To assess the feasibility of a cognitive rehabilitation program in breast cancer survivors (BCS) with persistent post-treatment cognitive complaints. BCS with cognitive complaints, 18-months to 5-years post-treatment, were recruited for a once-weekly, five-week, group cognitive training intervention. Outcome measures included self-reported mood and cognitive function, and neurocognitive tests administered at pre-intervention, immediate-, two-month and four-month post-intervention. A sub-study in eight participants evaluated resting state quantitative electroencephalography (qEEG) changes from pre- to immediate post-intervention in relationship to post-intervention changes in cognitive complaints. Twenty-seven BCS completed the protocol and tolerated the intervention well. We observed significant reductions in total and memory-specific cognitive complaints from pre-intervention to immediate post-intervention (p = 0.031 and p = 0.009, respectively) and at four-months post-intervention (p < 0.0001 and p < 0.001, respectively). Significant improvement in neurocognitive tests were found for Symbol Digit, Stroop, and Trails A tests (df = 26, all p's <0.05). Effect sizes for changes from pre-intervention to immediate and to four-month post intervention ranged from 0.429 to 0.607, and from 0.439 to 0.741, respectively. Increase in qEEG absolute alpha power over the course of the intervention was associated with reduced complaints at immediate post-intervention (r = -0.78, p = 0.021), two-months (r range = -0.76 to -0.82, p-value range 0.004 to 0.03), and four-months (r = -0.71, p = 0.048). A five-week group cognitive training intervention is feasible and well tolerated. Cognitive complaints and neurocognitive test performances showed positive changes. qEEG may serve as a potential biomarker for improvement in self-reported complaints. A randomized clinical trial is underway to test the efficacy of the intervention.