We use a spin-1/2 model to analyze tunnelling in a double well system coupled to an external reservoir. We consider different noise sources such as fluctuations on the height and central position of the barrier and propose an experiment to observe these effects in trapped ions or atoms.
Two mechanisms of decoherence in ion traps are studied, specially related to the experiment [Kielpinski et al., Science 291 (2001) 1013]. Statistical hypothesis are made about the unknown variables and the expected behaviour of the visibility of the best experimental pattern is calculated for each mechanism. Data from the experiment are analyzed and show to be insufficient to distinguish between them. We suggest improvements which can do this with slight modifications in the present facilities.