Sleep and perceptual learning

old_uid4152
titleSleep and perceptual learning
start_date2008/02/21
schedule14h
onlineno
summaryRecently, a considerable number of studies have suggested that sleep plays an important role in learning consolidation, although it is still controversial. In the present study, we examined consolidation related brain activity during sleep after a visual perceptual training. We measured brain activity in sleep for 90 min simultaneously using polysomnography and functional magnetic resonance imaging at 3 Tesla magnetic field. First, subjects spent 2 adaptation nights in which they slept inside the MRI scanner with electrodes attached so that they familiarized themselves with falling asleep inside the scanner. Then, we measured BOLD signals in the visual cortex before and after an intensive training of a texture discrimination task. Perceptual learning of this task has a location specific training effect and may involve the primary visual cortex (V1) (Karni & Sagi, 1991). The task stimulus was presented only in one visual field quadrant. The post-training sleep session was conducted at the same night as the training, approximately 6 hours after the training session. Relative BOLD changes during sleep compared to that in the wake period after sleep were calculated in both the trained and untrained regions in V1. Sleep stages were scored according to the standard criteria. V1 in each subject had been localized in advance in a separate fMRI session by a standard retinotopic mapping technique. In the post-training sleep period, the relative brain change in the trained region in V1 was significantly higher than in the untrained region while there was no significant difference between the untrained and trained regions in the pre-training sleep. In the re-test behavioral session, which was conducted immediately after the post-training sleep, subjects' performance was significantly higher than in the initial training. The significant correlation was found between performance and the brain activity in the trained region. The performance improvement is attributed to the sleep rather than the mere time passage between the initial training and the re-test sessions, since no performance improvement was observed in the control condition where no sleep took place in the same time interval as in the main experiment. These results indicate that sleep consolidation process for the perceptual learning occurs in a retinotopically highly localized circuit specific to the location of a trained stimulus.
responsiblesInformation non disponible