In the Lives of an Avatar

Does an avatar's life begin at conception? That moment, where one imagination is fertilised by another?

Being mere models of narrow aspects of reality, avatars have no sanctity. We need not try defining some objective milestone dividing the 'living' from the non-'living'. The important question, particularly with games, is subjective: when do we start feeling attached to avatars? While I agree that 'immersion' is a poor goal for game developers, describing a game as a whole as 'engaging' is petty praise. While the absence of such an adjective would be a chirping canary, its use leaves one wanting: what exactly separates a given game from the immense crowd of engaging media?

On top of engagement, attachment requires consequences to matter. Does a game instill awe of ingenuity or courage? Make you feel remorseful for your actions? Add to or challenge your understanding of the world? Content can be engaging - attachment demands context.

From Brenda Laurel:

I used the Renaissance Faire as an example of how the clever selection and arrangement of materials (to quote Aristotle) could predispose individuals with differing traversals through the space to have dramatically satisfying experiences. This moves the notion of the dramatic from a line to a field. Interaction designers can also think about what sorts of predispositions are set up by the arrangement and potential ordering of experiences and encounters. [emphasis mine]

Having played through Heavy Rain recently, this quote resonates. It suggests that good design is good curation. This is bound to become an increasingly limited perspective of design as technology advances. Design-as-curation assumes both a stable "arrangement", and 'look but don't touch' (or only-touch-in-exactly-the-ways-we-expect) behaviour from players. This kind of context is fragile to transgressions of unspoken (and not just agreed-to) rules. It exalts the designer and infantilises the player.

Since an essential relationship between designers and players is one of trust, the question is how such relationships are established. The traditional - sometimes questionable - response has been to make trust unnecessary, for example by empowering players with better means to express themselves in interesting ways, yet (intentionally or not) limiting the harm they can do to their own or others' experiences. In digital environments at least, the expressiveness (and damage-control) of such kinds of 'sandbox' environments is sure to increase over time.

A wiser approach may be organic forms of social engineering. Expecting trust in both directions, rather than just from players to creators. Often social experiences are reward enough in themselves for good behaviour. I would love to see social science research on the feasibility of more explicit rewards for people who would otherwise need sanctions. Brenda Laurel talks about communities with effective soft sanctions, and seems quite optimistic about being able to replicate their results on a larger scale. The legal system has often taken a harder line; consent is very powerful: a trust that has turned, for example, acts of graffiti from vandalism in to expressions that can enliven urban environments.

Combined, these practices can be seen as ways of helping communities (or systems in general) become more robust - even antifragile - to unexpected events. This is the skill of game/interaction designers: fostering serendipity in emergent systems.

Instead of technology (no offence to the programmers), Heavy Rain looked to non-linear narrative to achieve this - better than other narrative-oriented games I've seen - in the most emergent system known: the human brain. Even when a narrative is predefined, so long as you can empathise with characters the mere sense that one could have done differently can be enough to trigger remorse. Games can take narrative further, by encouraging and allowing you to change your behaviour in later, similar parts of the game. To see different consequences within the same world. To feel growth, rather than just be witness to it, and not just by seeking a better 'score,' but a 'better' story.

I have felt attached to game characters before, but their nature as levers for advancing a narrative - however enlightening of oneself or society these narratives may be - rather than challenging my motivations or reasons for doing, means they have a very different appeal. They deal with the large and intimidating rather than the small and malleable. While both experiences are obviously important, the former are most faithfully (though perhaps not as entertainingly) conveyed via audiovisuals or text.

To make player actions matter, Heavy Rain's creators had to give up a large amount of control over the progress of the story. A lack of 'game-overs' places definite limits on the way stories can be told; for want of a term, I refer to such narratives as 'persistent.' For example, If a designer relies on the player's reptilian reflexes in combat situations, they are forced to choose at which points in the story it is acceptable for a character to die.

Persistent narratives demand that we question the idea of closure: apart from cutting the power, what circumstances lead to a game's conclusion? While it depends on the nature of the story, without a combat mechanic - or a very limited one - it is easy to consider other ways for a story to end: characters could be aged (remember Jason Rohrer's Passage?), exposed to pathogens, fired from a job, whatever. Most generally, closure is the point in a game when all player characters have permanently changed to non-player characters. In a standard narrative this general perspective of closure is pretty mundane, in a persistent narrative it would endow designers with huge creative freedom.

Given that death is such a permanent change for a character, it is interesting that its presence in games has been so chronically unreflective. "Health bars" are such a shallow facade - a one dimensional gap between alive and dead. The change from one to the other is isolated to such a narrow band of time - and with game-overs can happen so frequently - that the idea is sucked of almost all meaning. It is much easier to give the idea meaning in manual than automated, repetitive storytelling. So far, attempts at automation have meant giving a character some form of blurry vision or fatigue (perhaps with the screen donning an increasingly red hue).

So long as games are oriented around skill, and developing the twitch reflexes so as to die less, this immediate link between cause and effect is useful if not necessary. At the same time, it makes 'how?' questions stupid, and doesn't invite anyone to ask 'why?' More cerebral choices offer both the player and the designer more latitude. Passage creates closure with a five minute timer in the background and a rapidly aging character in the foreground. Other games could offer branches in the story for the player (and creator) to decide how - and more significantly, when - they experience (or create) the transition.

A friend of mine wishes for a game that would create content on-the-fly as the player goes. Rather than wait around for a personal Robin Williams clone, for now it is a worthwhile thought to ponder. 

Games broaden the definition of 'content' beyond any other medium, because players are able to develop meaning from more than what they have experienced via their senses. Content is also the experience of what they didn't do (and as a result, didn't sense; a parallel to the economists' opportunity cost). This understanding can build on itself by impacting ongoing play (and later missed choices).

In a perceptive discussion between Peter Thiel (PayPal co-founder, all-round smart guy) and Garry Kasparov, Moravec's Paradox is drawn from the annals of Artificial Intelligence. They use the paradox to highlight a predilection within Silicon Valley to try and automate everything, when many problems would often become much more tractable by offloading some of that work to humans. I think, on a smaller scale, that this becomes even more true for games.

Duolingo makes for a powerful case study of this.

Algorithms replace designer ego with player ego, and so does offloading content creation. Both are desirable, but the need for a coherent story also limits the amount of creative effort that can be offloaded. Economically, it is also a matter of incentives: offloading creates a need to give work a playful face. With Duolingo for example, the desire to learn a language is more engaging than any of the game mechanics it uses. It is also instructive in that the learning experience can be influenced by the actions of the global community, without making the game 'multi-player' in the classical sense.

Games dissolve many differences between creation and consumption, leaving chiefly the contrast of incentives - and therefore power - to express. Exploiting the paradox, this form of crowdsourcing therefore offers particular benefits to smaller developers.

The paradox is also echoed in the nature of games that inhabit each genre. Put more forcefully, the choices made in the above balancing act delimit the kind of game one can make. Existing sandbox games, like The Sims, edge closest to achieving my friend's dream. At the same time, they very strictly limit the gamut of experience that players can have, in turn precluding them from deep attachment.

If dying between the panels condemns a comic-book character to a thousand deaths, the lives (and deaths) of an avatar are magnified in potential significance still more. Design decides whether this potential becomes reality. That can and should include all of us; I can't wait to see where we go with it.