Humans of Speed: The Video Game Theory of Travis Ross

November 13, 2017

A relatively recent addition to the JB Speed School of Engineering, Travis Ross is focused on the intersection of engineering and the human experience. Working originally in Undergraduate Affairs, Ross is now Director of Data Analytics, a position that puts him in charge of our data, including student data, graduation numbers, courses, enrollment, etc., effectively any and all statistical data in the school in an effort to create consistent definitions of our data terminology. During his undergrad, Ross pursued Computer Science, which he ultimately parlayed into a Master’s Doctorate in Communication and Cognitive Science, a degree that culminated in a dissertation involving social engineering and game theory through video gaming.

A New Direction

Following the completion of his undergrad, Ross took a position in the corporate world as an application developer/computer programmer at Samtec. That experience provided him the opportunity to know what he didn’t want as a career path, launching in an altogether different direction away from industry.

“I decided I didn’t like the way that things were going, so I joined the Peace Corps. I managed a computer lab and taught computer programming in the middle of the rainforest, in a little gold mining town called Bartica, Guyana. It’s only accessible by boat and part of the Essequibo river system. The country is majority west Indian and African,” says Ross.

He continues, “Cognitive science looks at how anything comes to know. I wanted to move more towards social science because I was fascinated by people in the Peace Corps. It made me want to find a way to do more than just work with computers. I was drawn to social science, to see how people interact with technology.”

Video Game Theory

The culmination of his research, his dissertation looked at social norms as a tool for changing behavior in game environments, a social engineering premise as filtered through his background in computer science. Utilizing economic game theory and the Institutional Analysis and Design Framework, a multi-level theory developed by a Nobel laureate Eleaner Olstrom, Ross’ work looked at groups of individuals and their preferences in a structured context. Through those preferences he examined the interactions between personal choice and value by means of a structured, mixed-motive game embedded in a video game called torchlight. 

He explains, “In a mixed-motive game, there is an incentive for someone to be selfish, but as a group it’s best that everyone work together. An easy example of this is the prisoner’s dilemma each prisoner is told if they talk then they go free, but If they both talk they get 10 years in prison, if neither talks they each get 5 years. If one talks they go free, but the other gets 10 years. In this game there is always an incentive to talk, but the group does best if no one talks.”

My game was a multiplayer mixed-motive resource management game, the system was designed with simplicity in mind. Each player had a pool of money available evenly to all players. Individually, players purchase equipment which is ranked according to its value. After you buy equipment, you travel through a structured dungeon crawl; the better the gear, the better the rewards. 

Ross says, “We were trying to do two things: to test norms as a means for increasing pro-social behavior in game communities and to look at how to sustain norms with different tools like sanctions. That was one aspect. The other thing was to take other work that had been done on norms, which was purely an in an economic laboratory setting, which was very simple. You have this many tokens, how do you use them. The goal was to take it to the next level. To take it into a place where you have still structured rules, because it’s a game, but to start to look at video games, games that have rule sets that aren’t as sterile as game theory in a lab.”

The Greater Good vs. Selfish Desire

Ultimately, the better equipped a party, the better the overall gain. However, anti-social behavior, someone working for their own best interest, would result in a higher net for the individual, as the experiment paid out actual money to participants. At most, you could make a total of $21 dollars if you cheated, although the total amount of money decreases on average if everyone cheats, lowering the available resources; mutual cooperation always yielded the highest award for the group.

“We found that norms can be used as a means to initialize pro-social behavior, but People will get greedy. Over time, all it takes is one bad apple. Once one person goes bad, the game evolves toward antisocial equilibrium, and it is really hard to get back to prosocial behavior. People also take advantage of the rules, you would get this stuff where people would step out of line at the end, because they couldn’t get sanctioned. Communication channels and sanctions are definitely key tools for sustaining cooperative norms,” says Ross.