葉怡玉Yeh, Yei-Yu臺灣大學:心理學研究所郭柏呈Kuo, Bo-ChengBo-ChengKuo2010-05-042018-06-282010-05-042018-06-282009U0001-1506200910101900http://ntur.lib.ntu.edu.tw//handle/246246/178724透過注意力的調控，我們可以將訊息暫時地維持在工作記憶表徵。然而，我們對於注意力與工作記憶互動的大腦神經機制仍缺乏清楚的認識。本論文利用行為測量方法、腦電波測量，以及功能性磁振造影，探討注意力調控在工作記憶中運作的神經表徵。第一章介紹本論文的研究背景與研究假設；第二章我介紹所使用的神經影像技術與資料分析原理。第三章利用事件相關電位檢驗當個體從工作記憶表徵中選擇目標項目時，是否涉及具有空間與視覺特定性的腦電波活化，且此一活化與個體從知覺表徵中選擇目標項目時所引起的活化相同。第四章與第五章利用事件相關功能性磁振造影探討注意力調控的有效性是否影響工作記憶表徵的維持與提取，研究結果展現了有效的注意力調控會引發視覺皮質強烈的神經活化以維持記憶表徵，且注意力調控的有效性與記憶提取的困難度在大腦前額葉與後頂葉皮質的活化上也會產生顯著的交互作用，這些與注意力調控有關的大腦區域反映了由前而後的階層性關係。第六章透過功能性聯結的分析方法，展現注意力調控在工作記憶中運作的神經網路，以及大腦區域間的交互作用關係。最後，第七章針對我的研究結果進行綜合性的討論，對注意力在工作記憶中運作提出一個整合性的認知神經看法。Attention and working memory are intrinsically bound, but the neural mechanisms of their interaction are still less understood. The goal of this thesis is to establish a framework that provides neural representations for understanding the operation of attentional control in working memory with the evidence integrated from behavioural measures, event-related potentials (ERPs), and functional magnetic resonance imaging (fMRI). Chapter 3 demonstrates the two ERPs experiments that examine whether selecting relevant target items from within working memory representations involves spatially specific, retinotopic biasing of neural activity in a manner analogous to that which occurs during visual search for target items in perceptual domain. Chapter 4 describes an event-related fMRI experiment that investigates the neural correlates of the effectiveness of orienting attention during retention and the mnemonic evaluation whilst comparing working memory representations with current perceptual stimuli. Chapter 5 shows an event-related fMRI study that demonstrates that the neural activity in the posterior areas is modulated by reflectively transient attention-based operation during working memory retention. Finally, I identify the neural network of the brain regions associated with the top-down attentional operation in working memory representations using coherence analysis in Chapter 6. In conclusion, I investigate the interaction within a distributed neural network for supporting attentional operation and how the cognitive framework of attentional operation in working memory can be implemented with a dynamic neural network in the brain. Based on the findings and implications, directions for future research are discussed.Table of Contentscknowledgments ibstract (Chinese) iibstract (English) iiihapter 1: Introduction 1-1. Attentional selection of relevant information in perception 6 1-1-1. Behavioral measures of stimulus salience biasing the competition in perception 8 1-1-2. The neural basis of visuospatial attention 10 1-1-3. Top-down control modulates visual processing efficiency 13 1-1-4. The neural correlates of top-down and bottom-up attentional processes 17 1-1-5. Conclusion 21-2. The nature of working memory 22 1-2-1. Cognitive models of working memory 23 1-2-2. Neural correlates of working memory 26 1-2-3. Neural source of top-down control 30 1-2-4. Capacity limitation and competition of working memory representations 33 1-2-5. Conclusion 36-3. The role of attentional operation in working memory 37-3-1. Orienting attention in working memory 38-3-2. Attention-based rehearsal hypothesis 43-3-3. Attentional prioritization in working memory 47-3-4. Conclusion 51-4. Hypotheses and predictions 52hapter 2: Neuroimaging methodologies 57-1. Principles and methodologies of EEG/ERPs 58-1-1. ERP physiology 59-1-2. ERP recording 60-1-3. ERP interpreting 63-1-4. Topographical analysis 65-2. Principles and methodologies of fMRI 66-2-1. fMRI physiology 67-2-2. fMRI experimental design 69-2-3. fMRI data processing 71-2-4. fMRI univariate statistical analysis 74-2-5. fMRI multivariate statistical analysis 76hapter 3: Searching for targets within the spatial layout of working memory: An ERP study 81-1. Introduction 81-2. Methods and materials 85-3. Results 96-4. Discussion 103-5. Conclusion 108hapter 4: The neural correlates of the effectiveness of orienting attention during working memory retention affects the difficulty of mnemonic evaluation 111-1. Introduction 111-2. Methods and materials 118-3. Results 125-4. Discussion 136-5. Conclusion 142hapter 5: Neural correlates of top-down control of refreshing in working memory 145-1. Introduction 145-2. Methods and materials 150-3. Results 156-4. Discussion 168-5. Conclusion 177hapter 6: Functional connectivity of attentional operation in working memory 179-1. Introduction 179-2. Methods and materials 183-3. Results 188-4. Discussion 196-5. Conclusion 203hapter 7: General Discussion 205-1. Objectives 205-2. Summary of the main findings 207-3. Discussion 210-4. Future studies 223-5. Conclusion 223eference 225ist of Figuresigure 1-1. The structure of the thesis. 53igure 2-1. The 40-channel EEG system. 61igure 3-1. Schematic of experimental trials. 89igure 3-2. Sensitivity measures (d’) and mean response times (ms) in target present trials across the two types of search tasks (visual and mnemonic) and two types of search load (2-item and 4-item) were measured in (a) Experiment 1 and (b) Experiment 2. 98igure 3-3. The N2pc effects in the visual and mnemonic domains showed equivalent time-courses (grand-averaged waveforms) and topographies in (a) Experiment 1 and (b) Experiment 2. 101igure 4-1. An example of (a) an early cue in the binding -change condition, and (b) a late cue in the color -change condition. 115igure 4-2. (a) Mean accuracy (% correct) and (b) reaction time (RT in ms) of correct response in each condition. 126igure 4-3. Areas of significant activation associated with the cueing effect. 128igure 4-4. Areas of significant activation associated with the effect of change type. 130igure 4-5. Areas of significant activation related to the interaction effect. 133igure 4-6. Areas of significant activation related to the effects of (a) color-change minus binding-change under early cueing and (b) binding-change minus color-change under late cueing. 135igure 5-1. Examples of (a) a sustained cue, (b) transient cue, (c) test cue, and (d) no-cue control change-detection tasks. 149igure 5-2. (a) Mean accuracy (% correct) and (b) reaction time (RT in ms) of correct response in each condition. 157igure 5-3. Areas of significant activation related to attentional and executive control from the contrast of change (grey bar) minus no-change (white bar) trials. 159igure 5-4. Areas of significant activation associated with the cueing effect. 161igure 5-5. An anatomical gradient effect from posterior regions (MOG, SPL, and PCu) to anterior regions (PMd and MFG) of the right hemisphere was observed under transient cueing. 167igure 6-1. Coherence maps of the left MFG seed (a) and the left SPL/IPS seed (b) for early-cueing and late-cueing conditions from data set 1 (Chapter 4). 189igure 6-2. The comparison of the coherence maps for early-cueing condition with late-cueing condition. 190igure 6-3. Coherence maps of the left MFG seed (a) and the left SPL/IPS seed (b) for sustained-cueing, transient-cueing, test-cueing and no-cue conditions from data set 2 (Chapter 5). 192igure 6-4. The comparison of the coherence maps for sustained- and transient-cueing conditions with test-cueing condition respectively under the left MFG as a seed region. 194igure 6-5. The comparison of the coherence maps for sustained- and transient-cueing conditions with test-cueing condition respectively under the left SPL/IPS as a seed region. 195igure 6-6. Summary of the findings of functional connectivity. 197ist of Tablesable 3-1. Mean d’ scores and reaction times (RTs in ms) of correct responses in each condition. 98able 4-1. Mean accuracy (% correct) and reaction time (RT in ms) of correct response in each condition. 127able 4-2. Brain regions and their Talairach coordinates in which significantly activated clusters were observed for successful detection and correct rejection. 129able 4-3. Brain regions and their Talairach coordinates in which significantly activated clusters were observed for binding-change minus color-change and color-change minus binding-change. 131able 4-4. Brain regions and their Talairach coordinates in which significantly activated clusters were observed for the interaction effect between cue onset time and change type. 132able 4-5. Brain regions and their Talairach coordinates in which significantly activated clusters were observed for color-change minus binding-change under early cueing and binding-change minus color-change under late cueing. 134able 5-1. Mean accuracy (%) and reaction times (RTs in ms) of correct responses in each condition. 158able 5-2. Brain regions and their Talairach coordinates in which significantly activated clusters were observed for the change versus no-change trials (p < 0.001). 160able 5-3. Brain regions and their Talairach coordinates in which significantly activated clusters were observed for each cueing condition versus no-cue control condition (p < 0.001). No significant activation was observed in test-cueing condition minus no-cue condition under no-change trials. 163application/pdf2770528 bytesapplication/pdfen-US事件相關電位功能性磁振造影大腦功能性聯結注意力調控工作記憶Event-related potentialfunctional magnetic resonance imagingfunctional connectivityorienting attentionworking memoryhttp://ntur.lib.ntu.edu.tw/bitstream/246246/178724/1/ntu-98-D92227008-1.pdf