Abstract: There exist many phenomena of affective priming in evaluative categorization task,such as congruency proportion effects,reverse priming effect and target-word-frequency effect etc. Until now,any affective priming model can not explain all appearances occurring in affective priming,which exposes deficiencies of these models. For this reason,four theorems were established in the thesis. In the first experiment,it is confirmed that the reaction time on congruent prime-target pairs when participants’ attentional weight is assigned to the congruent trials was shorter than RT on incongruent prime-target pairs when assigned to the incongruent trials,which is shorter than RT on congruent prime-target pairs when assigned to the incongruent trials. In the second experiment,a sequence of the reaction speed to prime-target trials was determined from fast to slow,which is negative-negative trial,positive-negative trial,negative-positive trial,positive-positive trial. In the third experiment,we confirmed an experimental dissociation of accuracy-emphasized task from speed-emphasized task. Using these theorems,any phenomenon of affective priming in the evaluative categorization task could be explained.

摘要： 评价分类任务中的情绪启动效应有很多现象，如比例效应、反转效应、词频效应。到目前为止，任何一个启动模型均无法解释所有的情绪启动现象，表明现有的模型均有理论不足。为此提出了四条定理，实验一证实了启靶一致RT(一致性期待时)<启靶不一致RT(不一致性期待时)<启靶一致RT(不一致性期待时);实验二证实了“启—靶”反应从快到慢依次是:N(负)—N、P(正)—N、N—P、P—P;实验三证实了在评价分类任务中准确性判断与速度判断出现了实验性分离。使用这些定理可以解释所有的评价分类任务中的情绪启动现象。

关键词: 情绪启动, 比例效应, 反转效应, 词频效应, 定理