AbstractBackgroundYoga is a low risk intervention that is accessible, adaptable for older adults, easily incorporated into daily life, and potentially sustainable. It combines mental and physical practice and includes instruction on breathing, stress reduction, and mindfulness meditation. Previous research documents that yoga can target modifiable risk factors for Mild Cognitive Impairment (MCI) progression, including stress, depression, cardiovascular health, insulin resistance, and inflammation. We describe a pilot randomized trial of yoga for individuals with MCI.MethodsParticipants were 37 individuals with amnestic MCI who were randomly assigned to receive 12 weeks of twice‐weekly yoga intervention (YI; n = 18) or healthy living education classes (HLE; n = 19). The yoga classes were 60‐min, designed for safety (using props and chairs), and followed a fixed sequence of breathing practices, meditation, warmups, standing and seated postures, simple inversions, and relaxation. HLE classes involved interactive discussion and presentations addressing topics relevant to aging and cognitive impairment. Participants completed neuropsychological and mood measures at baseline and immediately post‐intervention.ResultsThere were no group differences for age, education, race, sex, class attendance rates, or percent taking dementia medications. Results of ANCOVA, controlling for baseline scores, did not show any statistically significant differences in neuropsychological domains between the groups. However, medium effect sizes in favor of the YI were seen for visuospatial skills (Cohen’s d = 0.47) and executive functioning/working memory (Cohen’s d = 0.58). Minimal to small effects were found for attention/speed, language, and learning and memory. The yoga group showed significant declines in perceived stress compared to HLE (p<0.05; Cohen’s d = 0.74). Surprisingly, HLE resulted in significantly greater reductions in depressive symptoms after the intervention compared to the YI (p<0.05; Cohen’s d = 0.78).ConclusionsAlthough small sample sizes likely impacted our ability to detect statistically significant results, these findings indicate that the YI was feasible and may improve aspects of neuropsychological functioning in individuals with MCI. Results support stress reduction as a possible mechanism for the YI. The interactive nature of HLE class may explain its effect on mood. Future studies should address a yoga intervention in a larger sample.