Computational modeling of the neural mechanisms underlying working memory
2005 (English)Licentiate thesis, comprehensive summary (Other academic)
The performance on various cognitive tasks, from language to selective attention and guiding future actions depends on working memory (WM), the capacity to hold and manipulate limited items of information. However, neural basis of WM and the capacity limitation are still unclear. The present work includes behavioral, functional magnetic resonance imaging (fMRI) and computational modeling studies of the visuospatial WM in order to identify the neural correlates of WM capacity. In the first study we used behavioral distracting stimuli in order to identify cellular mechanisms that accounts for the observed behavioral decrease in mnemonic accuracy as a function of distractor distance. The study provided theoretical support that independently of the cellular and synaptic properties, increased neuronal firing rates accounted for higher mnemonic accuracy and resistance against distractors. In the second study we performed fMRI experiments on adults and children to monitor brain activity during a WM task. We isolated the delay-related activity and analyzed group differences and the distractor influence both behaviorally and in terms of changed brain activity. The fMRJ study showed higher brain activity in inferior frontal and intraparietal cortex in adults compared to children during the delay periods of WM tasks. Furthermore, adults were more accurate and less distractible than children. In a subsequent study we addressed the cellular changes during WM development. The study combined a computational analysis with fMRl in order to establish putative maturational processes governing developmental changes in brain activity. We found that the increase in activit' together with higher resistance against distractors could be explained by computational models having stronger connectivity between network areas. Our studies suggest that increased firing rates of the cortical areas involved in the maintenance of visuospatial information accounts for the developmental related increase of activity in areas associated with WM processes as well as for the higher resistance against distractors. Therefore, increasing the neural activity of the WM circuitry using either psychophysiological training protocols or pharmacological manipulation may have a beneficial effect on the WM capacity and distractibility.
Place, publisher, year, edition, pages
Linköping: Linköpings universitet , 2005. , 34 p.
Linköping Studies in Science and Technology. Thesis, ISSN 0280-7971 ; 1196
IdentifiersURN: urn:nbn:se:liu:diva-29963Local ID: 15390ISBN: 91-85457-29-9OAI: oai:DiVA.org:liu-29963DiVA: diva2:250782
List of papers