对象(语法)
代表(政治)
计算机科学
感知
启动(农业)
参照系
人工智能
帧(网络)
参考坐标系
领域(数学)
过程(计算)
计算机视觉
沟通
心理学
数学
程序设计语言
电信
植物
发芽
物理
量子力学
神经科学
政治
政治学
纯数学
法学
生物
作者
Daniel Kahneman,Anne Treisman,Brian Gibbs
标识
DOI:10.1016/0010-0285(92)90007-o
摘要
A series of experiments explored a form of object-specific priming. In all experiments a preview field containing two or more letters is followed by a target letter that is to be named. The displays are designed to produce a perceptual interpretation of the target as a new state of an object that previously contained one of the primes. The link is produced in different experiments by a shared location, by a shared relative position in a moving pattern, or by successive appearance in the same moving frame. An object-specific advantage is consistently observed: naming is facilitated by a preview of the target, if (and in some cases only if) the two appearances are linked to the same object. The amount and the object specificity of the preview benefit are not affected by extending the preview duration to 1 s, or by extending the temporal gap between fields to 590 ms. The results are interpreted in terms of a reviewing process, which is triggered by the appearance of the target and retrieves just one of the previewed items. In the absence of an object link, the reviewing item is selected at random. We develop the concept of an object file as a temporary episodic representation, within which successive states of an object are linked and integrated.
科研通智能强力驱动
Strongly Powered by AbleSci AI