Children treated with radiation therapy for brain tumors consistently demonstrate a decline in intellectual functioning that is associated with academic failure and reduced quality of life. Medulloblastoma is the most common malignant brain tumor in children and survival rates approach 80% with current multimodal therapy. With a growing survivor population, efforts to improve long-term cognitive outcomes become imperative. A primary goal of this research program is development of interventions that mitigate the impact of CNS-directed therapies on the quality of life of childhood cancer survivors. While efficacious, pharmacological and therapist- delivered interventions are limited by medical contraindications and availability of local providers. Recent research suggests impairments in attention and working memory (WM; online maintenance and manipulation of information) may be proximal contributors to intellectual declines among brain tumor survivors. A computer- based WM intervention has demonstrated efficacy with individuals with developmental and acquired attention disorders. Improvements have been achieved on measures of attention, WM and executive functions with benefits persisting months after training. Functional neuroimaging (fMRI) conducted before and after WM training suggests neuroplasticity for supporting brain systems. This intervention has recently been shown to be feasible and efficacious for childhood cancer survivors with evidence for associated brain-based changes. It is now crucial to determine whether we can preserve cognitive function by intervening prophylactically, before the emergence of late effects, and combine interventions to optimize outcomes. New research with children with chronic disease indicates aerobic exercise offers cognitive benefits. Further, animal models reveal exercise- related recovery following injury and suggest underlying brain mechanisms (e.g., new nerve cell formation). In the proposed study, we will recruit children treated for newly diagnosed medulloblastoma on a front-line protocol that includes a randomized exercise trial during radiation therapy. Immediately following treatment, we will randomly assign patients to a computerized WM intervention or a time and attention control group (generic social support). Both groups will complete pre- and post-intervention cognitive testing and fMRI. Study aims include: testing the efficacy of a prophylactically administered computerized WM intervention (Specific Aim 1), comparing the cognitive impact of a computerized WM intervention administered in conjunction with an aerobic exercise intervention (Specific Aim 2), and using fMRI to examine neuroplasticity associated with WM training (Specific Aim 3). It is hypothesized that: 1. participants randomized to the WM intervention will demonstrate less WM decline relative to controls, with benefits that are generalizable to higher-order cognitive skills and sustainable over time, 2. participants receiving both aerobic exercise and WM interventions will demonstrate better cognitive outcomes than participants receiving either intervention alone, and 3. participants randomized to the WM intervention will demonstrate meaningful changes in brain activation patterns after WM training.