You all know about Twitch.tv, right? I can’t actually hear your answer, so to be on the safe side I’ll go ahead and explain that it’s a website that makes it super easy to stream your gaming sessions out to the Internet and for Twitch viewers to find those streams. All it takes for PC gamers is a little software configuration, an Internet connection, and a webcam if you want a picture-in-picture showing your grinning face. More advanced users have also been streaming console games for a long time, but the Playstation 4 features built-in Twitch streaming and the Xbox One will implement the feature later this year. So the practice is only going to grow in the coming months.

I’ve thought about creating a Twitch channel, but honestly one of the things stopping me is that I’m afraid of sucking at games in front of an audience. I’m not even talking about hardcore competitive DotA 2 or Street Fighter matches. I’m afraid of loading up ANY game and sitting there going “Herpa derpa DERP DERP!” into the webcam while my avatar just walks into walls. Should I worry about that? Can knowing you have an audience affect your performance in video games?

Research says …yes! It also says …no! Because it depends!

Studies on what’s called “social facilitation” were all the rage during the early 1900s, but they started out noting how competing with others spurred people on to better performance on tasks such as cycling or reeling in fishing lines.1 Unsurprisingly, we tend to perform better on tasks while in competition, which can be chalked up to increased motivation to look good and/or using information about the performance of others to figure out how much we should be doing.

Eventually, though, people began researching what effect the presence of mere spectators had on solo performance –something dubbed “the audience effect.” Early investigations found that yep, having even a small audience caused people to perform better on tasks like solving easy math problems or simple demonstrations of hand-eye coordination.2 Other studies, however, showed that having an audience caused people to perform WORSE when the task was difficult, like memorizing lists of nonsense words.3

What was going on?

The best current explanation is called “Distraction-Conflict Theory” and was proposed by University of Illinois researchers Robert Baron and Glenn Sanders.4 They say that audiences distract us for a variety of reasons, including our preoccupation with what they may be thinking of us. This causes conflict between our intention to perform a task or play a game and our intention to see (or even just ponder) what the audience thinks of our performance. In the case of simple tasks or tasks that we’re really good at, we’ve got mental resources to spare because these tasks require little cognitive juice. Thus we can handle the distraction of the audience and choosing to focus on the task benefits performance.

But say the audience is full of jerks who are screaming at us, or the task is really difficult or novel. In that case the total demand on our attention and mental energy is higher, and performance suffers. Even if the distractions go away, Sanders and Baron found 5 that the drain on resources caused the drop in performance to persist for a while. This is why trash talking opponents in fighting games sometimes works: You’re distracting them and maybe even making them angry to the point where mental energy needed to play the game is being diverted to dealing with your incessant stream of “Yo mamma…” jokes. Other research has shown that if we think the crowd is hostile or judging us, the audience effect intensifies.

In 1993 researchers James Moore and Jody Brylinsky discovered a unique opportunity to test this theory.6 They dug up data from a series of basketball games in the 1988-1989 North Atlantic Conference season when a measles outbreak had caused the team from one school to play a series of games under quarantine –no audience, in other words. It was just the two teams facing off against each other on a weirdly quiet basketball court. By comparing players’ statistics during these spectator-free games with their performance just a few weeks earlier in front of a roaring audience, the researchers showed that lacking an audience caused them to excel in the complex, demanding task of playing a basketball game. Or to put it another way, they choked less.

And while research on the audience effect and social facilitation theory is underrepresented in the video game literature, researchers Nicholas Bowman and his colleagues do provide one recent example.7 In their study, they had subject play games of the first person shooter Quake 3: Arena on either high or low difficulty levels either in front of a small audience or alone. Even after controlling for certain aspects of skill like hand-eye coordination and 3-dimensional spatial ability, the researchers found support for the idea that at least in the low-difficulty condition, performance was better when an audience was there.8

And so I imagine it would be so for someone playing a complex and challenging game FIFA or League of Legends on Twitch.9 Even if you’re not keeping an eye on the chat or the viewer count, you’re still probably thinking about what the audience thinks and this will subtract from the mental resources that are in demand because of the game. But if you’re just screwing around with an easy game, like Minecraft or the latest Lego game, your performance might actually improve in front of an audience.