Verbal working memory delays are found in many deaf children with cochlear implants compared with normal-hearing peers, but the factors contributing to these delays are not well understood. This study investigated differences between cochlear implant users and normal-hearing peers in memory scanning speed during a challenging verbal working memory task. To better understand variability in verbal working memory capacity within each sample, associations between memory scanning speed, speech recognition, and language were also investigated. Twenty-five prelingually deaf, early implanted children (age, 8-17 yr) with cochlear implants and 25 normal-hearing peers completed the Wechsler Intelligence Scale for Children, Fifth Edition, Letter-Number Sequencing (LNS) working memory task. Timing measures were made for response latency and average pause duration between letters/numbers recalled during the task. Participants also completed measures of speech recognition, vocabulary, and language comprehension. Children with cochlear implants had longer pause durations than normal-hearing peers during three-span LNS sequences, but the groups did not differ in response latencies or in pause durations during two-span LNS sequences. In the sample of cochlear implant users, poorer speech recognition was correlated with longer pause durations during two-span sequences, whereas poorer vocabulary and weaker language comprehension were correlated with longer response latencies during two-span sequences. Response latencies and pause durations were unrelated to language in the normal-hearing sample. Children with cochlear implants have slower verbal working memory scanning speed than children with normal hearing. More robust phonological-lexical representations of language in memory may facilitate faster memory scanning speed and better working memory in cochlear implant users.