To pile on: it’s in Scientific Reports. Not high impact, not highest quality. It’s a 9 day in lab study on mystery taste lab brew coffee (whose brewing was deemed so insignificant to a coffee study that its brew recipe was excluded from the methods section). These sort of setups are useful from a “help me isolate some test cases” point of view, but is obviously as much a perturbation in participants natural lives that it might even outweigh the impact of drinking coffee. Further, it draws conclusions from an acute 5 day caffeine exposure, whereas in real life these participants relationships and adaptations to caffeine are built over a long period of time. The power of the study is small and not scoped for conclusions, but rather pointing to an area of interest for further funding.