Examining the emotions that players seek in videogames gives you a first step towards understanding play. Now you will explore what happens not just with a single player, but also among several players as they experience a game together. The principles discussed in Step 2 are taken from two areas of social science that have a lot to offer in terms of assessing what is happening between people moment-to-moment: social psychology and communication. This step begins with an overview of why social interaction matters for game designers, before looking at the specific mechanisms at work when people play together, recommendations for design choices that encourage and heighten social interaction, and considerations of player differences to keep in mind while designing games.
Game designers and developers might find that they focus most of their efforts on thinking about how an individual player will react to what has been created, moment-by-moment (unless they are designing what they know will be a multi-player game). Certainly there are tremendous challenges at this level of analysis, and this step is not meant to say it isn’t so. Rather, it’s meant to point out a whole additional layer at which a game is being experienced by players, which is fundamental to how they experience the world, and even, fundamental to why they play games in the first place. This is true even if you are designing a single-player game!
How could this be? Well, the fact is, human beings are social creatures at heart. We are not solitary animals—we live in social groups from the very beginning of our lives. Being social and figuring out how to do it well takes up a tremendous amount of our time in life. Biologists and neuroscientists tell us that the basic ways in which we perceive the world—what we notice, how our senses function, even what is happening in the neurons in our brains—is colored by our social nature. Examples of evidence gathered to support this claim include the following:
Even infants orient very quickly to the sound of a human voice. They can recognize their mother’s voice and face at two weeks old, and seem to prefer the sound of their mother’s voice even in the womb [Kisilevsky03].
Because of experimental research confirming that people can recognize faces even when they are at a great distance or highly distorted, researchers have concluded that humans most likely have a neural resource that is specialized for and devoted to facial recognition [Sinha06].
When we see another person experiencing and demonstrating signs of an emotion, we automatically mimic that emotion in our own face, and in our brain, it seems to be the case that special mirror neurons are firing with sympathetic imitations of the muscle movements of that person [Iacoboni99].
Most communication between people is happening not in the semantic content of the words, but in what are called “paralinguistic cues”—body language, facial expression, and tone of voice. So we all have to become experts at reading these signals in everyday life [Knapp02].
This means that anytime there is a social situation going on in combination with the playing of a game, players will be heavily involved in interpreting and reacting to that social situation, and it will color how they experience your game. And more often than not, the player will be in a social situation when playing.
Although most designers are familiar with the growing popularity of multi-player online games, many designers may not realize that most gaming is done in social groups—the Entertainment Software Association 2008 report [ESA08] says that “59% of gamers play games with other gamers in person.” That’s over half of the time!
Why is this so? Perhaps because playing with others is often simply more fun. Some interesting research done using biosensors (hooking up heart rate, sweat, and other monitors to players) showed that players had more emotional signals when they scored a goal against a real human being as opposed to a computer opponent [Mandryk07]. Playing games is not just a form of individual entertainment; it is something that people seek out as a way to be together doing something fun. This means most gameplay exists within a social context. A group of friends is doing all kinds of social “work” when hanging out playing a game together—becoming closer to one another by building trust and shared experience, establishing hierarchies of skill and dominance, letting off steam safely from conflicts among the group, carrying out tensions or grudges within gameplay, and so forth. This was what Nicole introduced in Step 1 as People Fun. Designers ignore all this at their peril, because it can have profound implications as to whether a game will turn out to be fun.
Another powerful reason to care about social play, even when designing a single-player game, is that researchers have shown that people have what you might call an overactive social sense. That is to say, that people end up making use of their social skills even with technologies that act as if they were human. This happens instinctively and despite a person’s conscious awareness that they are dealing with a machine, not a person. Reeves and Nass demonstrated this phenomenon in a series of studies that they called the “Media Equation” [Reeves96]. They tested out findings from Social Psychology using computers. They would replicate a study, substituting a computer for a person in the study, and see if they got the same results. Time after time, they did. For example, they demonstrated that people can’t help being polite to a computer. In one study they had a computer tutor a person, and then that person gave feedback on how well the computer did.
In one condition of the experiment, the participants gave feedback directly to the computer they were tutored by. In the other condition, they went into another room and told a second computer how the tutoring computer did. Reeves and Nass showed that people rated the tutor more positively when telling it to “its face” how it did, as opposed to when they told another. This mimics politeness norms in human interaction. The participants in the study denied that they thought the computer was at all human and certainly were not consciously being polite to the computers. They simply couldn’t help it.
This means that anytime a designer makes a game social at all, such as when there are non-player characters (NPCs) on-screen, they are evoking the player’s social skills in a powerful and unconscious way. It thus becomes even more vital that designers understand the building blocks of social play and social interaction when making design choices.
As Nicole already noted in Step 1, when people play together it evokes all kinds of emotions and behaviors. Here, you will discover some of the key elements of social play.
On a very simple, primitive level, social play introduces a powerful effect among players called emotional contagion [Hatfield94]. Because people are so easily influenced by seeing how others feel, suddenly all of the emotional effects discussed in Step 1 become magnified among players. Someone watching a friend play a round of Katamari Damacy (Namco, 2004), for example, will vicariously experience the emotions of the player as he unfolds in body language, exclamations, and facial expressions (if they can be seen). The spectator will also be influenced by the emotional cues (music, movement, and the like) in the game that are directed at the player. In turn, the spectator’s emotions will have an effect on the player, creating a loop of emotional contagion that designers can take advantage of.
As mentioned, this can also occur when a player encounters NPCs (or has a view of the avatar in a third-person game). Such players will pick up and begin to feel the emotions they see on-screen without realizing it.
Communication and Social Psychological research has shown that a person’s actions are fundamentally altered simply by having someone else observing [Cottrell72]. If a person is already good at something, her performance in the task may get even better when others are watching. If she is not so good, it may make her performance worse. Some people really enjoy performing for others, and get more of a charge out of playing for an audience. Designers can create opportunities for this kind of showing off. Designers can also use the embarrassment of performance to heighten the fun of play—the Wii game WarioWare: Smooth Moves (Intelligent Systems, 2007) is a great example of a game that leverages social embarrassment to powerful effect in terms of fun. Everyone gets embarrassed making the silly gestures that drive the game mechanics, and this can even be a bonding experience among players to share such a vulnerable moment. Interestingly, this kind of social facilitation effect also seems to occur when the watcher is a computer character [Rickenberg00]. So even an NPC audience in a game might enhance a solo player’s experience.
One very important advantage of being social animals is that human beings do not have to learn everything from scratch all by themselves. Instead, each of us gains a tremendous amount of knowledge about the world not just from our own senses, but also from the people around us who already know many valuable things. One way this happens is through explicit teaching—people are mentored by parents, educators, and friends. This happens when people play games together, too. Part of the fun of mastering a game is being able to show off through teaching other players how it’s done. And teaching others to play helps to ensure that the player has others to play with later.
People also learn very effectively merely through observing what others do—Bandura called this social learning [Bandura77]. He demonstrated that small children in a room where a video was being played of a teacher hitting a doll with a mallet, would be more likely later to hit such a doll in this way during their play, compared to children who had not seen this video. Watching is a powerful part of learning how to play a game—making it possible for a person to learn “over the shoulder” should increase the number of new players who find it easy to take up this game.
When people play together, they are usually not just there because they want to engage with the game itself and its challenges. They are also there because they want to be social with the people with whom they are playing. Games provide opportunities for people to achieve important social aims such as:
Building trust and friendship. Working closely on a shared goal, or even fighting one another but in a setting with close interaction, helps build trust and friendship and warm feelings toward others [Tschannen-Moran00]. So at a very basic level, playing together helps to deepen relationships. In a physical sense, gameplay puts people close together and may also synchronize their movements and lead them to share some powerful emotional experiences [Capella96]. All this adds to the deepening of relationships.
Establishing power relationships and alliances. Social groups have internal dynamics that games can support and sustain [Shaw81]. For example, there may be power struggles going on in a group as to who is a leader. The group may work on this question in the context of gameplay—who leads the group in collaborative play? Who wins in competitive play? How do the group members treat one another as they play? Perhaps gameplay becomes a way for some group members to let off some steam or set someone down a notch in a safe context, if they have been tipping the power balance in an unwelcome way, or have done something to irk the group.
Trying on social roles and identities. Gameplay also allows people to do some social identity exploration [Turkle95]. Players can select an avatar that has different social qualities than they do, and can play out those qualities in social gameplay. This kind of play can end up deepening relationships as co-players learn other aspects of one another that may not come across in everyday life and get to try on different ways to relate that may actually enhance their core relationships with one another.
All this has some very practical and tangible implications for game design, and you will now see some of the ways that you can leverage the understanding of social play for the benefit of the development process.
As you already saw in Step 1, part of what game designers aim to do is to create interesting emotions from play. Emotional contagion can be a powerful tool for doing this. If people are in the same room playing in front of a shared screen, a designer can definitely take advantage of that in thinking through the emotional cues in the game, like NPC and avatar actions, music, camera angles and cuts, and figure on the player audience being affected by these as well, which shifts the mood for all the players in a room. Even without a shared screen, when players are playing in the same room (for example, when a group gathers with Nintendo DS systems for a play session), the designers can assume they are hearing one another and seeing one another’s frantic actions out of the corner of their eyes. One important thing to keep in mind is that one can’t always predict how emotions play out socially from observing solo players. It is crucial for designers to watch groups of people playing the game, in order to get a feel for how emotional contagion will play out. Ideally, the team will want to do group playtesting internally among the development team (as Harmonix does [Kohler07]), to really get a feel for these dynamics and how to maximize them. This is true for all of the social effects in this step, and I’ll touch on this again at the end.
In a social group, there is likely to be one or more people who really enjoy performing for others, and whose performance is enhanced by having an audience. Building improvisation, extra moves, opportunities for finesse, and the like into a game allows these people a chance to show off for the group, which will enhance everyone’s experience. True performers will of course find unexpected ways to show off in the game—like the vehicle jumping in Halo: Combat Evolved (Bungee, 2003)—and designers can capitalize on this by spreading the word about how to do these things once they’ve been discovered.
Designers can also reduce the embarrassment of anyone who is especially shy in the group, surprisingly, by requiring everyone to do something that is embarrassing but fun. Performance games like Karaoke Revolution (Harmonix, 2003), Rock Band (Harmonix, 2007), or the aforementioned WarioWare: Smooth Moves require everyone to perform embarrassing actions, and thus end up building trust and affiliation by causing people to become vulnerable in front of one another. Creating variant difficulty levels that are easy to access helps to ensure that people whose performance is impaired by having an audience can still take part and improve at their own pace.
The best possible scenario for quick learning is when players can use social learning and observation—in other words, simply by watching, they can figure out a lot about how the game is played. Games with physical movement linked to game mechanics, like the Dance Dance Revolution series (Konami, 1999 onwards), have this quality. So do games with well-thought-out menus and without intense time limitations, such as The Sims series (Maxis, 2000 onwards).
This may not be possible for complex, real-time focused games without physical controllers. In this case, creating a great tutorial level that offers social learning for observers (for example, through the voice-over instructions and menus that can be seen by the audience) can help. Designers might also consider incorporating a mentor mode in which one player can coach and assist another as that person gets up to speed on a game’s mechanics—something like those cars for driver education that allow teachers to co-pilot when needed. It’s also helpful to create web space where players can upload replays for others to learn from.
If the development team wants a very broad audience for a game that includes casual gamers, it is very important to design for and to test out over-the-shoulder learning, as well as tuning the game for individual player learning curves.
Building relationships is a side effect of playing together—few players choose to play a game with the goal of working on their relationships, but whenever people play together it is inevitable that these interconnections will form. When designing with this aspect of social play in mind, a few key points should be considered:
Build in breaks. This may sound obvious, but isn’t always taken into account—designers need to allow room for socializing to occur in the first place. That is to say, a game that has relatively short play cycles, with breaks in between for things like avatar selection or other selections (for example, a new minigame to try out) allow people to take a break, get a snack, talk about how the last round went, and the like. If designers are working on a game with longer play cycles, they should think about natural break points for groups of players, and how they might handle finishing asynchronously. It’s boring in a group to sit out a very long period in which one has been knocked out of play. This is something the team will want to test out with groups of players pretty early on.
Form factors matter. As discussed earlier, it’s important to think about how the platform a game is played on impacts what people can share. A console game with a big screen that players share allows for more common input than individual handheld machines. A game with physical controllers changes social dynamics as well. Designers should use what they have well!
Encourage trust. There are some social actions that end up contributing to trust and affinity between people, which can be built into a game. For example, getting into synchrony (performing the same action at the same time), being on the same team, not to mention standing close together or having to do some kind of shared physical action. Nass, Fogg, and Moon conducted an interesting experiment in which people were induced to view computers as their teammates, for instance [Nass96], which demonstrates that a game design could help induce a cooperative and trusting environment.
Let the players form their own power relationships. Social dynamics are complex, and far beyond any game designer’s ability to anticipate, but fortunately most games support the establishment of power relationships and alliances better than many of the other aspects of social play. Providing team dynamics that include competition among teams as well as among individuals allows social groups to explore their own power dynamics and alliances fluidly through gameplay. It’s still wise to do group playtesting to see how these dynamics will play out, and to tune the offerings, of course.
Another aspect of social expression in videogames is the opportunity for players to roleplay different identities and try on roles that they would perhaps not normally explore in real life.
The explosion of avatar customization in recent games is a tribute to the importance and value of this social aspect of games. Investing in socially interesting opportunities for customization of avatars for your target audience is one way to encourage and enhance social play. There is little academic research as of yet on the particulars of which specific avatar qualities are more valuable than others, but there has been some preliminary research about how avatar choices are made and what they may mean socially, such as Nick Yee’s Daedelus project [Yee04]. Development teams are probably best off doing some initial prototype testing with the target player groups to see whether they are really offering the right set of “knobs and sliders” to encourage people to forge really interesting and compelling social identities in the game.
Providing multi-layered built-in social groups such as clans or races, as well as guilds, is another way to allow people to try on alternate traits at a broader level than personal appearance and characteristics. It also provides ready-made team-building support in the game world. Although the races in games are usually not directly correlated to real-world groups, of course this brings up issues and challenges (as well as opportunities) related to everyday experiences and understanding of race—there’s a helpful introduction to this topic in the book Race in Cyberspace [Kolko00]. Researchers who study social interaction in games confirm that these kinds of social groupings provide important points of entry and ongoing value in play [Taylor06].
When designing for social play, it is important to keep in mind that the player community is probably not made up of identical hardcore players of the same age and expertise, but more likely, of a range of people with widely varying qualities. Part of designing for social interaction is taking this into account.
When designers are aiming at a heterogeneous group of players, with different interests, aims, and skills, they need to design the game to appeal to this range. They can even use this to establish a niche social gaming market for the game. Two commercial examples are Go, Diego, Go! Safari Rescue (High Voltage, 2007) for the Wii, designed from the beginning to appeal to preschoolers, with some special parts of gameplay that parents direct with a second controller [Bryant07], and LEGO Star Wars (Traveller’s Tales, 2005 onwards), a highly successful series of games known for their family-friendly co-op mode. In the Wii game, the preschooler’s level of motor control is taken into account and the motions are very simple; pressing any button leads to the same action. This is not the case for the parts where the parents step in and play. In LEGO Star Wars, reviewers on the Amazon website report that the combination of the appeal of the Star Wars and LEGO characters, and the approachable difficulty level, make playing co-op mode a great diversion for themselves and their children.
There is also an interesting game created by a research lab entitled Age Invaders that is designed for three generations—grandparents, parents, and children [Khoo08]. In this game, based upon Pac-Man (Namco/Midway, 1980), the grandparents and grandchildren play together, and the parents can monitor from a remote location if they want and make adjustments. The children must stick to tougher timing and more complex mechanics than the grandparents, to keep things fair. Both groups enjoy the physical activity involved in the game. You’ll learn more about difference in player skills in Steps 8 and 9, but it’s always worth bearing in mind that no two players have the same competencies, and this is especially true with games that are intended to be played by players from different age groups.
Designers may also want to consider the various types of player personalities and how they may impact play. I touched a bit on performance and how some people really enjoy showing off for others—there are also personality characteristics that people bring to social interaction that will affect how they want to play in a game. Designers may want to consider personality profiling players, and building in types of play that suit multiple types of players within a single game—an idea explored in an interesting fashion in Alessandro Canossa’s work on play personas [Canossa08].
Although it was mentioned earlier, I would like to reiterate the importance of testing designs early and often in a social setting that is as close to the one the development team thinks the game will be in when released as is possible. It is simply impossible to predict how a game will “read” in a group without seeing it unfold, and this will also allow the team to discover and capitalize on unexpected social dynamics that arise in play. It doesn’t have to be more trouble for the testing team, and it’s possible to include testing among internal team members in early stages to help with this. Such a process can give everyone on the team a good feel for how the game plays in a social context and what needs to be enhanced, fixed, or simply kept true to the end of the development cycle. Companies like Harmonix, which has really succeeded in the social gaming area with multiple hit titles, swear by this method [Kohler07]. Larger commercial game companies have begun doing social testing with players in their labs—for example Microsoft [Amaya08]. It can be done. Developers, please make social testing a part of your team culture!
Over half of all gameplay occurs in social situations—designers ignore them at their peril, even if they think they are designing a single-player game. I hope this step has convinced you of the importance of designing for social play, and given some concrete steps to take to ensure that a game has strong social appeal.