Spatiotemporal dynamics of memory and attention in recognition of dynamic scenes

动态场景识别中记忆和注意力的时空动态

• 批准号: 11610075
• 负责人: SAIKI Jun
• 金额: $ 2.69万
• 依托单位: KYOTO UNIVERSITY
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (C)
• 财政年份: 1999
• 资助国家: 日本
• 起止时间: 1999 至 2000
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-11610075/
• 关键词: Visual short term memory; Visual attention; Information integration; Change blindness

(1) To study the integration of different visual features in dynamic situations, we investigated the integration of color and spatiotemporal location in visual working memory by using a modified version of the flicker paradigm. Inconsistent with the previous claims that we can store four to five object information simultaneously in visual working memory, it was extremely difficult to store the information of only three objects in a dynamic and multidimensional situation. These data strongly suggest that in dynamic situations involving objects' motion, we do not construct multiple object representations integrating multidimensional information. The results also showed that the difficulty in storage and transformation cannot be attributed to the lack of local motion signals, and that spatiotemporal predictability had some facilitatory effects on performance. Furthermore, in a task where the combination of shape and color is necessary, the availability of object representations where perc … More eptual features are integrated is quite low in the process of dynamic transformation and storage of visual information.(2) Change blindness phenomenon suggests that visual short-term memory has severe capacity limit. To investigate what kind of visual information is stored in visual short-term memory, we conducted a change detection task with subjective contour figures. In a task where participants were asked to search the presence of shape change across blank periods, we have shown that even when the amount of physical figural change is equivalent, the search was more accurate when the changing objects contains subjective surface figures. Moreover, this effect occurs not at the level of subjective contours, but at the level of subjective surface representations.(3) We conducted a series of experiments to examine whether an occluded object functions as a single object for the visual attention mechanism. In the previous study supporting this hypothesis, the effects of amodal completion and symmetry of objects were confounded. We examined whether the effect of amodal completion is obtained with asymmetric objects. The results revealed that the effects of amodal completion completely disappeared with the asymmetric objects. Less

(1)为了研究动态情境下不同视觉特征的整合，我们采用改进的闪烁范式，考察了颜色和时空位置在视觉工作记忆中的整合。与之前的观点不一致的是，我们可以在视觉工作记忆中同时存储四到五个物体的信息，在动态和多维的情况下，只存储三个物体的信息是非常困难的。这些数据有力地表明，在涉及对象运动的动态情况下，我们不会构建集成多维信息的多个对象表示。结果还表明，存储和转换的困难，不能归因于缺乏本地运动信号，和时空的可预测性有一定的促进作用的性能。此外，在需要形状和颜色组合的任务中， ...更多信息 在视觉信息的动态转换和存储过程中，视觉特征的整合度很低。(2)变化盲视现象表明视觉短时记忆存在严重的容量限制。为了研究视觉短时记忆中储存的是何种视觉信息，我们采用主观轮廓图进行了变化检测任务。在一个任务中，参与者被要求搜索的形状变化的存在跨越空白期，我们已经表明，即使当物理图形的变化量是相等的，搜索是更准确的变化对象时，包含主观的表面图形。此外，这种效应不是发生在主观轮廓的水平上，而是发生在主观表面表征的水平上。(3)我们进行了一系列的实验，以检查是否被遮挡的对象功能作为一个单一的对象的视觉注意机制。在先前支持这一假设的研究中，非模态完成和物体对称性的影响被混淆了。我们研究了非模态完成的效果是否可以通过不对称物体获得。结果表明，非模态完成的影响完全消失的不对称对象。少

项目成果

Saiki,J.: "Occlusion, symmetry, and object-based attention : Comment on Behrmann. Zemel, and Mozer(1998)."Journal of Experimental Psychology : Human Perception and Performance. 26・1. 424-433 (2000)

Saiki, J.：“遮挡、对称性和基于对象的注意：对 Behrmann. Zemel 和 Mozer (1998) 的评论。”实验心理学杂志：人类感知和表现 26・1 (2000)。

齋木潤: "視覚情報処理ハンドブック(分担執筆)"朝倉書店. 7 (2000)

齐木淳：《视觉信息处理手册（合着）》朝仓书店7（2000）。

Saiki, J.: "Handbook of visual information processing"Asakura-Shoten. 7 (2000)

Saiki, J.：《视觉信息处理手册》朝仓书店。

遠藤信貴,齋木潤,齋藤洋典: "新奇図形の照合課題におけるネガティブ・プライミングの生起の規定因."心理学研究. (印刷中).

Nobutaka Endo、Jun Saiki、Hironori Saito：“小说人物匹配任务中负启动发生的决定因素”（正在出版）。