Cole Geoff G, Kentridge Robert W, Gellatly Angus R H, Heywood Charles A
Department of Psychology, University of Durham, UK.
J Vis. 2003;3(1):22-31. doi: 10.1167/3.1.3.
The human visual system is particularly sensitive to abrupt onset of new objects that appear in the visual field. Onsets have been shown to capture attention even when other transients simultaneously occur. This has led some authors to argue for the special role that object onset plays in attentional capture. However, evidence from the change detection paradigm appears contradictory to such findings. Studies of change blindness demonstrate that the onset of new objects can often go unnoticed. Assessing the relative detectability of onsets compared with other visual transients in a change detection procedure may help resolve this contradiction. We report the results of four experiments investigating the efficacy with which onsets capture attention compared with offsets. In Experiment 1, we employed a standard flicker procedure and assessed whether participants were more likely to detect the change following a frame containing an onset or following a frame containing an offset. In Experiment 2, we employed the one-shot method and investigated whether participants detected more onsets or offsets. Experiment 3 used the same method but assessed whether onsets would be detected more rapidly than offsets. In Experiment 4, we investigated whether the effect obtained in Experiments 1-3 using simple shapes would replicate when images of real-world objects were used. Results showed that onsets were less susceptible to change blindness than were offsets. We argue that the preservation of information is greater in onsets than in offsets.
人类视觉系统对出现在视野中的新物体的突然出现特别敏感。即使其他瞬变现象同时发生，物体的出现也已被证明能吸引注意力。这使得一些作者认为物体出现在注意力捕获中起着特殊作用。然而，来自变化检测范式的证据似乎与这些发现相矛盾。变化盲视的研究表明，新物体的出现往往会被忽视。在变化检测过程中评估物体出现与其他视觉瞬变现象相比的相对可检测性，可能有助于解决这一矛盾。我们报告了四项实验的结果，这些实验研究了物体出现与消失相比在吸引注意力方面的效果。在实验1中，我们采用了标准的闪烁程序，评估参与者是否更有可能在包含物体出现的帧之后或包含物体消失的帧之后检测到变化。在实验2中，我们采用了单次呈现方法，研究参与者是否检测到更多的物体出现或消失。实验3使用了相同的方法，但评估物体出现是否比物体消失能更快被检测到。在实验4中，我们研究了使用简单形状在实验1 - 3中获得的效果，当使用真实物体图像时是否会重现。结果表明，与物体消失相比，物体出现对变化盲视的敏感度更低。我们认为，物体出现时的信息保留比物体消失时更多。
