The Secret World of Sleep by Penelope A. Lewis

Author:Penelope A. Lewis
Language: eng
Format: epub
Publisher: St. Martin's Press
Published: 2013-07-17T04:00:00+00:00

Fig. 22 Japanese character task

More abstract examples of the same general principle come from tasks in which people are asked to extract general statistics from a set of information arranged according to probabilistic rules.5 For instance, if participants had been exposed to a continuous stream of auditory tones which were organized probabilistically (tone A was followed by tone B with a 90 percent probability, tone B was followed by tone C with a 90 percent probability, and so forth) for a few minutes, they were subsequently better able to differentiate short snippets of tone sequence which conformed to the same probabilistically determined pattern from randomly ordered snippets after a night of sleep. In fact, their ability to recognize structured sequences improved significantly overnight.

Taken together, these studies demonstrate that sleep is important for combining information from multiple sources. It helps us to extract statistical regularities, pull out general principles, integrate newly formed memories with older knowledge structures, and piece together a larger picture from a set of interrelated fragments. But how does sleep facilitate these processes? This problem remains to be solved, but there is at least one promising potential explanation.

In 2011, with my colleague Simon Durrant, now at the University of Lincoln, I developed a model called information overlap to abstract (iOtA), which attempts to explain how replaying memories during slow wave sleep, in combination with the downscaling of connections between neurons which also occurs in that sleep phase (otherwise known as synaptic homeostasis, see chapter 5), could explain all of these phenomena.6 The basic principle of this model is extremely simple: If more than one memory is replayed at the same time, the neurons associated with areas of shared replay, or “overlap,” will be more strongly activated than the other neurons (Fig. 23). This means, for instance, if you replay memories of two or three different birthday parties, all of which involved cake, presents, and balloons, but each of which was held in a different place and with a different set of guests, then responses in the neurons which code for cake, presents, and balloons will be stronger than responses associated with the locations of individual parties or the people who attended them. Furthermore, based on the general principle that neurons which fire together wire together (see chapter 3), the linkages between neural representations of cake, presents, and balloons will also become stronger than other linkages associated with these memories, such as between a specific birthday girl and her presents or the other people who were at her party (Fig. 24). All of this strengthening is important because it means that when the synapses are subsequently downscaled these representations of overlap may be the only thing that is retained. In fact, as multiple memories are replayed across a night, the more often a specific representation (say, a birthday cake) or pair of representations (say, both cake and presents) is triggered, the more likely that specific aspect of a memory is to be retained.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.