On Experiments for Hypothetical Reasoning

Suppose that HY P is a set of hypotheses which we currently entertain about some state of aaairs represented by a propositional sentence. In a diagnostic setting, HY P might consist of all the diagnoses of some device whose description is given by , although our analysis is not restricted to diagnosis. Our concern is with experiments { how they can be designed, and what conclusions can be drawn about the hypotheses in HY P as a result of performing experiments. Speciically, we deene the concept of an experiment and the concept of the outcome of an experiment. We characterize those experiments whose outcomes refute or connrm an hypothesis, and discriminate between competing hypotheses. These characterizations are in terms of the prime implicates of , and hence are implementable using assumption-based truth maintenance systems. Finally, we provide some results on diierential diagnosis in the cases of consistency and abductive-based diagnosis.