This study presents 2 experiments investigating whether listeners with cochlear hearing loss (hearing impaired; HI) and listeners with normal hearing (NH) show differential susceptibility to masking in double-vowel identification. Experiment 1 addressed how double-vowel perception changes as a function of differences in fundamental frequency (deltaF0) of 0 and 2 semitones and the relative amplitudes of the constituent vowels (target-to-masker ratios: -10, -5, 0, 5, 10 dB). When deltaF0 is 0 semitones, listeners in the HI group often perceive the presence of only 1 vowel, whereas listeners in the NH group generally perceive the presence of 2 vowels. In both groups, deltaF0 benefits target-vowel identification, with the greatest benefit occurring when the target-to-masker ratio is -10 dB. When identification rates of specific vowels (in d') were rank ordered, different patterns of vowel dominance were found between the NH and HI groups. In Experiment 2, the effects of deltaF0 (0 to 4 semitones) were compared for monaural and dichotic presentation of double vowels. Both groups show significant dichotic benefit. In addition, individual listeners in the HI group showed trends toward greater dichotic benefit. In both experiments, identification of competing vowels was significantly worse in HI listeners. The results of this study support the idea that increased susceptibility to masking is a primary factor underlying the degraded double-vowel perception in listeners with hearing loss.