BackgroundBrain training is currently widely used in an attempt to improve cognitive functioning. Computer-based training can be performed at home and could therefore be an effective add-on to available rehabilitation programs aimed at improving cognitive functioning. Several studies have reported cognitive improvements after computer training, but most lacked proper active and passive control conditions.ObjectiveOur aim was to investigate whether computer-based cognitive flexibility training improves executive functioning after stroke. We also conducted within-group analyses similar to those used in previous studies, to assess inferences about transfer effects when comparisons to proper control groups are missing.MethodsWe conducted a randomized controlled, double blind trial. Adults (30–80 years old) who had suffered a stroke within the last 5 years were assigned to either an intervention group (n = 38), active control group (i.e., mock training; n = 35), or waiting list control group (n = 24). The intervention and mock training consisted of 58 half-hour sessions within a 12-week period. Cognitive functioning was assessed using several paper-and-pencil and computerized neuropsychological tasks before the training, immediately after training, and 4 weeks after training completion.Results and conclusionsBoth training groups improved on training tasks, and all groups improved on several transfer tasks (three executive functioning tasks, attention, reasoning, and psychomotor speed). Improvements remained 4 weeks after training completion. However, the amount of improvement in executive and general cognitive functioning in the intervention group was similar to that of both control groups (active control and waiting list). Therefore, this improvement was likely due to training-unspecific effects. Our results stress the importance to include both active and passive control conditions in the study design and analyses. Results from studies without proper control conditions should be interpreted with care.