Semantic cuing and the scale insensitivity of recency and contiguity.

In recalling a set of previously experienced events, people exhibit striking effects of recency, contiguity, and similarity: Recent items tend to be recalled best and first, and items that were studied in neighboring positions or that are similar to one another in some other way tend to evoke one another during recall. Effects of recency and contiguity have most often been investigated in tasks that require people to recall random word lists. Similarity effects have most often been studied in tasks that require people to recall categorized word lists. Here we examine recency and contiguity effects in lists composed of items drawn from 3 distinct taxonomic categories and in which items from a given category are temporally separated from one another by items from other categories, all of which are tested for recall. We find evidence for long-term recency and for long-range contiguity, bolstering support for temporally sensitive models of memory and highlighting the importance of understanding the interaction between temporal and semantic information during memory search. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

How inhibitory oscillations can train neural networks and punish competitors

We present a new learning algorithm that leverages oscillations in the strength of neural inhibition to train neural networks. Raising inhibition can be used to identify weak parts of target memories, which are then strengthened. Conversely, lowering inhibition can be used to identify competitors, which are then weakened. To update weights, we apply the Contrastive Hebbian Learning equation to successive time steps of the network. The sign of the weight change equation varies as a function of the phase of the inhibitory oscillation. We show that the learning algorithm can memorize large numbers of correlated input patterns without collapsing and that it shows good generalization to test patterns that do not exactly match studied patterns.