Two distinct routes for inducing emotions in HCI design

While there is a proliferation of research on emotion detection, little research has focused on emotion induction in HCI literature. We propose an emotion-induction model that points out the two fundamental design goals (i.e., sensemaking vs. exploration) as antecedents of the four different types of emotions, which, in turn, introspectively reinforce cognitive outcomes (i.e., attitude; behavioral intention). Design elements that facilitate the sensemaking goal (i.e., coherence; legibility) induce emotions along a low-arousal-positive (LAN) to a high-arousal-negative (HAN) continuum (e.g., “quiescence” to “agitation”). By contrast, the design elements that facilitate exploration goal (i.e., mystery; complexity) induce emotions along a high-arousal-positive (HAP) to a low-arousal-negative (LAN) continuum (e.g., “cheerfulness” to “dejection”). Among them, only the two types of emotions (i.e., HAP; HAN) have a significant mediation effect of reinforcing attitude. The HAP emotions (e.g., cheerfulness) signal a success in exploration efforts. The HAN emotions (e.g., agitation) signal a failure in sensemaking efforts. Thus, the design elements for sensemaking goals are responsible for the emotions that enhance positive (but not negative) attitude, while the design features for exploration are responsible for the emotions that enhance negative (but not positive) attitude. Incidentally, our data reveals that people place more weight on emotional signals that are consistent with their concurrent motivational concern than those inconsistent with it.