Last September, in opening remarks during a National Academy of Sciences workshop on the arts and aging, I posed some questions about research in this field. As reported in Aging and the Arts: Building the Science (2013), an account of the workshop, jointly sponsored by the NEA and the National Institutes of Health, I asked, "Are there theoretical models that explain how participation in the arts affects the health and well-being of older Americans?" For researchers in the social sciences, assumptions of this type are often illustrated through a "theory of change" diagram or a logic model. But in biomedical research, there's an additional layer of terminology to describe how we think different variables rub off each other to produce a result. We talk about plausible "pathways" or "mechanisms of action." And every so often a study emerges that invokes this impressive argot to reflect on how art works. A couple of months ago, the peer-reviewed journal Medical Science Monitor published an article titled "Recreational Music-Making Alters Gene Expression Pathways in Patients with Coronary Heart Disease." The study, backed by Yamaha Corporation of America, shows how a specific music-making intervention brings about molecular changes that can lower the impact of stress in heart disease patients. The arts intervention, Recreational Music-Making (RMM), is described as "a unique stress amelioration strategy encompassing group music-based activities for individuals without prior musical experience. RMM focuses on personal expression and group support, rather than mastery and performance." The study design itself was creative. Researchers took 34 subjects with a history of heart disease (19 men and 15 women, all white) and, well, stressed them out. Participants were all asked to assemble a 500- to 1,000-piece jigsaw puzzle, ranked by its manufacturer as "most difficult" because of its "repetitive images, dual-sided images offset by 90◦, extra pieces, and photo-to-mosaics within each piece," the article helpfully notes. The subjects had one hour. Cash was promised to the person who assembled the most pieces in that time. And, as if that weren't enough, the subjects were reminded periodically of the time remaining. After this so-called "stress" phase of the study, participants were assigned randomly to one of two forms of relaxation: either RMM or "quiet reading" of newspapers or magazines. RMM sessions consisted of "no more than seven participants at individual electronic keyboards and the facilitator physician sitting in close proximity." Peripheral blood was taken from the subjects at baseline and then again at the stress and relaxation phases of the study. The samples were analyzed for changes in gene expression pathways relevant to modulating responses to stress and relaxation. The result? RMM participants showed changes in expression for "12 pathways governing immune function and genetic information processing." By contrast, the quiet readers showed a "significant change in expression" in only two pathways. These findings impel the authors to write that "RMM may be more clinically useful for stress amelioration and may confer greater benefit to cardiovascular patients than traditional modes of relaxation." Given the well-documented  role of psychosocial stress as a risk factor for heart disease, this research has potentially broad applications for clinical practice---although as with any study worth its salt, the authors note that "additional studies are required." The study team included Marina Vernalis, MD, medical director of the Integrative Cardiac Health Project at Walter Reed National Military Medical Center. The lead author is Barry Bittman, MD, chief innovation officer at the Meadville Medical Center in Pennsylvania and CEO of the Yamaha Music and Wellness Institute. Another key contributor was Darrell Ellsworth, PhD, senior director of the Integrative Cardiac Health Program at Windber Research Institute, also in Pennsylvania.