Download What Is Consolidated During Sleep-Dependent Motor Skill Learning_ 2005-3916.pdf

Contents : Commentary/ Walker: A refined model of sleep and the time course of memory formation tion (Gaskell & Marslen-Wilson 1997 Luce & Pisoni 1998 McClelland & Elman 1986 Norris 1994). Hence an acid test of whether a spoken form has been lexicalized is whether or not it engages in lexical competition and thereby affects the activity within the mental lexicon. In one of our experiments (Gaskell & Dumay 2003 Experiment 3) adults learned nonsense-speech sequences that overlapped strongly with existing words (such as cathedruke for cathedral ). The influence of the newly learned words on lexical activity our indirect measure of learning was then assessed using the pause detection paradigm. Here participants made speeded decisions as to whether a short silence was present towards the offset of the existing words (e.g. cathedr al ). Mattys and Clark (2002) demonstrated that pause detection latencies are positively correlated with the number of words activated in lexical memory on hearing the speech portion preceding the pause. As indicated by a direct two-alternative force choice recognition test (e.g. cathedruke vs. cathedruce ) the 36 exposures to each novel word during learning resulted in a good immediate explicit knowledge with no significant change when retested one week later (96% of correct responses on both occasions). More crucially whereas no change in lexical activity was observed in pause detection immediately after learning a new competitor a clear effect of the novel competitor had emerged during the time interval between exposure and retest seven days later. This is strong evidence that in contrast to phonological (episodic) storage lexicalization (and thus integration) of spoken words requires a substantial amount of time. In a follow-up experiment (Dumay et al. 2004 Experiment 2) we examined more closely the timecourse of lexicalization tracking the effect of exposure on lexical activity at three time points: immediately after exposure 24 hours later and a week later. Again there was no evidence of immediate lexicalization but 24 hours after exposure as well as a week later pause detection performance on the existing words demonstrated that the new competitor was now contributing significantly to lexical activity. Concurrently the performance in explicit recognition and free recall improved across sessions (from 82% to 87% and from 8% to 20% respectively). From these results we can therefore narrow down the critical time period for the lexicalization of a spoken word form to somewhere between one and 24 hours after exposure. Whether the integration of new representations into long-term lexical memory is primarily dependent on sleep (or some sleep-specific brain state or states) is still to be determined. However our findings are clear evidence that both consolidation-based enhancement and integration of new declarative memory representations can be obtained after a posttraining interval that includes sleep. Rather than being a distinct stage in the process of memory formation enhancement might be the sign that integration has taken place. In fact it would seem quite uneconomical to engage into some sleepdependent additional learning if it were not to integrate the corresponding representations in a long-term associative network or repertoire. Walker (sect. 2.2) speculates that the effect of sleep on declarative memory could be more protracted and one of subtle maintenance in order to prevent decay over time. Our data indicate that this may not be the case. They suggest instead that following sleep newly acquired declarative memory representations are not only enhanced that is more easily accessed or specified but also able to affect a highly automatized perceptual skill and therefore its underlying procedural memory system. ACKNOWLEDGMENTS This research was supported by a grant from the UK Medical Research Council (G0000071) to Gareth Gaskell. We thank Graham Hitch for useful comments. 70 BEHAVIORAL AND BRAIN SCIENCES (2005) 28:1 What is consolidated during sleep-dependent motor skill learning Luca A. Finellia and Terrence J. Sejnowskia b a Howard Hughes Medical Laboratory The Salk Institute for Biological Studies La Jolla CA 92037 bDivision of Biological Sciences University of California San Diego La Jolla CA 92093. lfinelli@salk.edu terry@salk.edu http://www.cnl.salk.edu Abstract: Learning procedural skills involves improvement in speed and accuracy. Walker proposes two stages of memory consolidation: enhancement which requires sleep and stabilization which does not require sleep. Speed improvement for a motor learning task but not accuracy occurs after sleep-dependent enhancement. We discuss this finding in the context of computational models and underlying sleep mechanisms. Procedural learning particularly the investigation of motor skill learning has attracted renewed attention in memory research over the past few

INFO HASH : 1cc61c5925ca0de4062eb3bba321f06086c3518d

Download now

File Size: 79.4 kb