This study aimed to determine the efficacy of cognitive training in a 10-week randomised controlled study involving 22 individuals presenting with mild cognitive impairment of the amnestic type (MCI-A). Participants in the experimental group (n = 11) learned face-name associations using a paradigm combining errorless (EL) learning and spaced retrieval (SR) whereas participants in the control group (n = 11) were trained using an errorful (EF) learning paradigm. Psycho-educational sessions on memory were also provided to all participants. After neuropsychological screening and baseline evaluations, the cognitive training took place in 6 sessions over a 3-week period. The post-training and follow-up evaluations, at one and four weeks respectively, were performed by research assistants blind to the participant's study group. The results showed that regardless of the training condition, all participants improved their capacity to learn face-name associations. A significant amelioration was also observed in participant satisfaction regarding their memory functioning and in the frequency with which the participants used strategies to support memory functions in daily life. The absence of difference between groups on all variables might be partly explained by the high variability of scores within the experimental group. Other studies are needed in order to verify the efficacy of EL learning and SR over EF in MCI-A.