Observed and predicted sentence recognition in babble and benefit of spatial separation were measured as a function of low-pass cutoff frequency with and without bilateral hearing aids. Predictions were determined with an importance-weighted speech-audibility metric (aided audibility index). Hearing aid benefit improved significantly as cutoff frequency increased, but only with spatial separation. Likewise, spatial benefit improved significantly as cutoff frequency increased, but only when aided. Hearing aid benefit was significantly less than predicted, whereas spatial benefit was significantly greater than predicted. Questionnaires administered to assess listeners' perspectives on hearing aid success supported an association between subjective and objective measures of speech recognition.