Regarding verifiably finding a bottle of water, I think that there may be more consistency than you seem to think. Following a north/south route works for finding a bottle of water during the daytime, but, if the study is done at night, one can find the water bottle regardless of whether they are searching a north/south or east/west path. The study I've designed does include a double blind and, where applicable, random assignment. I've also built in controls to account for time of day, directionality, distance from the bottle of water, and false positives. Besides Educational Psychology, three of my areas of expertise are experimental design, evaluation, and statistics. I've already put myself through the entire rigor of my experimental design.One of the most consistent things about dowsing is... there is no consistency. Everyone seems to have their own peculiarities, requirements, or limitations. I'll bet there's a dowser out there who can only do it in an east/west route, but for most dowsers direction makes no difference.
If you're a psychologist then you are surely familiar with scientific test protocols, such as double-blind clinical trials used in, say, psychiatric studies. The same sort of thing can be applied to dowsing. In the past, I've designed and applied such methods to willing participants and, in every single case, their dowsing ability completely disappeared. You might want to apply some rigor to your self-evaluation before you go too far down the rabbit hole.