Can Sleep Enhance both Implicit and Explicit Processes?

Can Sleep Enhance both Implicit and Explicit Processes? Andrea C. Smyth (andrea.smyth@gmail.com) University of Winnipeg, 515 Portage Avenue Winnipeg, MB R3B 2E9 Canada Shuichiro Taya (s.taya@qmul.ac.uk) Experimental Biology and Psychology Centre, Queen Mary University London London, E14NS, UK Chris Hope (c.hope@surrey.ac.uk) Dept of Psychology, University of Surrey, Guildford, GU2 7XH, UK Magda Osman (m.osman@qmul.ac.uk) Experimental Biology and Psychology Centre, Queen Mary University London London, E14NS, UK Abstract This experiment examined the effects of sleep on learning, while employing an experimental design that minimizes time of day and fatigue effects. Using a modified two-phase contextual cuing task, we show that sleep benefits consolidation and offline learning minimally, and hindered subsequent conscious awareness on an explicit memory test. These differential effects of sleep on implicit learning and explicit memory can be taken as evidence that these types of information are processed differently and based on entirely distinct memory stores. Keywords: Contextual cuing; offline learning; sleep Introduction Although there is a lack of consensus concerning the exact function of sleep, recent empirical evidence substantiates claims that a good night’s sleep is more than just a biological necessity. Playing an important role in homeostatic restoration, thermoregulation, tissue repair, immune control, and memory processing (Walker, 2008), sleep may just be Mother Nature’s version of a miracle drug. A key issue of interest is whether sleep can also lead to offline learning – that is, when sleep enhances learning such that performance following a nights sleep is comparably better than without a period of preceding sleep. Studies using associative learning tasks have demonstrated that indeed, sleep after learning shows offline consolidation of knowledge acquired during training (Walker & Stickgold, 2004). Furthermore, it is speculated that consolidation benefits are mediated by overnight neural reorganization of memory resulting in more efficient storage of information, affording improved next-day recall (Gais, Molle, Helms, & Born, 2002). Sleep before learning also appears to be critical for brain functioning. Specifically, one night of sleep deprivation markedly impairs hippocampal function, imposing a deficit in the ability to commit new experiences to memory. Despite the apparent benefits of sleep on both implicit and explicit memory, recent evidence has suggested that many of the demonstrations of offline learning in the above studies are an artifact of the type of averaging methods used to reveal sleep effects, or biased by time-of-day testing (Keisler Ashe, & Willingham, 2007), and can often be artificially enhanced as a result of the gradual build up of amassed fatigue effects through repeated or concentrated training periods (Rickard, Cai, Rieth, Jones, & Ard, 2008). Rickard et al’s (2008) demonstration of these factors involved training participants using a typical motor task in which people typed out a sequence of 5 button presses (with a reliably repeating sequence) across 12 training blocks and 2 test blocks. This research has serious implications, particularly because the criticisms apply to techniques commonly employed by many sleep studies (e.g., Gais et al, 2002; Robertson, Pasual-Leone, & Press, 2004; Wagner et al, 2004; Walker & Stickgold, 2004). One concern with Rickard et al’s (2008) study is that their criticisms are based on evidence from a motor learning task, in which fatigue effects are more likely to be generated, and so may not generalize to visual search tasks, or tasks involving explicit memory. Therefore the current study is concerned with examining the issues raised by Richard et al (2008), but using a task designed to examine both implicit and explicit processing in learning: the spatial contextual cueing paradigm (Chun & Jiang, 1998). Contextual cuing refers to improved visual search performance with repeated exposure to a configuration of stimuli. Participants are shown displays containing a set of 12 letter stimuli and are required to detect a target stimulus (a letter T) within the subset of distracter stimuli (11 letter L’s). Crucially, the location of the target in half of the displays appears repeatedly with the same arrangement of the distracters surrounding it. This learning is expressed through the gradual development of search efficiency for these repeated displays, indicating that repetitive exposure to these distracter configurations results in the acquisition of a mental representation that becomes relied upon to guide search. The benefits of employing the contextual cuing paradigm in the study are that massed practice involves visual search instead of motor processing and employs within-subjects comparisons between learned and random trials, and so the