Participants will be given a stationary exercise bike for home use. They will be instructed to use the exercise bike five times a week for thirty-minute sessions. The exercise intensity prescription will be based on the subject's VO2max determined on pre-test day. The exercise program will start at 60% of intensity per session, and then will be increased by steps of 5% intensity every 2 sessions until participants reach 30 minutes of training at 80% intensity. Participants will be contacted weekly by e-mail or phone to answer any questions about the exercise protocol and will be instructed to log each training session. Subjects will record duration of exercise, perceived exertion, average heart rate, maximum heart rate, and distance.

A physical therapist will tailor a home balance training program for each participant based on pre-training capabilities. Subjects will be asked to perform exercises five times a week for thirty-minute sessions. Both dynamic and static exercises will be performed in sitting and standing positions. Exercises will start with stabilizing in a challenging static position and progress to dynamic arm and leg movements in the same or modified position. Participants will be contacted weekly by e-mail or phone to answer any questions about the exercise protocol and will be required to log their exercise effort in terms of frequency and level of balance challenge. Individuals will be instructed to perform more difficult exercises if balance challenge scores are low.

Aerobic training on stationary bicycle for 30 minutes a day, 5 days a week for 6 months

Ataxia severity will be measured using the Scale for the Assessment and Rating of Ataxia (SARA).17 SARA evaluates the degree of ataxia by measuring gait, stance, sitting balance, speech, finger-chase test, nose-finger test, fast alternating movements, and heel-shin test.

For the walking assessment, participants will walk as fast as possible on a 10-meter runway six times, and we will average the times of trials 3-6. We will also collect marker position data from infrared emitting diodes placed bilaterally at the first and fifth metatarsal heads, heels, medial and lateral malleoli, medial and lateral condyles of the knee, head of the fibula, and anterior and posterior superior iliac crests pelvic crests using a three-dimensional Vicon motion capture system (Vicon, Denver, CO). Custom Nexus and Bodybuilder software will be used to calculate joint position and determine the following walking parameters: stride length, stride length variability, percent time in double limb support, and degree of pelvic rotation and tilt.