Talk:BONUS - Advances in Methods and Practices in Psychological Science May 2018/@comment-35481611-20180504134701

Man I just wrote another 500+ words about this paper and the wiki ate them. Beware! Write your comments and posts in another program first!

Cliffnotes version: How representative do you think that .27 R in the Open Science Foundation sample is of cognitive psychology. I can think of lots of cognitive psychology that I would think of as high-R. Think Randy Engle, pursuing a high-powered design in which he must feel pretty assured that he'll find the working memory something or other that he's looking for. The good science part of that is not the materialization of a sexy new finding, but the rich description of a known effect.

To your point, you are right that a majority of the well-reasoned, well-executed science that we pursue based on intuition and the information at our disposal are not likely to take root in this publishing environment. A [exceedingly] charitable view of publisher behavior is that this is a deliberate incentivization of high-R research; of rich description of known psychological phenomena. They err in deciding what a 'known psychological phenomenon' is, CLEARLY, but they're still trying to plant a flag on the side of unexciting elaboration rather than well-reasoned, well-executed exploration. Recent shifts in publisher behavior might suggest that they want to stay this course, and just do a better job of really getting the real stuff.

To be clear, all of this is in stark opposition to what we *actually* think happens, which would suggest that the best thing that could happen to Will Whitham is to mock up some junk science about balloon psychology and take it to Nature...