At the last three meetings of the Society, we have described MDCC experiments with unmasked cues. When the signal (and therefore the cue) is a sinusoid of frequency below about 1100 Hz, the cue may either aid or detract from the detection of the monaural signal. The direction of effect depends on the phase difference between the signal and the cue. As the cue is made louder, performance at bad phase drops steadily; but over a wide range of cue intensities, performance at optimum phase is near that of the signal‐known‐exactly (SKE) observer. However, performance drops drastically for very loud cues whether the phase is good or bad. In this paper, we consider the effect of masking the cue. If the detrimental effect of the loud cue at all phases were due simply to “contralateral masking,” adding more noise should further reduce performance. If, on the other hand, the loud cue provides some sort of misinformation to the detector, cue masking should improve performance. The latter proves to be the case; the masked cue acts like a softer cue.