Frustrating the user on purpose: a step toward building an affective computer

Using social science methods to induce a state of frustration in users, we collected physiological , video and behavioral data, and developed a strategy for coupling these data with real-world events. The eeectiveness of the proposed strategy was tested in a study with thirty-six subjects , where the system was shown to reliably synchronize and gather data for aaect analysis. Hidden Markov Models were applied to each sub-ject's physiological signals of skin conductivity and blood volume pressure in an eeort to see if regimes of likely frustration could be automatically discriminated from regimes when all was proceeding smoothly. This pattern recognition approach correctly classiied these two regimes 67.4% of the time. Mouse-clicking behavior was also synchronized to frustration-eliciting events, and analyzed, revealing NN distinct patterns of clicking responses