Mutuality, group dynamics, and the iterated prisoner's dilemma

It seems like everyone is talking about game theory these days. As those who know me can attest, I'm much more interested in game studies (applying literary/social/design theory to games) than I am in applying game theory to life, and my forays into game theory of late mostly result in my getting caught up in the particulars and implications of certain grammatical conventions. But I digress.

After reading Nick's game theory post on cpsrenewal.ca, I decided to post some of my own perspectives on the applications of certain theoretical paradigms to a real world (workplace) context. By the way, if Nick's post got you interested in backwards induction and Nash equilibriums, I wholeheartedly, emphatically suggest that you play Frozen Synapse.

In a recent discussion with a friend regarding symbiotic relationships and inherent motivations for cooperation, we ended up tossing around a few ideas related to the prisoner's dilemma. Often used as an explanation for why individuals might opt not to cooperate even if it is in their best interests to do so, it's less about doing the right thing than it is about minimizing personal risk and covering your own butt. And I thought that this sounds an awful lot like some work environments.

(from Wikipedia)

Cooperating holds the maximum potential gain, but also the greatest potential risk. For this reason, most players will opt to defect (that is, not cooperate) when faced with the prisoner's dilemma. This is true whether you are playing one round of the game or, theoretically, any limited and specified number of rounds.

But that's not what most people deal with in the workplace, is it?

The iterated prisoner's dilemma

The work environment can be more appropriately modelled as a multi-player indefinite iterated prisoner's dilemma. That is to say that the individuals will play a non-specified (potentially indefinite) number of rounds in which each individual can recall the previous decisions of the other players and adjust their own strategy accordingly. The iterated prisoner's dilemma game is fundamental to certain theories of human cooperation and trust. Even in cases where the number of rounds is known (and theoretically the best strategy would be to consistently defect), the establishment of an indirect relationship between the two players results in cooperation being a more dominant strategy.

And research shows that purely self-interested and rational players interacting for indefinitely long games can sustain the cooperative outcome, quite simply because when you're in it for the long haul as opposed to a one-off task, you have a vested interest in both fostering your relationship with other individuals (to ensure their cooperation) and in maintaining your best possible long-term outcomes.

While choosing your own partner (unsurprisingly) expedites the emergence of a mutually cooperative equilibrium, establishing this pattern is possible for people who do not know or like each other, even in cases when no communication between players is possible.

Though hotly contested, the most effective strategy (in terms of personal gain and in terms of establishing mutual cooperation) is perhaps the "tit for tat with forgiveness" approach. Under this strategy, the player defaults to cooperation at the start of the game, and henceforth does what the other player chose to do in the previous round (defects as punishment for defection), while occasionally choosing to forgive instances of betrayal in order to avoid falling into a cycle of mutual defection.

This strategy appears to be relevant in the workplace as well as in theory. The best approaches are those that are nice, assume good faith but are not afraid to react to instances of betrayal, readily forgive, and strive to maximize personal gain as opposed to simply besting the score of another player.

Viewed in such a light, personalities tending towards risk-aversion (which are often regarded as opponents to openness and collaboration in the workplace) can actually promote cooperation; because even though the payoff for selfishness can be higher, the risk is also greatest.

The question is: how do we create a culture where it is obvious to all people that cooperation (or collaboration, if you will) is not only most likely to benefit the largest number of people (appealing to the community-minded) but also that not cooperating comes with a greater risk to the individual (appealing to the risk-averse).

In most cases, this environment already exists. Increasingly, as networking and sharing within communities becomes a great way to access new information and resources, those who choose not to share risk being left behind. The problem is that many who are sitting comfortably with more or less unassailable job security do not perceive this risk as actively as one might if they were in a less secure position. But set on a long enough timeline, everyone's comfy status quo is threatened by stagnation and lack of cooperation.

What do I mean by cooperating, here? Put simply: cooperating means doing your part in as transparent, communicative and accessible a fashion as possible, as an individual component of a network of mutuality.

The network of mutuality

In describing society (or any group, such as colleagues) as an inescapable network of mutuality, I am respectfully alluding to Martin Luther King Jr. and, more recently, Fred Clark as a way of saying that we are all responsible to each other and it is in our own best interests to embrace that responsibility.

"We are all responsible, but we are not all responsible in the same way. We each and all have roles to play, but we do not all have the same role to play, and we do not each play the same role all the time.

Relationship, proximity, office, ability, means, calling and many other factors all shape our particular individual and differentiated responsibilities in any given case. In every given case. Circumstance and pure chance also play a role, sometimes a very large role, as when you alone are walking by the pond where the drowning stranger calls for help, or when you alone are walking on the road to Jericho when you encounter the stranger who has fallen among thieves."

By failing to support the efforts of others, by failing to proactively share information and access to the work we are doing, we are contributing to a culture that will not support our efforts, or share information when we ourselves need it. Call it karma or call it game theory, the bottom line is that it is in our own, individual best interests to take care of each other, whether in the context of a small team or a staff as large as the public service.

But just to make sure, let's examine that premise through another paradigm and think about the really important question: how do encourage this?

Avoiding the tragedy of the commons

Garret Hardin's 1968 article on group dynamics (based on a scenario outlined by William Forster Llyod) cast a dim view of the possibility for group cooperation when it is in the individual's short term interest not to do so. He argued that even if it is in the individual's long-term interest to cooperate and contribute actively to a shared responsibility, individuals will opt for the more short-term profitability of exploiting their peers. (In game theory, this is often called the commonize costs - privatize profits game, credited to Garrett Hardin.)

The tragedy of the commons has been used as an argument for sustainable development, and as an argument against restrictions to private property. But at its heart the commons dilemma is a paradigm that can be applied to any situation in which people's short-term selfish interests are at odds with their long-term interests and the common good.

So, based on this model, how do we account for myopically selfish behaviours in the workplace?

1. Community-building. In this paradigm, motivation to contribute to a shared responsibility is increased with the number of social ties an individual has within a group. When we identify with the other members of our community, we are not only more likely to exercise personal responsibility, but also more willing to compensate for the shortcomings or overindulgences of others. (Lesson for the risk-averse and uncooperative: if you do not invest in people, they will not help you out or cover for you.) Group identity promotes a long-term perspective on shared responsibility.

2. Transparency. Research has shown that people are less likely to take advantage of the group in public situations than in private. Additionally, those who contribute actively to group needs and exercise restraint in the interest of their peers gain greater prestige and influence within the group. Working in a transparent environment (in which information is freely accessible) not only makes people feel accountable, but it also amps up the perceived and actual value of social capital.

3. Ownership. Feelings of accountability to the group actually increase the smaller the group size happens to be. At first this might seem counterintuitive: why should I feel more responsible to fewer people than I would to more? But upon examination of social realities, it makes sense. In smaller groups, you are personally more responsible for the results produced by the group: your work and your choices enact considerable influence. The fact is that not all of us can work in a small group (and not all of us want to), but by tasking individuals with projects that really matter and in which they can perceive the impact of their work will, as I've argued before, result in better work and more productive groups.

3. Low barriers to entry and concrete, attainable goals. To quote Scott Johnson, "if you don't have to pitch in much to fix a huge problem, you'll take that deal. If the solution to the problem seems out of reach, you may consider contributing to be a waste until it looks tractable; then you'll want to pitch in." People are less likely to cooperate if they consider the goals of the group to be intangible (hence the reluctance to cooperate exclusively for the preservation of long-term interests) or difficult. We need to build transparency into the way we work, through using platforms such as wikis and fostering movements that favour openness. And anyone can do this: start sharing your drafts and posting your ideas before they are finished. Make your work findable and contribute to the culture of group-focussed productivity in order to re-align social norms.

5. Leadership. One of the best ways to motivate cooperation is to demonstrate it and lead by example, but there is more to it than that. Several responses to the tragedy of the commons suggest (and I am inclined to agree) that people are more likely to cooperate when a respected person not only endorses and exemplifies cooperation, but also makes it clear that it is expected. When openness, creativity and cooperation are normalized, even the reluctant individuals are more likely to engage because that's just how we roll.

6. High stakes. As discussed above in the context of the iterated prisoner's dilemma, the fact of the matter is that some people are more motivated by what they risk losing than by potential personal gain. While I certainly think we should empasize the benefits of cooperation first and foremost, it's worthwhile to consider the risks of doing nothing in how we frame our value propositions.

The bottom line is that our actions as individuals must be more than efficient and inspiring -- each of us must also strive to be propulsive, for the sake of our own interests as well as those of Canadians at large.