The last decade has seen a dramatic increase in our understanding of sleep-dependent memory consolidation, moving it from a generally discredited (or at best ignored) concept to a largely accepted tenet among both memory and sleep researchers, even found in undergraduate psychology textbooks. This work has established a firm connection between sleep and memory function. But given the complexity of both sleep and memory, each of which comprise multiple stages and subtypes, even the most basic characterizations of this phenomenon remain unclear. To date, most studies have sought to assign consolidation of particular classes of memory to discrete sleep stages, for example consolidation of procedural memories to REM sleep or declarative memories to slow wave sleep (SWS). But exceptions to this simple dichotomy come close to outnumbering supporting studies.Several new studies, presented here, lead us to now propose a unified sleep-dependent memory consolidation hypothesis, to describe how sleep stages contribute to memory consolidation. We propose that sleep plays a more nuanced role in memory processing than previously considered, with sleep stages being selectively involved not with specific memory types, but with specific components of the memory consolidation process. Specifically, we propose that SWS stabilizes recently acquired declarative and procedural memories, while REM and Stage 2 nonREM sleep subsequently enhance them, selectively reinforcing their most valuable components and integrating them into pre-existing networks of stored information.This new hypothesis is consistent with the growing literature of sleep-dependent consolidation across a range of memory functions, and is strongly supported by new results presented in this talk. These new results, in concert with the unified memory consolidation hypothesis, move the study of sleep and memory beyond a discussion of classical memory consolidation (i.e., stabilization), into the realm of more powerful and valuable forms of sleep-dependent memory processing that (i) enlarge the neural networks in which memories are stored, (ii) extract patterns and rules from large bodies of encoded information (iii) integrate them with other, older memories into rich semantic networks, and, perhaps as a consequence of these other processes, (iv) selectively enhance those aspects of memories of greatest value to the organism.Background: There are few data on the sleep patterns of Australian children, or the associations between sleep and sociodemographic factors [age, sex, socio-economic status (SES)], temporal context (type of day, season), type of day (school day, weekend, holiday), and individual characteristics (weight status). Methods: Between 2001 and 2007, 4,033 9-18 year old Australians reported the time of falling asleep and waking up on 9,053 individual nights. Using a computerised use-of-time recall, the adolescents also reported all activities performed on that day, choosing from 250 different activities and reporting in time slices as fine as 5 minutes.
Resul...
