Behavioral responses for double-vowel stimuli. (A) Accuracy for identifying both tokens of a two-vowel mixture. Performance is poorer when concurrent speech sounds contain the same F0 (0ST) and improve ~30% when vowels contain differing F0s (4ST). (Inset) Behavioral FO-benefit, defined as the improvement in %-accuracy from 0ST to 4ST, indexes the benefit of pitch cues to speech identification. F0-benefit is stronger for clean vs. noisy (+5 dB SNR) speech indicating that listeners are poorer at exploiting pitch cues when segregating acoustically-degraded signals. (B) Speed (i.e., RTs) for double-vowel identification. Listeners are marginally faster at identifying speech in noise. However, faster RTs at the expense of poorer accuracy (panel A) suggests a time-accuracy tradeoff in double-vowel identification. Error bars = ± 1 s.e.m. *p < 0.05, **p ≤ 0.01, ***p ≤ 0.001.