Suppr超能文献

双稳态感知中的时间动态。

Temporal dynamics in bistable perception.

作者信息

Mamassian Pascal, Goutcher Ross

机构信息

CNRS & Université Paris 5, Boulogne-Billancourt, France.

出版信息

J Vis. 2005 Apr 27;5(4):361-75. doi: 10.1167/5.4.7.

Abstract

Bistable perception is fundamentally a dynamic process: Our perceptual experience continuously alternates when an ambiguous or rivalrous stimulus is observed. Here we present a method to analyze instantaneous measures of dominance and transition between percepts. The analysis extracts three time-varying probabilities. First, the transient preference represents the probability of perceiving one interpretation at one instant. Second, the reversal probability is the probability that the current percept will change at the next evaluation. Finally, the survival probabilities are the probability that at one instant the current percept will not switch to the alternative interpretation. We derive the relationships between these probabilities and offer a test of independence between consecutive percepts. We also introduce a simple technique to sample the observer's perception at regular intervals. The analyzing method is illustrated with the example of binocular rivalry. We demonstrate Levelt's second proposition with the survival probability measure and show that the consecutive rivalrous percepts are not independent.

摘要

双稳态感知本质上是一个动态过程

当观察到一个模糊或竞争性刺激时,我们的感知体验会持续交替。在这里,我们提出一种方法来分析感知之间优势和转换的瞬时测量。该分析提取三个随时间变化的概率。首先,瞬态偏好表示在某一时刻感知一种解释的概率。其次,反转概率是当前感知在下一次评估时会发生变化的概率。最后,存活概率是在某一时刻当前感知不会切换到另一种解释的概率。我们推导了这些概率之间的关系,并提供了对连续感知之间独立性的检验。我们还介绍了一种简单的技术,以定期间隔对观察者的感知进行采样。以双眼竞争为例说明了该分析方法。我们用存活概率测量证明了莱尔特的第二个命题,并表明连续的竞争性感知不是独立的。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验