7. PROJECT SUMMARY Motor imagery, especially when used as an adjuvant treatment with physical practice, promises to be a powerful tool for improving function in individuals with movement disorders. Yet, due to its very nature, motor imagery cannot be directly observed. This makes it difficult to assist and evaluate a patient's motor imagery efforts. Brain activity associated with motor imagery is, however, observable through neuroimaging. Moreover, with the recent development of technologies like real-time functional magnetic resonance imaging neurofeedback (rtfMRI-NF), motor imagery “behavior” can be displayed to both the patient and the clinician. We hypothesize that if patients could learn to “exercise” their own motor brain networks directly, they could optimize their rehabilitation. In this proposal, we seek to examine the feasibility of applying rtfMRI-NF imagery training to individuals with cerebellar ataxia (CA), a movement disorder that results from progressive cerebellar degeneration. Current treatments can slow the rate of motor loss through methods such as physical therapy and core strengthening, but they focus on physical manifestations and do not target the underlying neural mechanisms involved, thereby missing the root cause. In addition to evaluating the feasibility of motor imagery rtfMRI-NF in CA, we will examine the utility of additional at-home therapy, subsequent to the rtfMRI session. Finally, we will use the rtfMRI-NF data for offline analyses for brain mapping, machine learning, and simulating additional rtfMRI approaches to develop future iterations of rtfMRI-NF protocols. Thus, future work aims to establish refined experimental medicine frameworks by identifying neural underpinnings (NF targets) of motor accuracy, and testing whether engaging these targets, through NF, improves motor performance. As outlined in the proposal, Aim 1 will use rtfMRI-NF during motor imagery to train CA participants to improve motor accuracy. Thirty CA participants will receive NF during motor imagery in an experiment in which we hypothesize that 1) CA participants will be able to control a NF interface; 2) imagery skill will be positively correlated to improvements in overt tapping accuracy; and 3) overt tapping accuracy will correlate with neurological signs, whereas motor imagery skill will correlate with assessed motor imagery ability. Aim 2 will translate rtfMRI-NF learning into at- home therapy strategies for three weeks of continued training in which we hypothesize that 1) continued practice with imagery strategies will lead to additional improvements in motor timing and precision, and 2) performance during rtfMRI-NF training will positively correlate with at-home motor imagery performance. In an exploratory Aim 3, we will examine three primary questions to establish future experimental medicine designs. Specifically, these question are 1) Are there group-level differences in fMRI activity in CA versus healthy controls?; 2) Are healthy...