The changing perceptual skills of paediatric cochlear implant (CI) users provide a unique opportunity to examine the process and development of speech perception in an altered auditory system. Habilitation specialists disagree as to whether children using a CI should develop their speech perception skills under auditory-alone (AA) or audiovisual (AV) conditions. Audiovisual information aids auditory perception by reducing some of the ambiguity in the speech signal, and in doing so reduces the processing load on working memory. Children using a CI continually process a degraded auditory signal. The influence of visual speech information to auditory perceptions and working memory was investigated in children using a CI and children with normal hearing. Across two test periods, comparisons were made between the children's perception of synthetic speech stimuli under AA and AV listening conditions. The speech stimuli varied along a voiced plosive continuum, /badaga/, and were presented AA or paired with visual /ba/ and /ga/ recordings in congruent and incongruent AV test conditions. Under each condition, children were asked to indicate what sound they heard by selecting a response button. Results were compared to performance on nonverbal measures of working memory. The first test period included 16 children using a Nucleus cochlear implant with an average 4.75 years device experience. Children with normal hearing were matched to the CI group on age, gender and Performance IQ. Children using a CI were less categorical in their perception of the AA stimuli and produced more variable responses. All children perceived the auditory stimuli more consistently when paired with visual speech cues. Results indicated that children using a CI benefited more from the visual speech information than did the comparison group. Children with better working memory skills also generally perceived the exemplar auditory stimuli at higher response rates, were more likely to select the response that matched congruent AV information and were less influenced by incongruent visual speech cues. Preliminary results are now available for a follow-up assessment conducted 24 months later, including 9 children from the CI group and 4 children from the comparison group. Few AA or AV responses differed between the first and second assessment, although at the second assessment the children's AA responses were more categorical for some stimuli. With further auditory experience, children using a CI appeared to respond more like their hearing-peers to AV presentations. The implications of these findings to habilitation practices will be discussed.