Maybe this is a stupid suggestion, but does all this mean that if users want the most accurate results they should keep their PicoMuon detectors in a fridge? And, if so, what would be the optimal temperature for the fridge to be set to?

Not a stupid suggestion at all. Temperature really does matter for detectors like the PicoMuon detector, especially because the silicon photomultipliers (SiPMs) inside produce thermal noise (“dark counts”) that rises…