PAXsims

Conflict simulation, peacebuilding, and development

The SIGNAL-to-noise ratio

BASsignal.png

At the Bulletin of Atomic Scientists, Heather Wuest offers some thoughts on the SIGNAL wargame-based research project on deterrence. We have previously mentioned some concerns about playability, and she expresses a similar view:

In an era where I can play complex dungeon crawlers on my iPhone, I was expecting more from the online version of SIGNAL (Strategic Interaction Game between Nuclear Armed Lands), the Project on Nuclear Gaming’s experimental wargame. In terms of playability, SIGNAL’s first big problem is the tutorial; it’s long and confusing. Second, you need to be matched with two other people on the server to play, but when I tried, no one else was playing. Admins are working around this problem by offering “Open Play Windows,” where they hope to have people online and waiting for you to play. But these windows all take place within in the 9-to-5 workday during the week, and SIGNAL just isn’t fun enough to risk getting fired over.

Don’t get me wrong; I was and am still excited about this game. SIGNAL was designed to collect data from as many people as possible, to study strategic interaction in wargaming scenarios. As a nuclear nerd, I found the implications for deterrence theory to be fascinating. Nuclear deterrence is hard to study because when it works, nothing happens, and it is difficult to identify exactly why nothing happened. When talking about nuclear weapons, the act of successfully deterring someone means that in the end, they didn’t nuke you. Not getting nuked is good, but hard to study because you can never be certain that the enemy intended to nuke you in the first place or what specific action or inaction or combination of both led the enemy to come to the decision that allowed your continued existence.

Anyway, I was excited for SIGNAL because it promised a game where people could practice nuclear deterrence, an experience practiced before only by armchair philosophers (like me) and policymakers. The game offers an environment in which data on decisions can be recorded and studied to grow academic understanding of how deterrence works and, hopefully, to be used later to help humans get better at not making nuclear war.

But the game does have playability problems. After making it through the tutorial (which I rage quit once, when my browser froze because I tried to nuke someone, apparently an unanticipated player action at that juncture), I quickly realized that I needed two other players if I were to play the game. Browsing the Open Play Windows, I found none scheduled for times when I was free to play, so I whipped out my short list of reliable and pliable friends to recruit some fellow wargamers. I will be honest; SIGNAL was a harder sell than I was expecting. After bribery (pizza and beer) and much whining as I helped speed them through the tutorial, we were ready to play.

I don’t know what this observation has to say about my friends or me, but gameplay went something like this: After we realized that one of the players did not have nuclear capabilities, that player got nuked, repeatedly, because it was an easy way to deny him/her resources. In the game, a nuke costs the same amount of money as a conventional missile, but predictably does a lot more damage. Once the poor, non-nuclear soul was shoved to the resource sidelines, the two remaining players duked it out over resources on the map. Game scoring is complicated, but having more resources makes more options available to you. The game lasts five rounds, so we just operated under the rule that he who has the most resources at the end of the fifth round wins. I am not sure that back end data collection got much of academic use out of our first couple rounds.

As the night wore on, and we better understood the advantages and disadvantages designated to each player at the start of the game, SIGNAL switched from an exercise in nuking to an exercise in avoiding being nuked. But it took quite some time to reach that deterrence focus….

h/t Stephen Downes-Martin

2 responses to “The SIGNAL-to-noise ratio

  1. dshlapak 11/06/2019 at 5:01 pm

    The biggest problem with SIGNAL is that the incentives baked into its mechanics explicitly contradict pretty much the most important single lesson we’ve learned about how nuclear powers behave: that they will go to almost any length to AVOID using nuclear weapons.

    There are a host of other issues with the game, but the scary part will be if it’s flawed representations of nuclear decision-making, international behavior, and coercive diplomacy actually start being seen as accurate “lessons” to be internalized by students, let alone policy makers.

  2. Rex Brynen 11/06/2019 at 5:46 pm

    I haven’t managed to actually play it yet, for some of the reasons described in the article. However, I do worry about the way in which the game design, interface, and player group (let’s face it, only an unusual group of individuals is going to play it) could skew the findings. I think experimental games can offer useful insight into crisis behaviour, but I’m inclined to see large n studies use simpler and more accessible game designs.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: