Despite the most common form of brain injury, there has been little progress in the prognosis and treatment of concussion/mild traumatic brain injury (mTBI). Current 'return-to-play' guidelines are conservative, deterring the initiation of physical and social activity until patients are asymptomatic; but the effects of post-injury exercise have not been adequately investigated. Therefore, this study examined the effects of voluntary exercise on concussion recovery. Using a translational rodent model of concussion, we examined the influence of exercise on injury-associated behaviours that comprise post-concussive syndrome (PCS) and gene expression changes (bdnf, dnmt1, Igf-1, pgc1-a, Tert) in prefrontal cortex and hippocampus. In addition, as we have previously demonstrated telomere length (TL) to be a reliable predictor of mTBI prognosis, TL was also examined. The results suggest that exercise initiated within 1-3 days post-concussion significantly improved motor and cognitive functioning, but had limited efficacy treating emotional impairments. What is more, when deprived of social interaction and exercise, a combination similar to clinical recommendations for rest until symptom resolution, animals did not recover and exhibited impairments similar to typical mTBI animals. Exercise aided in restoration of mTBI-induced modifications to gene expression in both brain regions. An inverse relationship between the exercise return interval and TL was identified, indicating greater recovery with acute exercise reinstatement. Although additional strategies may need to be employed for emotional functioning, these findings support re-evaluation of 'returnto-play' guidelines, suggesting that exercise is valuable for the treatment of concussion.