The compellingness of the interaction between vision and auditory localization (the ventriloquist effect) was investigated using a discrimination task. A tone sequence was presented either from the same location or from two locations that alternated along the horizontal plane. In synchrony with the tones, lights were presented either at the same or at alternating locations. Subjects had to decide whether the tones alternated or not, thereby ignoring the lights. The main result was that presenting non-alternating tones together with alternating lights increased the number of [HEX 145]alternating'-judgements, even though subjects were told to ignore the lights, and even though they received corrective feed-back after each trial.