精度需求和特征数量对视觉工作记忆容量资源分配的影响

发布时间：2018-07-06 14:27

本文选题：视觉工作记忆容量 + 精度需求　；参考：《首都师范大学》2014年硕士论文

【摘要】：作为大脑认知加工的核心部分,工作记忆的容量决定了我们所能够同时加工的信息数量。尽管工作记忆的容量已被大量研究证明是极其有限的,但决定工作记忆中能够存储的项目数量的根本原因是工作记忆容量资源的分配模式。在对这一分配模式进行解释时,出现了各种不同的假设和理论。争议点集中于项目复杂度对容量分配的影响、项目特征数量对容量分配的影响和表征精度与容量分配的关系。考虑到这三个问题间的重叠性,当前研究试图利用对简单刺激的适当操控从复杂度因素中分离出精度需求和特征数量的影响,在控制记忆特征本身的差异和单个项目所包含的特征数量的前提下,操控对记忆项目的精度需求来考察纯粹的精度需求对视觉工作记忆容量资源分配的影响,并分别在不同的精度需求条件下检验特征数量对容量资源分配的影响。研究一使用单色方块作为记忆目标,通过改变记忆项目和测验项目的颜色相似性来操控对记忆项目的表征精度需求。研究结果发现,排除了测验难度的影响而获得的容量估值与电生理学指标都表明精度需求的提高未能使容量下降,尽管可能提高了表征的精度。研究二为了检验这一结果是否适用于记忆多特征项目时的情况,使用了双色方块作为记忆目标,通过变化记忆项目和测验项目间可能发生改变的颜色数量来操控精度需求,并加入单色块条件和两种双色块条件的比较来考察特征数量的效应。结果表明,对于多特征项目精度需求依然不会对容量分配造成影响,并且同样可能提高了与精度相关的控制资源的投入,但特征数量的增多则会导致更多容量资源的消耗,却不一定会导致更多控制资源的投入。总之,当前的两项研究表明,不管是记忆单特征还是多特征项目,精度需求的提高尽管可能会诱发表征精度的提高但不会使容量下降,而特征数量的增加会使容量资源的消耗增多并与精度需求无关。以上结果与视觉工作记忆容量的插槽-资源模型更为相符,即存储数量限制与表征精度限制是由相互分离的两种资源决定的。
[Abstract]:As the core part of cognitive processing, the capacity of working memory determines the amount of information we can process at the same time. Although the capacity of working memory has been proved to be extremely limited by a large number of studies, the fundamental reason for determining the number of items that can be stored in working memory is the allocation pattern of working memory capacity resources. Various hypotheses and theories have emerged in the explanation of this distribution model. The controversy focuses on the impact of project complexity on capacity allocation, the effect of item feature number on capacity allocation and the relationship between representation accuracy and capacity allocation. Considering the overlap between these three problems, current studies attempt to separate the effects of precision requirements and the number of features from complexity factors using appropriate manipulation of simple stimuli. Under the premise of controlling the differences of memory features and the number of features contained in a single item, the effect of pure precision requirements on the allocation of resources of visual working memory capacity is investigated by manipulating the accuracy requirements of memory items. The effect of quantity of feature on capacity resource allocation is tested under different precision requirements. In this study, monochromatic squares were used as memory targets to manipulate the accuracy requirements of memory items by changing the color similarity of memory items and test items. It is found that both the capacity estimation and the electrophysiological indexes obtained after eliminating the influence of the difficulty of the test indicate that the increase of accuracy demand fails to reduce the capacity although it may improve the accuracy of characterization. In the second study, in order to test whether this result is applicable to memory multi-feature items, we used two-color squares as memory targets to manipulate precision requirements by changing the number of colors that may change between the memory items and the test items. The effect of characteristic quantity is investigated by adding the comparison of monochromatic block condition and two color block conditions. The results show that the accuracy requirements of multi-feature projects still have no effect on capacity allocation, and may also increase the input of precision related control resources, but the increase of the number of features will lead to the consumption of more capacity resources. But it does not necessarily lead to more investment in controlling resources. All in all, two current studies have shown that, regardless of whether it is a memory single feature or a multi-feature item, increased accuracy requirements may lead to improved representation accuracy, but not to a decrease in capacity. The increase in the number of features will increase the consumption of capacity resources and independent of accuracy requirements. The above results are more consistent with the slot resource model of visual working memory capacity, that is, the storage quantity limit and the representation accuracy limit are determined by the two resources separated from each other.
【学位授予单位】：首都师范大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：B842.3

【参考文献】