Using systems to explain systemic problems in Parable of the Polygons

Parable of the Polygons came out of the left field a few weeks ago. It's an article which applies the ideas of Thomas Schelling, most specifically his ideas of Micromotives and Macrobehaviour, and applying them into a game system to express the problem of small personal bias against diversity leading to greater segregation in the general population. 

It's an incredibly effective way of showing a systemic problem, and builds on foundations it demonstrates to you until you can fiddle around with the model yourself, just to see what happens when you change how much someone wants to have all their neighbours be similar to themselves. 

The article is published by Nicky Case and Vi Hart, and I talked to them about their motivations, how they implemented the model into a game system, and what happened to the cute little green pentagon. 

Before I get into the nitty-gritty, can you explain the intent behind Parable of the Polygons?

Vi: We wanted to show just how easy it is for systemic bias to exist, even in a world where everyone's actions seem faultless.

Nicky: "I hear and I forget. I see and I remember. I do and I understand." ~ Confucius.

Obviously he wasn't talking about half-game-half-articles, but still.

We made Parable of the Polygons to show the mechanics of how small individual bias turns into large collective bias. Talking about systemic prejudice can be scary and confusing. But hopefully, with our cute shapes and interactive models, we've given people a friendly way to intuitively understand how systemic prejudice emerges, and how we can reverse it.

One of the things I think Parable does so effectively is to dissociate from the prejudices of real life by using triangles and squares to make a more salient point. What led to the choice to use those specific shapes? Was there ever a consideration about the connotations of shapes, and particular shapes you avoided?

Nicky: It would've felt really weird to have "black people" and "white people" that you drag around. Or "female people" and "male people". Doing that would also have the unfortunate implication that race/gender are binary and immutable. So, abstract shape-people it was.

Vi: We didn't want to limit the applications of PotP to any particular situation, since the system is the point. I've used simple shapes to represent people in many of my films, and I know from experience that works. And Flatland, by A. Square, sets a historical precedent for allegorically using simple shapes to make a political point.

Nicky: Why a yellow triangle? It's Vi's cartoon persona, seemed like the best way to get The Vi Hart Brand in there.

Why a blue square? I was like, well yeah "square" would be the obvious next polygon to use after triangles. And blue is the complementary color to yellow. (Well.... violet actually.) Apparently, blue squares are a second-class citizen to me.

Then, when we playtested (play...read?) Parable of the Polygons... everyone asked, "hey what about a green pentagon?"

We didn't want to complicate the model, but we did want to add that green pentagon somewhere! It would also reinforce the idea that race/gender's not a cut-and-dry binary categories, so if you scroll to the very bottom of Parable, you'll find this lil' fella hanging out:

P.S: Someone took our code, and actually remixed our model to include pentagons! A great testament to open source/creative commons.

Similarly, by using code and systems to make your point you are essentially able to sidestep presenting any of this as overtly subjective; yes, you made the code and the systems, but they're taken from real life. Is this what led you to this approach in the first place?

Vi: It's possible to code systems in a misleading way, but we worked to make the system as transparent as possible by going through it step by step. Schelling originally used coins on a chessboard, when there's no way for a computer to "cheat". We felt it was important to start with moving the shapes by hand, so that later when shapes move automatically you know there's nothing going on behind your back. Once you understand the algorithm, then we take advantage of computer's ability to do simple things that humans already know hot to do, but a lot of them, really fast.

Nicky: A train of thought I'm currently on is "art meets science", or more accurately, "STEM meets humanities".

And Vi does this all the time, making mathe-musical videos! For me, I'm fascinated with taking models/systems from the arts & humanities, and translating them into game models/systems. Most games taht address social issues are merely themed around them, without addressing them in their mechanics. ((e.g. WATCH_DOGS. And I'll admit, one game I'm in the process of making.) Also, see: Ian Bogost's procedural rhetoric.

But now it seems obvious to me that, hey, games are systems, so what better way to understand systemic issues?

Were you surprised by the behavior of the systems that you set out? Were there any preconceptions you went in with that were refuted by what you'd made?

Nicky: The entire second part of Parable, actually, was made because we'd found a new surprising aspect of Schelling's model! At first we were just going to explain Schelling's Model exactly as he'd created it. But, playing around with variables, we found that just as a small bias can turn a society segregated, a small anti-bias can reverse that!

Vi: We had a clear idea of how Schelling's model worked going in, but our addition of the "nonconformity bias" surprised us with its effectiveness. We knew some amount of nonconformity would mix up the population, but were pleasantly surprised to see it work even with the very minimum level of "I don't want absolutely everyone around me to be like me." It suggests a very clear and feasible goal for every human.

Nicky: That was really surprising, and would also let us end Parable of the Polgyons on an optimistic note -- small local changes really can change institutions from the bottom-up. (and in fact, that's probably the only way real lasting social change is made -- top-down measures tend to be ineffective and intrusive.)

Also, in the Sandbox, we found that higher levels of segregation are possible with sparser populations. Maybe that explains why rural areas tend to be more xenophobic? (Probably not, but it's an interesting hypothesis.)

Whenever you deal with issues of discrimination people tend to be resistant to changing their opinions. How have you found the reaction to Parable of the Polgyons?

Vi: When you talk to people about bias, they often think you're talking about strong individual biases. That's a hard opinion to change, because most people know they're not horribly overtly biased, and they're usually right. We worked hard to make it clear that that's not the opinion we're trying to change, we're just adding new information about systemic bias that's completely compatible with people's existing knowledge about themselves. Personally, I'd much rather expand minds than change them.

Even so, I have come to expect long hateful diatribes in response to even my most straightforward mathematical work. The reaction to PotP has been more positive than anything I've ever done. It's almost disturbingly eerie.

Nicky: It's been surprisingly positive and polite, especially for something that touches on a, uh, touchy subject!

One of the healthiest things we could do for discussions about race/gender is to make it less scary to talk about. Because it's totes scary. I think Parable of the Polgyons helps with this in three ways: 1) it's cute shapes. 2) one main message is to not take it personally, since large collective bias can exist even with small individual bias. 3) shows that combating institutional bias isn't hopeless - it's possible with small bottom-up efforts!

Now that there's a little time since you published it, is there anything you would have done differently or focused more on?

Vi: Quite the opposite, actually. We were planning to tweak some stuff after release, but the reception has been so overwhelmingly positive that I'm hesitant to change the magic formula.

Nicky: "I DON'T KNOW WHY IT WORKS BUT IT DOES. NOT TOUCHING IT." ~ every programmer, ever.

Many people commented that if you turn the bias all the way up, "segregation" will be at 0%, only because everyone's constantly moving around and never settling down. They quickly realized that was just a quirk of the model, but I hope it wasn't distracting. (Maybe we could have even used that to show that massive segregation can't be the result of high bias, because if it was, everyone would be moving around randomly.)

But that's a small problem, and any fixes we could do to it might weaken or confuse the message. Parable's doing shockingly well, so we're not gonna touch it.

Are there any plans to expand on what you've done with Parable?

Vi: Perhaps! :]

Nicky: Peeps really loved the innovate interactive format -- I'm definitely going to explore "playable posts" more! I've already got sketches and prototypes for loads of other "arts and humanities" models I want to port over to an interactive form. Meanwhile, Vi's working on a super top secret* video, talking about Parable & similar models for segregation. Should be out soonish!

*not really that secret.