The effect of sleep in perceptual learning with complex objects.

Recognition of objects improves with training, but task performance also improves between sessions without further training. This offline learning seems to be influenced by post-training sleep, as is evidenced in perceptual learning studies with simple stimuli. In this study we aim to investigate the role of sleep in perceptual learning with complex natural and man-made objects. Participants were trained with a backward masking task during four sessions with 12h between each training session (morning-evening-morning-evening or evening-morning-evening-morning). A larger improvement on performance was found after a night's sleep, than when subjects performed the task without having slept between training sessions. This effect was not influenced by the participants' chronotype or non-verbal intelligence. In addition, we replicated some key characteristics of perceptual learning with complex objects. Participants were retested six days after the last training session with the previously trained stimulus and new stimuli. The performance gains were long-lasting and specific to the trained stimulus set.