Recently, several people have asked me exactly what it is I’m doing here. “Just working. Doing some research. You know, usability stuff.”
Lame, I know. But when it comes to work, its hard to know how much is enough and how much is too much. That’s the good thing about a blog: no one’s being held captive. If I’m boring, you can just hit the back button and get back to reading Dlisted.
I work in the Affective Technology Research Group at the Open University of Catalonia. Our goal is to understand how a person’s emotional response to a learning environment influences their engagement and desire to learn. To do that, the team is continually evaluating new tools designed to measure emotion (a notoriously tricky endeavor). From observation techniques (quite literally pulled from a script of the t.v. show Lie to Me), to physiological sensing tools like face readers and galvanic skin sensors to neuropsychological measures like EEG, the team triangulates different methods in their pursuit of reliable insight into learners’ expectations, feelings, aspirations, desires, interests and aesthetic preferences.
Of course, it would be a whole lot easier (not to mention cheaper) just to ask people how they feel, but it turns out human beings hide and repress our emotions for a million big and small reasons. We lie. We forget. We just don’t know. Even in seemingly mundane tasks, emotions can be quite complex – a messy mix of different high and low intensity feelings that are difficult to sort out on a therapists’ couch, let alone in a usability lab.
My main project at UOC is a research study to evaluate a self-response survey method called “affect-tagging”. It looks something like this: Bring 50 students into a lab and record them performing tasks in the UOC’s Virtual Campus – the online portal through which all content delivery and student/teacher interactions are facilitated. After the tasks, ask them to complete a survey about their experience -- half get a verbal survey and the other half get a non-verbal survey using animated characters depicting a relevant set of emotions: happiness, desire, fascination, satisfaction, sadness, disgust, boredom and dissatisfaction, each relating to different aspects of the portal’s usability, likeability and aesthetics. We then compare their self-reported responses against an expert emotional heuristic assessment (Again, see Lie to Me) or biofeedback data (Again, face scanners and skin sensors). That's it. Measure, analyze, rinse, repeat. The goal is to find the self-report tool (cheap and easy) that most reliably captures a user’s true emotion, as indicated by the other, more expensive and time consuming methods.
Before I left, I told my friend Carol a little about the work I’d be doing. Her first question was so obvious, yet I hadn’t really thought about it. “How will you do research on a portal written in another language,” she asked “and with users who don’t speak English?” Carol, you’ll be interested to know that I still haven’t quite figured that out yet.