In masked priming paradigms that incorporate “online” prime visibility judgments, participants are required to respond to the prime on each trial, either through a subjective report or an objective discrimination task. Our research is based on the observation that paradigms involving such trial-by-trial prime assessments constitute dual-task situations. We conducted three experiments using metacontrast masking to investigate unconscious priming from a dual-task perspective. We manipulated different characteristics of the direct (prime-related) task to examine their potential influence on the indirect (target-related) task and the resulting priming effects. In Experiment 1, we manipulated response modality and the complexity of the direct task, which involved a subjective perceptual awareness scale. Specifically, we compared a unimodal response modality condition (manual-manual) with a cross-modal condition (manual-vocal), and a high-complexity condition (4-point rating scale) with a low-complexity condition (2-point rating scale). Experiment 2 aimed to replicate the findings from the complexity manipulation, but instead of varying response modality, we varied hand usage, comparing a one-hand condition to a two-hands condition. Experiment 3 was a variation of Experiment 2, featuring different stimulus materials and employing an objective direct task, requiring participants to identify the direction of the prime arrow. As expected, response times were longer in the dual-task compared to the single-task condition, while the priming effects showed no clear pattern. Consistent with previous research, we found dual-tasking costs in the unimodal and high-complexity conditions. However, unexpectedly, the two-hands condition showed an advantage. We will discuss implications of these findings and suggest directions for future research.