自我运动中Optic flow对物体运动知觉的影响机制研究

Humans and animals could approach a target by two different types of movements. In object movement, the moving target approaches the stationary observer; while in self-motion, the observer approaches the stationary target. From the relative movement perspective, observer's responses in both motion conditions should be identical as long as the relative velocities are the same. However, from the cognitive perspective,different visual patterns in the two conditions will be perceived. In object movement, only the expansion of the target image on the retina was useful to the motion perception. On the other hand, in self-motion condition, in addition to the same expansion cue, the pattern of the images of the background objects moving off the view field (optic flow) could also be observed. Therefore, it is important to understand whether such a visual discrepancy could result in observer's different behaviors in these two movements. In present study, we are going to investigate this issue psychophysically. The characteristics of different available cues in self-motion were manipulated in a series of experiments. These characteristics include: optic flow velocity, direction, the expansion of the image of a distracter, the visual location of optic flow, the eccentricity of optic flow, and the combinations of some of them. By comparing observer's responses between the two movement conditions, we could find what role optic flow plays in the perception of the expansion of the target image. Furthermore, we are also able to discuss the cognitive mechanisms underlying the visual motion perception. This study has important implications in many fields in our daily life, such as the design of video games, the interface of human-computer interaction and human reaction training in driving-simulator and flight-simulator.

人类和动物为了接触到一个物体，可以使用两种方式。一种是让物体运动到自己面前（物体运动）；另外一种则是自己走到物体所在位置（自我运动）。从物理学的角度来看，当相对速度相同时，这两种运动方式在所用的时间上是等同的，因而观察者应该有同样的行为反应。然而，从认知角度来看，二者在视觉上有着明显的差异。当物体运动时，视野内只有物体自身在视网膜上的映像发生扩张；而在自我运动时，不仅物体映像在扩大，周围背景物体的视网膜映像也在发生位移（optic flow）。由此引出本研究的主题：这种视觉差异是否会导致人们在两种运动方式中的时间判断和行为反应的不一致？本研究试图使用心理物理法来探讨这一问题。通过设计和使用一系列的实验，我们将从多个角度对认知系统对optic flow，分心物的视网膜映像扩张和目标物映像扩张的敏感性以及不同视觉信息之间的相互关系进行详细研究。

人类可以通过自身运动或者让目标物发生运动这两种不同的方式来接触（或碰撞）到目标物体，在此过程中，人类需要估计从当前时刻到发生接触所需要的运动时间并做出恰当的行为反应。本项目使用心理物理学方法，研究当人类自身处于静止状态或者运动状态时，物理意义上完全相等的运动时间是否能够如实地被其视觉感知到，并做出完全相同的行为反应。结果发现，当人自身处于静止状态时，其倾向于低估目标物体与其发生碰撞所需的运动时间。与之相比，当人运动时，其对同一运动时间的感知和估计与静止状态下的估计有明显的差异。具体来说，当向被试呈现模拟其前向运动的视觉信息时，被试倾向于相对地高估所需的运动时间；当向其呈现后退运动的视觉信息时，通常会发生更明显的低估现象。因此，我们提出一种观点：自我运动引发的视觉optic flow信息同映像扩张信息在视觉上具有功能等同性，观察者觉察到的目标物体视觉扩张是一种基于其直接背景的相对扩张。这一结果对于运动能力训练，虚拟现实设备使用，驾驶技能培训等方面有重要的指导意义。