The effects of positive feelings and arousal on privacy decision-making

The goal of this study is to contribute to the underexplored interplay of unrelated positive feelings and arousal and their effects on users' willingness to provide personal data. To this end, we conduct an online survey (n=368) based on a hypothetical social network sweepstake scenario in which personal data, such as a photo, must be disclosed for participation. We perform structural equation modeling based on an extended privacy calculus model. Drawing on the feelings-as-information theory, more positively valenced feelings, even when evoked by an unrelated stimulus, should lead to a higher willingness to disclose personal data and participate in the sweepstake. We examine how arousal influences this effect as well as users' willingness to disclose. We find three significant crossover interactions for unrelated positive feelings and arousal, i.e., for more positively valenced feelings under higher arousal levels users are more willing to disclose data which is in line with the feelings-as-information theory. Surprisingly, less positively valenced feelings under low arousal levels also lead to a higher willingness to disclose personal data. We explain these results by additionally drawing on the affect regulation theory, which assumes that individuals try to protect their feelings in positive affective states, and take actions in order to improve their feelings in less positive affective states. We interpret the arousal level as a "switch" that helps to determine which theory is suited best to predict the direction of the effect of unrelated positive feelings and thus, explain the observed crossover interactions: for high arousal levels, the feelings-as-information theory and for low arousal levels, the affect regulation theory fits best.