Apparent motion is an illusory perception of movement that can be induced by alternating presentations of static objects. Already in Wertheimer's early investigation of the phenomenon [Wertheimer, M. (1912). Experimentelle Studien über das Sehen von Bewegung. Zeitschrift fur Psychologie, 61, 161-265], he mentions that voluntary attention can influence the way in which an ambiguous apparent motion display is perceived. But until now, few studies have investigated how strong the modulation of apparent motion through attention can be under different stimulus and task conditions. We used bistable motion quartets of two different sizes, where the perception of vertical and horizontal motion is equally likely. Eleven observers participated in two experiments. In Experiment 1, participants were instructed to either (a) hold the current movement direction as long as possible, (b) passively view the stimulus, or (c) switch the movement directions as quickly as possible. With the respective instructions, observers could almost double phase durations in (a) and more than halve durations in (c) relative to the passive condition. This modulation effect was stronger for the large quartets. In Experiment 2, observers' attention was diverted from the stimulus by a detection task at fixation while they still had to report their conscious perception. This manipulation prolonged dominance durations for up to 100%. The experiments reveal a high susceptibility of ambiguous apparent motion to attentional modulation. We discuss how feature- and space-based attention mechanisms might contribute to those effects.