Toy tweets were sent in the 4 days prior to the conference starting compared to 61 from people.
For previous conference games, hashtags were used to separate the game from the official conference tweets, but for this event a single hashtag was used for all conference interactions.This worked particularly well for the game and for delegates at the conference itself.However, it could be seen as problematic in terms of conference broadcast as those remotely following the Twitter hashtag were getting a lot of toy tweets, which could potentially disrupt the academic backchannel.
When delegates registered at the conference, they also registered their toy which received its own delegate pack including toy mission cards (Figure 2).The mission cards were based around the game Sneaky Cards, a pay-it-forward game encouraging positive and daring behaviour.The mission cards became a trading game for the conference and encouraged more Twitter conversations, as well as face-to-face interactions between the delegates themselves, and the delegates and organisers.Every time a mission was completed, toys received either a special sticker and a new mission or switched cards with another player.
At registration delegates also had the chance to adopt a toy, if they had not brought one, to ensure that everyone had an opportunity to join in the game.There was also a toy pamper salon where they could embellish and enhance their toys.
Organisers managed numerous alter ego accounts including other toys, toy keynote speakers, and, as had proved popular at previous conferences, an evil Twitter presence.

(page number not for citation purpose)
These were used to help feed a puzzle for delegates to solve as to the identity of the kidnapper of the esteemed keynote Professor Prod Eagle.A toy keynote, a TED talk, was staged with restrictions on only toys attending and a riddle for where they should sit (Figure 3).
After toys were dropped off, a kidnap was staged and clues left as to the real identity of the kidnapper.While there were a large number of toys who attended the lecture, the mystery did not attract huge numbers and it was felt that in some way this was an unnecessary element of the game, especially when so much other activity was occurring.
At the final debrief session for the toy game, delegates devised their own travel games in the form of Twitter challenges.Engagement with this activity was lesser, but the toys still engaged with Twitter at the end of the conference.

Outcomes
A note on the statistics that has been taken from Twitter.Consent to publish has been requested, and obtained, from the owners of the two tweets fully reproduced here.Two further tweets have been anonymised but reproduced in full after careful consideration of Townsend and Wallace's (2016) guidelines.All aggregated data have been anonymised.The tweets from PL highlight some interesting statistics.The pre-conference activities were much more popular with the 'toys' which was unsurprising given that they were the target of this game, with over three times as many tweets from toys as people in the 4 days prior to the start of the conference (Figure 4).
On the middle day of the conference (13 July 2017), a tipping point was reached where people overtook the toys in the amount of tweets.However, over 11 days (including pre-and post-conference), overall, toys tweeted more than people.This high level of toy engagement was primarily those at the conference itself, and with 68 toys tweeting this was a large proportion of those who attended (c.100).The figures from people included retweets and engagements from those following the conference from afar (Figure 5).
While some toy accounts were clearly linked to a human 'owner' and used primarily for fun, others morphed through the conference into 'serious' accounts which were used as the toy 'owners' primary Twitter account (Figures 6 and 7).
Of the top 10 who tweeted, aside from the organisers, 8 of them were toys and only 2 of them were people.The individuals who appeared in the top 10 were extremely active with 200 tweets but there were 388 from 8 different toys.

Evaluation
The game was evaluated in two different ways through a face-to-face session and general paper evaluation.The face-to-face session explored the successes and failures of the game.With regard to Twitter interactions, the comments were overwhelmingly positive.There were a number of delegates who had felt uncomfortable about the whole premise prior to the conference and by the end of the conference wanted the toy Twitter accounts to continue and run into the next conference iteration.Some spoke of the freedom the toy Twitter account gave them and the rebellious activity they felt able to do from their alter ego accounts.Many were even converted to the notion of using Twitter, or resuming their Twitter accounts having entered a hiatus.0 7 / 8 / 2 0 1 7 7 / 9 / 2 0 1 7 7 / 1 0 / 2 0 1 7 7 / 1 1 / 2 0 1 7 7 / 1 2 / 2 0 1 7 7 / 1 3 / 2 0 1 7 7 / 1 4 / 2 0 1 7 7 / 1 5 / 2 0 1 7 7 / 1 6 / 2 0 1 7 7 / 1 7 / 2 0 1 7 7 / 1 8 / 2 0 1 7 7 / 1 9 / 2 0  Approximately, a quarter of the delegate paper responses mentioned the toy conference game in their evaluation of the conference and all of these mentions were in response to their favourite aspect.
There have also been comments in blog posts commenting on the positive aspects of the game.Farley (2017, online) touches upon an aspect of networking that the organisers were keen to use Twitter to promote '[The toys were] a great way to break the ice among delegates, and the toy backchannel on Twitter was fabulous'; and Beaumont (2017, online) discusses the way that the toys '…allow [ed] behaviour that might otherwise have been considered odd or bad' and how the organisers allowed delegates to explore ideas around professional identity.
Many Twitter users in academia blur the boundaries between personal and professional usage, in some cases to the point where it's hard to tell in what capacity they are tweeting (Bowman 2015).However, if one is using Twitter as a professional and connecting with others in this sphere, it means there is usually some balance needed between revealing too much or too little personal information or deciding how far to engage in certain discussions.In the words of Marwick and Boyd (2010, p. 124), 'the tension between revealing and concealing usually errs on the side of concealing on Twitter'.One unforeseen consequence of the PL Twittersphere game was the way in which the use of alter egos allowed individuals the opportunity to break free from such constraints as the audience of the toy's account was complicit in the game creating new opportunities for the individual as well as the wider PL CoP.
The fun and playful atmosphere of the game had an impact on other users of the building.On the first day, the conference exhibitors asked to join in and became prolific mission hunters.On the second day, delegates from another event in the building also asked to join in the game.

9
(page number not for citation purpose)

Conclusions
Hard and fast conclusions are hard to draw about the success of using Twitter to disrupt a conference by bringing in playful elements.However, the authors have been involved in different iterations of similar games and the game run at the PL Conference was a culmination of ideas trialled elsewhere.The following observations were made after the Toy game in July 2017 and could be used to further develop this type of game in the future: • Know your audience: having run games with Twitter elements at different conferences understanding the delegates' motivations for attendance is an important part of the planning.PL is, as the name suggests, about disruption and participation.While other conferences may have such aspects as a theme, delegates may not engage as happily if they see this as a distraction from the main conference.One of the reasons Toy Twitter was so effective is because so many delegates engaged with it, and the fun aspect of it was something most wanted to get involved with.• Don't do too much: while a range of activities is needed to appeal to as many delegates as possible, you can do too much.The TED talk was fun, but the staged kidnap and subsequent puzzle to solve was a step too far with organisers having to spend too much time trying to whip up interest and enthusiasm.For the time it took to plan this activity, there was little interest.• Make it fun: games should be fun and not too much like hard work.They should be easy enough to engage with varying degrees of difficulty in the tasks given and the levels with which you can choose to participate.The kidnap was a case in point where the organisers created a game that became too difficult, either in terms of knowledge, time, or interest, for delegates to engage with.• As organisers, capitalise on it: A fun, informative, useful, spirited Twitter feed is what most conference organisers want.This works to keep the CoP engaged, as well as advertising for the conference the following year.
What is clear from the PL Conference experience is that the innovative approaches to Twitter use changed the dialogue for the conference and encouraged delegates to be more creative and playful in their interactions.The feedback received was extremely positive from both delegates and organisers and has encouraged the authors to continue to experiment with this tool in the future, both at conference events and at the workplace.The PL Conference is unique in terms of willingness to engage with immersive activities, and it will be interesting to see if this can be reapplied into more conventional situations where further can be understood about players and non-players.

Figure 2 .
Figure 2. Example of a Toy Mission card.

Figure 6 .
Figure 6.Example of accounts kept separate throughout the conference.

Figure 7 .
Figure 7. Example of an account where the toy tweeted as you may expect a person to.