Cognitive control during selective long-term-memory retrieval
In every waking moment of our life, we are focusing our conscious processing of information on certain perceptions, ideas, feelings, and memories. In other words, we are filtering information, or selectively attending to specific occurrences in our environment (external perceptions) and to thoughts...
|Main Author:||Kizilirmak, Jasmin|
|Contributors:||Rösler, Frank (Prof. Dr.) (Thesis advisor)|
PDF Full Text
No Tags, Be the first to tag this record!
In every waking moment of our life, we are focusing our conscious processing of information on certain perceptions, ideas, feelings, and memories. In other words, we are filtering information, or selectively attending to specific occurrences in our environment (external perceptions) and to thoughts (internal perceptions) all the time. For example, imagine you want to bake a marble cake: How much flour, butter, and sugar do you need? How do you accomplish the task of remembering the amount of the flour without confusing it with the amount of sugar? The present thesis addresses this problem of selective long-term-memory (LTM) retrieval, especially in a situation where you have to continually retrieve several internal representations (in the example above: the amount of flour, butter, and sugar for a certain cake) associated with the same cue (marble cake, Figure 1A). The aim is to shed light on the question which cognitive processes enable such a focused, selective processing of memory representations considering the vast amount of possibly interfering information. In a series of three studies, the present thesis investigated how selective LTM retrieval is accomplished when relevant and irrelevant associations with a retrieval cue can switch their roles dynamically. We manipulated changes of to-be-retrieved associations and relevant associative networks from trial to trial, and measured neuronal correlates of the involved processes by means of early and late occurring event-related potentials (ERPs), and hemodynamic activity as measured by functional magnetic resonance imaging (fMRI). We found reliable processing differences between trials in regard to (1) whether associations of the same network (i.e., related to the same cue) or different networks had to be accessed in consecutive trials, (2) whether consecutively to be retrieved associations were of the same type, and (3) whether all or one other association of the same network had been accessed in the previous trial. These processing differences became manifest in amplitudinal and topographical ERP differences, as well as in different activation patterns of cortical regions as indicated by fMRI, and a reliable effect of positive priming for the target association in the behavioral data. Our findings provide evidence that selective retrieval with dynamic shifts between to-be-retrieved and to-be-ignored associations with a retrieval cue is predominately accomplished by enhancing the relevant information through focusing internal attention on the relevant, instead of inhibiting potential retrieval competitors. The involved brain areas further bear a high correspondence to a cluster of frontal and parietal cortical regions that have been proposed to form an attentional-control network. Our data therefore support the recent claim that similar brain regions are important for the selection of memory (i.e., internal) representations as they are for the selection of perceptual (i.e., external) representations, and encourage further investigation in regard to the degree and nature of this overlap.