Identifying neuromodulation targets for pain in the human brain Neuropathic pain is prevalent across many different neurological and psychiatric disorders and has one of the largest adverse effects on quality of life. Pharmacological pain treatments often have limited efficacy and undesirable side effects, including dependence and addiction. Neuromodulation treatments such as transcranial magnetic stimulation show promise for pain while avoiding the side effects associated with pain medications. An obstacle to improving neuromodulation treatment is identifying the human brain regions responsible for pain. Traditional functional neuroimaging studies only identify correlates of pain which could be causing symptoms, compensating for symptoms, or simply serving as a marker of symptoms. This ambiguity is a problem when seeking to identify therapeutic targets. Unlike functional neuroimaging, lesion studies allow for casual links between symptoms and human neuroanatomy. The PI has developed and validated a new method to identify neuromodulation targets in the human brain based on brain lesions and a map of human brain connectivity. Called lesion network mapping, this approach has been used to map numerous neurological and psychiatric symptoms to brain networks (Fox 2018, NEJM). Brain networks identified using this approach have proven to be effective neuromodulation targets for symptoms such as tremor, Parkinson’s disease, dystonia, and depression. Our preliminary data suggests this method is equally valuable for mapping pain, including identification of therapeutic targets. Here, we will use this approach to map pain to a human brain circuit based on lesion locations previously reported to cause pain (Aim 1), a dedicated dataset focused on thalamic lesions causing or not causing pain (Aim 2), and a prospective dataset of patients longitudinally assessed for development of post-stroke pain (Aim 3). We hypothesize that; (1) lesions causing pain will map to a common brain network; (2) this network will show abnormal metabolism in patients with post stroke pain, (3) this network will predict who will develop post-stroke pain; and (4) this pain network will include primary motor cortex, our best validated neuromodulation target for pain. Completion of these aims will identify a candidate human brain network causally linked to pain. Once identified, this network can be validated in future trials with more detailed characterization of post stroke pain and used as an improved target for neuromodulation that can be tested in clinical trials.