Long-term working memory: leveraging prior knowledge to bypass working memory limitations

In our everyday interactions, we are exposed to large streams of information from which we construct new thoughts, form sentences and plan future actions. The system responsible for those processes is called working memory (WM) and is known to be severely limited in its capacity. In order to overcome this constraint on our ability to carry out complex cognitive tasks, we could capitalize on our (potentially) unlimited prior knowledge from LTM. The present proposal is aimed to understand how specifically contributions from episodic LTM modulate the storage, maintenance, and retrieval of information in WM and thereby bypass its limitations.  

Despite the growing number of investigations on how semantic properties of memoranda stored in LTM affect WM, the role of episodic LTM – memory for personal experience – has yet to be systematically investigated. The present project will fill this gap by providing a systematic empirical and theoretical examination of the effect of experimentally controlled prior knowledge to WM. We will examine under which conditions WM is benefiting from leveraging of episodic LTM, and whether it can shield itself from interference and therefore harmful influences that are connected to LTM representations.

Based on the present literature on the interface between LTM and WM we have isolated three main research questions on the effect prior knowledge from episodic LTM could have when used in WM: (1) Under which conditions does LTM contribute to performance in a WM task? (2) How does prior knowledge contribute to WM? And (3) What characterizes the LTM representations that contribute to WM performance?