First off, on my own personal level, I am feeling guilty right now.

I declined to accompany Felicity to Overeaters Anonymous for the second week in a row tonight, and I feel terrible about it. She sounded really sad and depressed when I talked to her and told her I was not going tonight, and honestly, I feel like shit right now.

I feel selfish and callous and petty and gross. I am so sorry Felicity. Next time for sure!

So that’s my own toxic behaviour. Lesson learned, I hope. But the main thrust of today’s entry is discussion of this very interesting article about the science of toxic behaviour in online gaming.

It’s from the wonderfully named blog Gamasutra, and it discussed, in some depth, the real science being done to study how this “toxic” behaviour that we hear about (verbal abuse, bullying, offensive language, etc) pervading online gaming comes from and how it spreads.

The science was done by Jeffrey Lin, Lead Designer of Social Systems for League of Legends, a massively popular online game where teams of good guys and bad guys vie for supremacy in a very fast paced multiplayer battle arena.

The first thing he and his team discovered is that this toxicity did not originate from a hardcore group of regular “trolls” that make life miserable for others all the time, and enjoy it.

I confess, this was a surprise to me. I just assumed there was a group of assholes peeing in the pool for everyone. Perhaps I was falling prey to the psychological error of attributing others’ behaviour to permanent attributes rather than temporary circumstances.

Well I was dead wrong. Lin’s team found that, instead, this negative behaviour originated most often in players with no previous pattern of toxic behaviour and who were not any more likely to be toxic later on.

Instead, the data suggested that the originators were otherwise nontoxic people having a “bad day”. This aligns with other social behavioural studies I have read about where a lot of the sorts of behaviour we think would only come from terrible people comes from, instead, normal people having a “bad day”.

So if the source is so intermittent, how does toxicity spread? Is there some kind of ripple effect, where one person’s bad day pollutes an entire channel of communication?

That would make sense to me. Often, one person’s angry, aggressive behaviour makes others angry and aggressive as well, both via empathy and purely in self-defense.

This led to an even more interesting result :

…investigate this idea the researchers conducted an experiment in which cross-team chat, as one of the main venues for negative interactions, was made optional for individual players. And indeed, they observed a significant decrease in all measures of toxicity (offensive language, obscenity, and displays of negative attitudes). Moreover, the total percentage of games using chat remained the same (only 46-47% included no chat, both before and after). Lin and team therefore concluded that shielding players from toxic behavior can in fact prevent it from spreading.

So simply by giving people the ability to turn the chat option off, they improved the standard of behaviour. Presumably, once everyone realizes that people can just stop listening to them at any moment, they realize that they had better behave or they will lose their audience.

There is a similar option in IRC, where users can “ignore” other users, but from what I have seen, that is only somewhat effective. The whole point of IRC is chat, after all, and people acting out only need one person listening to get the feedback they want.

Taking the social experimentation to a marvelous new level, Lin’s team then added a Tribunal to the game, where players could vote on whether or not another’s behaviour was worthy of being banned from playing the game, and for how long.

This harnesses the wisdom of crowds. It might surprise a lot of misanthropes, but most people are good people most of the time, and the thing about toxic behaviour is that most people can recognize it right away and most people will agree on what it is if given concrete examples.

Sure, there might be vindictive people who vote the maximum penalty every single time because of their own unresolved anger issues. And there might be softhearted people who vote the opposite. But those two minority groups will cancel each other out and therefore have nil effect on the final outcome.

I think this sort of research has implications far outside the world of online gaming. Bad behaviour is a problem throughout society, and while much has been written about the sorts of bad behaviour considered to be a breach of the social contract and therefore criminal, we know relatively little about the little crimes of manners, consideration, behaviour, and so forth that we merely call “rude”.

And it is very tempting, when you meet someone who is being very rude and unfair to someone else, to assume that this is a horrible person who is like this all the time.

It’s easy, it’s quite satisfying, and it’s very human. After all, we are making the judgment based on all the information that we have. The fact that this is not nearly enough information to make a judgment is irrelevant. We humans do not have the luxury of always waiting for enough information. We have to make judgments and decisions and act upon them in realtime, inferring as much as we can from whatever facts we have, and that leads to snap judgments of others based on very little info at all.

But maybe that person is just having a “bad day”. the sort of bad day that any of us might have when our lives are filled with work and stress and we are tired and cranky.

When it is us, we have all the information we need to know that an occasional sharp remark or outburst of temper comes from many complex situational factors, and not from a fundamentally rotten personality.

We should strive to show such understanding to others.