工作记忆是暂时存储和加工信息的记忆系统，在词汇学习、阅读理解、数学计算、问题解决等各种各样认知活动中起到重要的作用。大量研究表明，工作记忆，特别是视觉工作记忆 （visual working memory）的容量是有限的。个体视觉工作容量的差异不仅能预测其智力水平，还可以预测学业成绩，同时精神疾病也会导致视觉工作记忆容量的缺陷。已有行为和神经研究证据表明，视觉工作记忆容量可能受到注意控制和注意范围等因素的综合影响。其中注意控制可以过滤无关信息从而提高视觉工作记忆容量，而注意范围较大则可以暂时存储更多信息。但目前围绕视觉工作记忆容量限制的神经机制还存在许多重要问题需要解决。首先，尚没有研究从大样本个体差异角度揭示影响个体视觉工作记忆容量的因素，特别是检验注意控制和注意范围是否同时影响视觉工作记忆容量的大小。其次，虽然额顶区域在视觉工作记忆中起到重要作用，但尚不清楚其在注意控制和注意范围中是否起到不同的作用。为了回答上述问题，本研究综合采用多模态脑影像和脑刺激技术，结合个体差异和因果操纵的研究思路，通过三个研究系统考察了影响个体视觉工作记忆容量的重要因素及其神经机制。 研究一从大样本（N = 439）个体差异角度出发，在同一个视觉工作记忆任务中分离出注意控制和注意范围这两个因素，直接比较它们对个体视觉工作记忆（采用空间2-back任务测查）及流体智力（采用瑞文高级推理测验测查）的影响。结果发现，注意控制和注意范围两者只存在微弱的相关关系（r = 0.102, p = 0.032），两者分别对个体的视觉工作记忆和流体智力产生贡献。这一结果在大样本上重复和验证了注意控制和注意范围共同影响视觉工作记忆的结论。 研究二在研究一的基础上，进一步使用多模态脑影像大数据，结合脑结构的灰质体积和静息态功能网络的连接强度，揭示注意控制和注意范围背后不同的神经结构和功能基础。结果发现，在灰质体积上，额下回（inferior frontal gyrus，IFG）的灰质体积能预测注意控制，而与注意范围不相关，顶上小叶（superior parietal lobule, SPL）的灰质体积能预测注意范围，而与注意控制不相关。在静息态功能网络上，额下回的功能连接强度（全脑节点中心度；degree centrality, DC）能预测注意控制，而与注意范围不相关；顶上小叶的功能连接强度能预测注意范围，而与注意控制不相关。 研究三使用经颅直流电刺激（transcranial direct current stimulation，tDCS）的方法进一步对上述研究结果进行因果验证，本研究包含两个实验，分别使用双侧视野呈现和中心呈现的范式，考察注意控制和注意范围在认知上的分离，及额叶与顶叶在注意控制和注意范围中的不同作用。结果发现，对右侧额叶施加20分钟的直流电刺激可以提高个体注意控制的能力，但不影响个体的注意范围；相反，对右侧顶叶施加直流电刺激可以提高个体的注意范围，但不影响注意控制能力。这个结果进一步从因果关系角度揭示额顶在注意控制和注意范围上的功能分离，并找到了无损提高个体视觉工作记忆容量的方法。 综上，本研究的结果表明：1）视觉工作记忆容量受多种心理过程的影响，其中包括注意控制和注意范围在内的多种认知过程共同贡献于个体的视觉工作记忆，这些因素也能预测个体的流体智力水平；2）个体额叶和顶叶的脑结构和脑功能的差异分别影响注意控制和注意范围；3）使用经颅直流电刺激的方法刺激额叶和顶叶可以分别通过影响注意控制和注意范围提高个体视觉工作记忆的容量。这些研究结果加深了我们对于视觉工作记忆的底层认知神经机制的理解，并为改善个体工作记忆容量提供了新的思路和方法。研究结果将在特殊人才培养和精神疾病治疗等方面具有潜在的应用价值。

Working memory is the ability to temporally maintain and process information, which plays an important role in verbal learning, reading comprehension, mathematical computation, problem solving and etc. Mounting research has shown that the capacity of working memory, especially visual working memory, is limited and varies significantly across individuals. Visual working memory capacity can predict individuals’ general intelligence and academic performance. Meanwhile, people suffering mental illness also showed deficits in visual working memory capacity. Both behavior and neural research suggest that visual working memory capacity is affected by factors like attention control and attention scope: whereas attention control facilitates visual working memory capacity by filtering out irrelevant information, attention scope determines the amount of information one can maintain in working memory. Nevertheless, the cognitive and neural mechanisms underlying attention control and attention scope which affect visual working memory capacity remain elusive. In particular, there is a lack of big sample individual difference research to dissociate the contribution of attention control and attention scope to individual variances in visual working memory capacity. Still, although the frontal and parietal cortices both play important roles in visual working memory, it is unclear whether these two brain regions play different roles in attention control and attention scope. The current research aimed to address these questions using multi-modal brain imaging and brain stimulation techniques. Specifically, in three studies, we combined individual difference and laboratory manipulation to systematically explore the cognitive and neural mechanisms of attention control and attention scope in visual working memory. Study One used individual difference approach with big sample data (N = 439). In the same visual working memory task, we decomposed attention control and attention scope, and examined their contributions to visual working memory performance (i.e., measured by spatial 2-back task) and fluid intelligence (measured by Raven advanced progressive matrices). We found that the correlation between attention control and attention scope was small (r = 0.102, p = 0.032) and both of them contributed to individual visual working memory and fluid intelligence. Our results thus replicated previous findings with large sample that attention control and attention scope cohesively influence individual visual working memory. Study Two further examined the neural correlates of attention control and attention scope using large sample and multi-modal imaging data, i.e., brain structure and functional connectivity. We found that the grey matter volume of inferior frontal gyrus predicted individuals’ attention control score but did not predict their attention scope score. In contrast, the grey matter volume of superior parietal lobule predicted individuals’ attention scope score but not their attention control score. In addition, resting-state functional connectivity data suggested that the functional connectivity strength (indexed by degree centrality) of the inferior frontal gyrus predicted individuals’ attention control score but not their attention scope score, whereas the functional connectivity strength of the superior parietal lobule predicted individuals’ attention scope score but not their attention control score. Together, these data revealed a clear double dissociation between frontal and parietal cortices in attention control and attention scope, respectively. Study Three used transcranial direct current stimulation (tDCS) to further examine the causal roles of frontal and parietal cortices in attention control and attention scope. We found that, 20 minutes of anodal stimulation on the right frontal cortex facilitated attention control but not attention scope. In contrast, anodal stimulation on the right parietal cortex enhanced individuals’ attention scope but not attention control. These results not only reveal a causal roles of frontal and parietal cortices in attention control and attention scope, but also suggest a non-invasive way to enhance individuals’ visual working memory capacity. Taken together, these findings suggest that: 1) individual visual working memory capacity is influenced by both attention control and attention scope; 2) structural and functional differences in frontal and parietal lobes respectively underlies individuals’ attention control and attention scope score; 3) Anodal stimulation on the frontal and parietal cortices can separately facilitate individuals’ visual working memory capacity by enhancing their attention control and attention scope ability, respectively. These results not only deepen our understanding of the cognitive and neural mechanisms underlying visual working memory, but also provide potential novel intervention to enhance individual working memory capacity, which has significant implications in special education and health care.