Reproducible research for neuroimaging

The increasing storage capacity and processing power of modern computing resources is dramatically changing the nature of scientific scholarship.  As a result, traditional peer-reviewed research articles are no longer sufficient for communicating most scientific work.  In the late 1980s, Jon Claerbout, a geophysicist at Stanford, coined the phrase "reproducible research" to refer to the complete software environment necessary to generate all the results and figures included in published articles. Over the last several years, a growing number of scientists have begun calling for wide-spread adoption of Claerbout's ideas about reproducible research.

In this poster, I will present the basic ideas of reproducible research, discuss the social and technical obstacles to achieving it, and describe some of the existing attempts to implement it.