One of the main problems for the scientific study of consciousness is methodological. At least prima facie, the kind of knowledge we have of our own experiences is direct and not mediated by an inference process. This kind of knowledge contrasts with the kind of knowledge we have of others' experiences, which relies on the observation of their behavior and their reports.Collecting data for the scientific study of consciousness requires scientists to go beyond their own personal experiences and study others' states. This, in turn, requires that subjects report or act a certain way depending on their experiences. Although we can, at least typically, report on our own experiences, there are two important methodological worries if:1.There are experiences on whose content we cannot report.
2.There are experiences in circumstances in which we cannot report.Theories that I have called Higher-Order Cognitive, like Rosenthal's HOT theory, maintain that the mechanisms that render a state phenomenally conscious depend on the kind of cognitive access that underlies our ability to report the content of the state. These theories deny that −in beings like us, with our reporting abilities unimpaired− 1) is possible.The question at this point is how we can empirically falsify this kind of theory. In order to do so, we would need a case of an experience on which the subject cannot report; but, if this were the case, how can we know that the subject is undergoing an experience? The paper I have presented offers a possible reply to this question.The results of Lau and Passingham's experiment suggest that the neural correlate of the cognitive access that underlies our ability to report lies in the dlPFC, making this area the most plausible candidate to implement the required kind of HOTs. Rosenthal endorses this later idea. Such a commitment is an empirical one and can be empirically falsified. We shouldn't −by no means− think of this as a weakness of the theory; quite the opposite, because the connection between reportability and consciousness is, in any case, a posteriori. Now, if there are circumstances in which a subject undergoes experiences without the corresponding activity in the dlPFC, then the kind of cognitive theories under consideration would be jeopardized. Dreams seem to present such a case.The problem is that dreams are instances of 2) and someone might raise doubts on whether dreams are conscious experiences. In "Not a HOT Dream" I presented the case of lucid dreams in favor of the reality of dreams as conscious experiences, given that subjects are able to make some simple reports during these episodes. Surely, as Weisberg notes and (2010) showed that when rapid eye movements accompanied goal-oriented motor behavior during REM sleep behavior disorder (e.g. grabbing a fictive object, hand greetings, climbing a ladder) the great majority were directed towards the action of the patient (same plane and direction) and they suggest that, when present, rapid eye movements imitate the scanning of the dream scene.I find th...