|
Abstract:
|
There is increasing concern about statistical power in neuroscience research, and not the least for published fMRI experiments. Not only does low power decrease the chance of detecting a true effect, but also reduces the chance that a statistically significant result indicates a true effect. Therefore, a power analysis is a critical component of any neuroimaging study. However, due to the complexity of the neuroimaging data many different thresholding strategies exist, and a power analysis should match the desired strategy. During this presentation, I will give an overview of the available methods and tools for power analysis, but I will also show which methods are missing. Furthermore, I will show how power analysis should not only be a statistical procedure during the design of an experiment, but can also serve for study monitoring during and after an experiment. While such methods have been around in the context of clinical trials and genetics, steps have been taken to also introduce power monitoring in neuroscience. I will give an introduction of the tools that have been developed for interim analysis and retrospective power analysis for fMRI.
|
