Computerised cognitive training (CCT) approaches to memory rehabilitation represent an attractive alternative to traditional approaches; however, there is limited empirical evidence to support their use. An AB with follow up single case design was repeated across five participants to explore the feasibility and effectiveness of CCT on subjective memory in patients with stroke. Target behaviour was subjective everyday and prospective memory failures which were assessed weekly. Following baseline (three weeks), participants completed six weeks of LumosityTM training in their homes. Data were analysed visually and statistically. The frequency of prospective memory failures decreased during intervention for one participant, while the frequency of prospective and everyday memory failures decreased significantly during the follow up period for another participant. Yet, significantly more everyday and prospective memory failures were reported following training by one study participant. No significant change in subjective memory ratings was found for remaining participants. Regarding secondary outcomes, meaningful changes on objective measures of memory were not observed, despite considerable inter-individual variability. Three participants reported improvement in individualised memory goals, while two participants described a decline. Overall, LumosityTM training appears feasible; however, no consistent evidence to support effectiveness of this CCT on subjective or objective memory was found.