حميده نوروزي

Working memory is a fundamental component of human cognition, supporting the short term maintenance an‎d manipulation of information an‎d underlying many higher order cognitive functions. The aim of this thesis is to investigate the neural dynamics of spatial working memory an‎d its relationship to behavioral performance using electroencephalography (EEG) signals an‎d eye movement measures. The data were analyzed across three complementary studies. In the first study, time–frequency patterns an‎d functional brain connectivity were examined throughout the task. Time–frequency analyses revealed a significant increase in alpha ban‎d power over frontal an‎d central regions during the memory interval. Functional connectivity, assessed using the phase locking value (PLV), showed that alpha ban‎d connectivity between frontal an‎d posterior regions significantly increased during the memory period, potentially reflecting interactions between these regions in maintaining spatial information. Moreover, increased fronto posterior connectivity during the memory interval was associated with reduced behavioral error. In the second study, the spatiotemporal dynamics of the brain were analyzed using classical EEG microstate methods. In addition to the conventional microstates (A, B, C, an‎d D), polarity specific microstates were examined separately an approach seldom considered in prior research. The results indicated that reduced coverage of microstate C during the memory interval corresponded to suppression of the default mode network, whereas increased duration of microstate D was linked to activation of frontoparietal control networks. Furthermore, transition features of D+ an‎d D− microstates showed significant relationships with behavioral accuracy an‎d were able to predict saccadic error. In the third study, a novel framework for microstate extraction was introduced in which microstates were defined based on functional connectivity patterns derived from normalized PLV matrices. Using clustering algorithms, four distinct network level states were identified, an‎d two of these states were found to be more dominant an‎d more stable during the memory interval. Differences in microstate parameters between high an‎d low accuracy trials further indicated that working memory performance depends on the temporal stability an‎d network organization of the brain. Overall, the findings of this thesis demonstrate that spatial working memory arises from dynamic, state dependent interactions within large scale brain networks. The proposed connectivity based microstate framework provides a compact representation of network level dynamics an‎d bridges traditional microstate analysis with connectivity measures offering a novel approach for investigating the temporal organization of brain networks an‎d their relationship to cognitive performance.

حافظه كاري , اكتروانسفالوگرام , باند آلفا , ارتباطات كاركردي مغز , ريزحالت هاي مغزي , معيار قفل شدگي فاز

working memory , EEG , Alpha Band , Functional Brain Connectivity , Microstate , PLV

Hamideh Norouzi

Mohammad Reza Daliri