Many scientists believe that science is experiencing a 'reproducibility crisis'.
The causes for this are many,I'm sure, but one apparent factor is the lack of standardized protocols for many key methods. Moreover, even when methods are reported, the details are sparse, and seemingly unimportant variables are unaccounted for.
In pain research, and neuroscience broadly, behavioral assays make up a huge component. We rely on them to assess the effects of our interventions and manipulations. Given their prominence, it'd be reasonable to have a set of standardized protocols that we as pain researchers established and followed. But in fact, we don't have anything like that. Instead, each lab, and even each person in the lab, uses their own variations of behavior protocols. Even with simple assays, like the Von Frey assay, or Hargreaves, there are many variables at play, and often not just the ones we focus on. The ones we don't discuss frequently, like time of day, location of testing, sex of the tester, etc., might make a big difference, and we often make decisions on the fly about these other variables.
So, I'm wondering if it would not be a good idea to establish a set of standardized protocols, as a community, and adhere to them. I know this wouldn't be a magic solution, but it could help combat the inability to reproduce other people's results. There is precedence for this in other fields, so I think we could do it to.
What do you think? Would this be worthwhile? And do you think it would even help?
@jmogil - What do you think?