Measuring Frustration through Facial Recognition: We Do It for Games, Why Not for Courses?

by Hap Aziz

While attending the 2012 Game Developers’ Conference last month in San Francisco, one of the presentations I sat in on was about measuring the level of frustration game players go through during certain game play bottlenecks. The presentation, “Arrggghh!!! Blending Quanititative and Qualitative Methods to Detect Player Frustration,” given by Janus Sorensen of Crystal Dynamics/IO Interactive (Square Enix), laid out the qualitative (observing and interviewing players) and quantitative (automatic data gathering) methods of research analysis that he used in assessing player frustration in the game Kane and Lynch 2: Dog Days. The methods were executed using a fairly standard academic methodology of observation and interviews, and while the results were meaningful, the manner in which they were obtained was labor intensive and not well-suited for broad application. And neither were the conclusions transferable to other games.

I wasn’t too surprised to hear that a member of the hacker community was working on a project to use the Xbox and a web cam to “read emotions” while watching television or playing video games. Dale Lane writes about his experiments on his blog located here. Using relatively simple and accessible technology along with some good ole’ programming ingenuity, Dale has crafted a way to roughly “measure” emotions through facial expressions. While much refinement still needs to occur, this is a great start to effectively gathering quantitative data regarding emotions evoked during different segments of game play… and thereby facilitating improvements in the games tested for this type of player interaction.

So why not apply this technique to online or computer-based coursework in the teaching and learning environment? It seems to me that we might be able to make some real and meaningful improvements to course content if we understood where students are struggling and frustrated as well as where students feel confident and happy about the way in which the content is presented? Perhaps I need to reach out to Dale and see if he’s up for a partner project….

Advertisements

2 Comments

Filed under computer games, education, education course content, education technology, emotions, future technology, games, Hap Aziz, qualitative research, quantitative research, technology, Xbox

2 responses to “Measuring Frustration through Facial Recognition: We Do It for Games, Why Not for Courses?

  1. Karen

    Is frustration always a bad thing? Can frustration motivate us to solve some sort of problem? Or is it a complete turnoff to the activity at hand?

    • Part of the answer to your question hinges around the level of frustration as well as how that frustration is addressed, if at all. If frustration is consistent around a particular theme or area, what we can infer is that there is little effective learning or understanding going on in that area (i.e., the frustration is not decreasing as the player/learner gains experience or knowledge). In this case frustration is bad, and measurement can point us to what needs to be addressed.

      However, there is also a distinction between frustration and challenge. In both games and education, we want people to be challenged. Challenge is a motivator that keeps people returning to an activity to try to master it. Frustration is more like the dark side of challenge, and may tend to turn people away rather than to “challenge” them to try again.

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s