School of Psychology, University of Sydney, Brennan-MacCallum Building A18, Sydney, 2006, Australia.
Mem Cognit. 2019 Jul;47(5):1024-1030. doi: 10.3758/s13421-019-00905-9.
Repetition blindness (RB) is the inability to detect both instances of a repeated stimulus during rapid serial visual presentation (RSVP). Prior work has demonstrated RB for semantically related critical items presented as pictures, but not for word stimuli. It is not known whether the type of semantic relationship between critical items (i.e., conceptual similarity or lexical association) determines the manifestation of semantically mediated RB, or how this is affected by the format of the stimuli. These questions provided the motivation for the present study. Participants reported items presented in picture or word RSVP streams in which critical items were either low-associate category coordinates (horse-camel), high-associate noncoordinates (horse-saddle), or unrelated word pairs (horse-umbrella). Report accuracy was reduced for category coordinate critical items only when they were presented in pictorial form; accuracy for coordinate word pairs did not differ from that of their unrelated counterparts. Associated critical items were reported more accurately than unrelated critical items in both the picture and word versions of the task. We suggest that semantic RB for pictorial stimuli results from intracategory interference in the visuosemantic space; words do not reliably suffer from semantic RB because they do not necessitate semantic mediation to be reported successfully. Conversely, the associative facilitation observed in both picture and word versions of the task reflects the spread of activation between the representations of associates in the lexical network.
重复盲视（RB）是指在快速序列视觉呈现（RSVP）中无法检测到重复刺激的两个实例。先前的研究表明，在呈现图片形式的语义相关关键项目时会出现 RB，但在呈现词刺激时则不会。目前尚不清楚关键项目之间的语义关系类型（即概念相似性或词汇联想）是否决定了语义介导 RB 的表现，或者刺激的格式如何影响这种关系。这些问题为本次研究提供了动力。参与者报告了在图片或词 RSVP 流中呈现的项目，其中关键项目是低联想类别坐标（马-骆驼）、高联想非坐标（马-马鞍）或不相关的词对（马-伞）。只有当类别坐标关键项目以图片形式呈现时，报告的准确性才会降低；坐标词对的准确性与不相关词对的准确性没有差异。在任务的图片和词两种形式中，关联的关键项目比不相关的关键项目报告得更准确。我们认为，图片刺激的语义 RB 是由于视觉语义空间中的类别内干扰所致；词不会可靠地受到语义 RB 的影响，因为它们不需要语义中介即可成功报告。相反，在任务的图片和词两种形式中观察到的联想促进反映了词汇网络中联想表示之间的激活扩散。
