Nerfing Trolls: Research for a Safer, Happier Internet

If you are reading this now, clearly you are an internet user.

If you are an internet user, then I suspect you are familiar with the term “trolling.”

Put a bit more scientifically, trolling is a global term used to classify antisocial behaviour online. This includes harassment, racist and/or sexist comments, inappropriate language, threats, and in extreme cases, sabotaging or compromising another person’s profiles or accounts. In the specific context of online gaming, this can also refer to playing the game incorrectly for the purpose of ruining the experience for the other player(s).

This phenomenon exists in all spheres of cyberspace, although it is particularly prevalent in websites with a comment feature (ex: Youtube) and games that involve a chat feature (ex: League of Legends). People who troll will use these opportunities to have fun at others’ expense, ridiculing or berating others, often with an extremely vulgar choice of words. In some cases, what passes as trolling in cyberspace would constitute hate speech offline.

This in of itself is totally unacceptable, but when you look at who and how many people this affects, the situation becomes even direr. The classic MMORPG World of Warcraft (WoW) has over 7 million active players around the world; League of Legends has another 27 million daily players. Youtube has 1 billion active users every month. Many of these people are women, minorities, or worst of all, children. Despite these sites’ and companies’ best efforts, millions of vulnerable people globally are subject to trolling of varying degrees on a daily basis. I’m not sure about you, but I’ve had quite enough of it.

ABOUT ME

My name is Chrissy Cook, and I am a Canadian student finishing up my Master’s at the University of Cambridge. I have been affected by trolling for as long as I can remember, facing sexual harassment for being female and verbal harassment for no particular reason, or more commonly because someone was having a bad game. What made it worse, however, was my feeling of powerlessness in its face. I could report the person, sure, but I knew they would just make another account on whatever game I was playing or site I was using and start up again. When others were the victim, it was even worse; any response was feed for the troll.

That is no longer the case. I have an honours degree in Psychology from the University of Moncton and I will be completing my Master’s in Social and Developmental Psychology at the University of Cambridge next month. I am armed with years of research experience in a variety of domains (psychometrics, cognitive psychology, social psychology, personality psychology, statistical modelling and programming) and the technological tools to get behind the scenes and design preventative measures to stop the spread of trolling and help make the internet a safer place for kids to enjoy the kind of upbringing I was lucky enough to have myself: playing games in safety, online and offline.

THE PROJECT

In the first instance, I will be focusing my doctoral research on online gaming, particularly the MOBA genre (i.e. League of Legends, Dota2, Heroes of the Storm, Smite, etc.). However, I hope to expand into studies that englobe other internet spheres, such as comment sections on public forum websites like Youtube.

So far, researchers have focused on the personality of the trolls. The results are as expected: trolls are a modern manifestation of sadism, generally male, and occasionally psychopathic. However, there are too many trolls out there for them to all be clinical sadists; it is statistically improbable. Even if it were the case, there’s little companies can do to get choosy about who uses their services; they are customers, and short of passing a battery of psychological tests at the outset of account creation, any kind of person can pass unidentified on the web.

Instead of looking at people, I have decided to look at the situation. If not all people who exhibit trolling behaviour are sadists, then there must be something happening that turns them to acting out. My hypothesis is exposure to more trolling from those clinical few. Think of trolling as a disease; if you get bit by a zombie, you turn into one. People aren’t all evil – they’re just getting infected with a nasty social bug.

To wipe out the virus, I’ve devised the following five step core research questions:

• 1.How does trolling spread through friendship networks? If we figure out how it spreads, we can have a better idea about how to stop said spread.
• 2.Do players behave differently outside of their network? For example, if there are key people in a friendship (the popular ones, shall we say) from which the bad behaviour spreads, then companies can enforce longer, stronger bans for key people to halt the spread.
• 3.Does prosocial behaviour act like antisocial behaviour? If good behaviour spreads like bad behaviour, we can focus on encouraging those really awesome nice people online, bolstering them up to brighten the days of more players.
• 4.How long does it take to become nice? By this point, we would know how many games it takes to become a troll. It would be time to figure out how many games it takes to become a knight in shining armour.
• 5.Does trolling work the same way around the world? Everyone deserves a troll-free experience, no matter where you live. If we need to come up with culturally-specialized models of behaviour to do then, then so be it.

These will take many studies over the course of my PhD (the next three years), and unfortunately, it isn’t cheap.

COST BREAKDOWN

The total cost of this project will come up to £75,900.00 . This is no small amount of money, so I want to make sure you know the cost breakdown before you decide on donating or otherwise.

• First off, this is for three years. The cost per year is £25,300.00 .
• £13,100.00 p/year goes toward tuition (University of Edinburgh) – this gives me access to other researchers and state of the art equipment with which to perform the research.
• £7200 p/year goes toward keeping me alive and healthy – rent, food, transportation, etc.
• £5000 p/year goes toward research costs – this would be paying participants for studies, buying any additional equipment required for more complicated studies, and going to present the research at conferences to get more researchers interested in tackling this ongoing problem.

If through your generosity I receive more than this minimum, any additional funding will go toward hiring research assistants and other research expenses.

WHAT YOU GET

Apart from the satisfaction of helping out a student in need, ideally you will get a gradually better internet/gaming experience over time! This is one first step forward in solving the problem. Companies have done their best, but we have to go beyond banning if we want to out the root of trolling. On a more concrete level, I will also be maintaining a webpage describing the research’s progress and, where applicable, giving you access to any articles that come out of this project. I will post a link to that page once I have begun my program.

Academics are still hesitant to fund these kinds of studies, so I really need your help to make this happen. If we can get this off the ground, more and more people around the world, academics and gamers and internet users alike, will be exposed to the hope for a solution. We just need a first step forward.

NOTE FOR COMPANIES

If you are a company interested in sponsoring or collaborating on this project, please contact me directly at [email protected]

EXTRAS

Link to the music in the video: http://freemusicarchive.org/music/Chris_Zabriskie/..

Link to the research website: https://nerftrolls.wordpress.com/2015/05/29/bonjou...