[en] Verbal working memory (WM) is strongly influenced by semantic similarity between items. Serial recall of lists of semantically similar words is better than of dissimilar words. No model currently exists that explains this support from semantic knowledge. We present a connectionist WM architecture integrating meaning. The core architecture encodes distributed representations of items and contexts through temporary binding. Meaning is not directly encoded into WM. Instead, semantic knowledge supports WM through sustained activation in a long-term memory semantic network. This way of integrating meaning into WM accounts for the impact semantic similarity on WM performance in different tasks, including: (1) the beneficial effect of semantic similarity on item memory, (2) the absence of impact of semantic similarity on memory for order and (3) the protective effect of semantic similarity against interference. Based on these results, we propose that semantic knowledge supports WM through activated long-term memory.
