ISSN 0439-755X
CN 11-1911/B

心理学报 ›› 2016, Vol. 48 ›› Issue (11): 1370-1378.doi: 10.3724/SP.J.1041.2016.01370

• 论文 • 上一篇    下一篇

任务设置影响负相容效应的机制 ——自上而下认知控制对阈下启动信息加工的影响

王佳莹1; 缴润凯1; 张 明2   

  1. (1东北师范大学心理学院, 长春 130024) (2苏州大学心理学系, 苏州 215123)
  • 收稿日期:2015-07-15 发布日期:2016-11-25 出版日期:2016-11-25
  • 通讯作者: 张明, E-mail:
  • 基金资助:


The mechanism of the effect of task setting on negative compatibility effect: The effect of top-down cognition control on subliminal prime processing

WANG Jiaying1; JIAO Runkai1; ZHANG Ming2   

  1. (1 School of Psychology, Northeast Normal University, Changchun 130024, China) (2 Department of Psychology, Soochow University, Suzhou 215123, China)
  • Received:2015-07-15 Online:2016-11-25 Published:2016-11-25
  • Contact: ZHANG Ming, E-mail:


结合掩蔽启动范式和Go-Nogo范式, 考察了任务设置影响负相容效应的机制。实验中以指向左或右的双箭头为掩蔽启动项, 在两个Block中的Go试次中分别以箭头和汉字为目标项, 要求被试对箭头或汉字做辨别反应, Nogo试次中不呈现目标项, 被试无需反应。结果发现:Go试次中, 只有在箭头任务设置下才出现负相容效应, 脑电结果表现为一致条件下P3潜伏期显著长于不一致条件; Nogo试次中, 箭头任务设置下的P3波幅显著大于汉字任务设置下的P3波幅。说明负相容效应受到任务设置这种自上而下认知控制过程的影响, 且这种影响发生在阈下启动信息加工阶段。支持注意敏化模型, 提示可以在更普遍的角度上理解负相容效应。

关键词: 负相容效应, 任务设置, 自上而下认知控制, 阈下启动信息加工


Negative Compatibility Effect (NCE) is the surprising result that the masked prime arrows inhibit the responses to compatible target arrows and facilitate the responses to opposite target arrows when the prime-target Inter-Stimulus-Interval (ISI) is between 100ms and 150ms in the masked prime paradigm. The theoretical argument is mainly about whether the origin of NCE is perceptual or reactive. But almost all the theories considered that subliminal information processing was automatic and can not be regulated by top-down cognition control process. However, recent studies showed that NCE can be affected by task setting, which is considered as a Top-down cognitive control processing. How does the task setting affect NCE? The Effect of top-down cognitive control on NCE might happen at the subliminal prime information processing stage or at the reaction stage. Therefore, this study manipulated task setting to explore the mechanism of the effect of top-down cognitive control processes on NCE. The research combined masked prime paradigm and Go-NoGo paradigm to investigate the effects and mechanism of task setting on NCE. Eighteen students participated in the study. The experiment adopted classical NCE procedure and stimulus. Primes were double arrows pointed either to left or right, the mask was composed of two double arrows (one pointed to the left and other to the right). Targets were characters or double arrows in Go condition, either compatible or incompatible to the prime. Participants were instructed to response to the arrows or the characters by pressing the left or right arrow keys on keyboard in Go condition. There’s no target in NoGo condition. Reaction times and accuracies were analyzed with a 2 (compatibility) × 2 (task setting) analyses of variance (ANOVA). Behavior data showed that the main effects of compatibility and task setting and the interaction between them were significant. Further analyses on interaction showed that NCE was significant only in the arrow target condition, in which prime is related to task setting and there was no significant effect in the character target condition, in which prime is not related to task setting. ERP waveform analyses revealed that P3 latency is longer in compatibility condition than in incompatibility condition only when the targets were arrows, which means that P3 is only longer in the compatibility condition when prime is related to the task setting in the Go trials. Mean amplitude of P3 has a significant difference between arrow target condition and character target condition in the NoGo trials. The results suggest that task setting has effect on the NCE through the top-down cognition control process. The cognitive control system regulates the top-down cognitive process according to the task requirements, which impacts on the subliminal prime information processing.

Key words: negative compatibility effect, task setting, top-down cognitive control process, subliminal priming information processing