Reproducibility, Transparency, and Lessons from Meta-research for Music Medicine

There has been increasing interest in empirical studies to understand the replication and reproducibility of scientific research. With close to 200,000,000 published scholarly documents across science, there is tremendous potential to synthesize the available evidence and to appraise also how research practices are evolving over time. Over 100,000 meta-analyses have tried to synthesize the available evidence across diverse scientific topics – with variable success. These efforts often reveal major biases in the design, conduct, and reporting of research studies in all fields of scientific investigation. Concurrently, targeted reproducibility efforts are trying to see how often high-profile investigations can be successfully reproduced in explicitly designed, repeat experiments and studies. The success rates are consistently low, although some fields perform better than others. A new discipline, meta-research, is integrating efforts to evaluate and improve research practices, including how to do, analyze, evaluate, disseminate, and reward scientific efforts. Work done to-date has highlighted the need for improving the integrity of scientific investigations and for enhancing their credibility through better alignment of the reward system with the standards of high-quality research. The lecture will discuss in brief these evolutions and their relevance to music medicine.