Several recent reports indicate that both localization and speech intelligibility in spatially separated noise are substantially improved by using cochlear implants (CIs) in both ears rather than in just one. Benefits appear to be largely derived from the effects of level variations at the two ears due to the head shadow whereas contributions from interaural time differences (ITDs) seem smaller than in normal hearing listeners. The effect of binaural unmasking estimated from speech studies to date varies from study to study and is possibly confounded by issues such as listening experience, bias or loudness effects when comparing the performance for the better ear with that using both ears. To improve the contribution from timing information at the two ears, it may be necessary to change present clinical sound-processing schemes that currently preserve only envelope cues so that they also preserve fine-timing information. However, recently published data show that basic psychophysical sensitivity to fine-timing ITDs in CI patients is very poor for rates beyond a few hundred hertz, suggesting that subjects do not actually hear ITD cues at those rates anyway. Data from a number of new studies are presented to discuss these and other issues related to the potential to benefit from bilateral implantation.