Art - ffytche (F)

Few imaging studies have investigated the conscious ‘pictures’ of the external environment that we associate with seeing (visual percepts). The problem that confronts the neuroscientist is recognizing the neural correlate of ‘seeing’ and differentiating it from afferent sensory activity, which is assumed to remain unconscious1. One solution is to study a visual system in which percepts have become dissociated from sensory input. Such dissociation can follow a sudden deterioration in visual abilities in patients who in other respects are neuropsychiatrically normal2–4. This syndrome is termed the Charles Bonnet syndrome (named after the Swiss philospher who first described it)5. The spontaneous visual percepts (visual hallucinations) experienced by these patients are identical to those associated with normal seeing, although they can be recognized because of their bizarre and often amusing character and because, given the patients’ impaired vision, they are seen in greater detail than real stimuli6. They differ from visual imagery experiences in that the hallucinations are localized to external space (rather than inside the head), have the vivid qualities of normal seeing and are not under voluntary control. We investigated the neural substrate of visual consciousness in a group of such patients, using two different but complimentary strategies, both of which have proven successful previously7–9. The first strategy (Experiment 1) was to ask the patients to signal the onset and offset of each hallucination during a fiveminute scan and to then correlate the timing of the hallucinations with the time-course of the fMRI signal. A second, indirect strategy, which did not depend on capturing a hallucination during a scan, identified functionally abnormal brain regions by scanning the patients while they viewed a nonspecific visual stimulus and comparing the results to those of a matched control group who had never experienced hallucinations (Experiment 2).