How can we know if patients in coma, vegetative state or minimally conscious state are conscious?

This paper examines the claim that patients in coma, vegetative state and minimally conscious state may in fact be conscious. The topic is of great importance for a number of reasons--not least ethical. As soon as we know a given creature has any experiences at all, our ethical attitude towards it changes completely. A number of recent experiments looking for signs of intact or partially intact cognitive processing in the various stages of decreased level of consciousness are reviewed. Whether or not vegetative or coma patients are in fact conscious is an empirical issue that we yet do not know how to resolve. However, the simple fact that this is an unresolved empirical issue implies that the standard behavioural assessment is not sufficient to decide what it is like to be these patients. In other words, different and more sophisticated methods are necessary. From a theoretical position, the paper moves on to discuss differences in validity between reports (e.g. verbal) and signals (e.g. brain activations) in the study of consciousness, and whether results from experiments on the contents of consciousness may be of any use in the study of levels of consciousness. Finally, an integrated approach is suggested, which does not separate research in level and content as clearly as in current practice, and which may show a path to improved paradigms to determine whether patients in coma or vegetative state are conscious.