According to learned helplessness theory, uncontrollable stress is assumed to be a critical etiological factor in the pathogenesis of depression. In contrast, unpredictability of stressors is assumed to facilitate the development of sustained anxiety. Despite the frequent co-morbidity of depression and anxiety disorders, these two factors have rarely been studied simultaneously in humans. Therefore, we investigated whether there are interaction effects of uncontrollability and unpredictability on anxiety response in healthy participants. Seventy-nine healthy participants performed a visual dot probe task with emotional faces, while receiving mild electrical shocks in four different conditions (2 × 2 factorial design). In (un)controllable conditions, participants were (not) able to attenuate shock intensity. In (un)predictable conditions, participants were (not) able to anticipate shock occurrence. Before the experiment, participants' subclinical depression and anxiety scores were measured using the Beck Depression and Anxiety Inventories (BDI/BAI). During the experiment, continuous skin conductance and self-reported state anxiety were assessed and attentional biases towards angry faces were calculated. As expected, participants showed greater anxiety in uncontrollable compared to controllable and in unpredictable compared to predictable conditions. Additionally, anxiety decreased within the test sessions in participants with low BDI/BAI scores but not in participants with higher BDI/BAI scores. Most importantly, controllability and predictability interacted with each other and with BDI/BAI scores with regard to anxiety. Our results provide evidence that uncontrollability and unpredictability of stressors not only have separate but also interaction effects on several anxiety measures in susceptible individuals and may provide insights into the psychological mechanisms underlying a depressive/anxiety co-morbidity.