Abstract: Perspective Effects is a kind of phenomenon which refers to the four-card selection task, when participants to test condition rules from different perspectives, the selection bias steadily changes in the card P, -Q and -P, Q. Perspective Effects has an important impact on the four-card selection task, to the date, its existing conditions and mechanism remains controversial. The current study employed three experiments to investigate the effect of perspective effects on four-card selection task and mechanism remains. Experiment 1 asks the participants to go through different four-card selection tasks of social conversion model, causal model and abstract model in different perspectives respectively to explore the perspective effects. we used a 3 ( rule type : social conversion rule model, causal rule model, abstract rule model ) ×2 ( perspective : P&-Q condition, -P&Q condition) repeated measurement design. P&-Q condition that makes participants more inclined to choose P&-Q combination of experimental conditions. -P&Q condition similar to this. In Experiment 2, we use the same design with experiment1 to explore which condition is easier to generate the abstract rules of perspective effects in the providing of counter-examples and the necessary and sufficient conditions under rich contents of abstract rules. Experiment 3 further studies the essential relationship between the four-card selection task and the corresponding conditional reasoning. In conditions reasoning, we use a 2 (perspective) × 2 (rule type) × 4 (reasoning form) mixed design. Perspective between the variables were tested with two levels (P&-Q conditions; -P&Q conditions), there are two levels of rule types which is the conversion of social contractual and non-contractual society , each rule type has four the form of reasoning, namely modus ponens, affirming the consequent, modus tollens, denying the antecedent . Rule types and reasoning form are within-subject variable. Four-card selection task has the same design with experiment2. The results show that: (1) the results of experiment1 show that the perspective effects do exist in the social contract model and causal model while do not exist in the abstract model. (2) A rich supplying and counter-examples providing are more conducive to generate perspective effects. (3) There is a separation phenomenon between the four-card selection task and conditional reasoning problems, between four-card problem and the corresponding conditional reasoning don’t have a logical consistency. Compared with prior studies, the present paper has systematically studied the conditions which would affect perspective effects. at the same time, by comparing the four-card selection task with the corresponding conditional reasoning to study the nature of perspective effect mechanism. In conclusion, we find that provide counterexamples than necessary and sufficient conditions are more likely to have perspective effect, domain-general and domain-specific cannot fully explain perspective effect of four-card selection task, we suggest the existence of a mechanism to regulate the field of general or special mechanism.

Key words: four-card selection task,conditional reasoning,Perspective effects,domain general,domain specific

摘要： 视角效应（Perspective Effects)是指四卡问题解决中，当被试从不同的角度对条件规则进行检验时，其选择偏向会在卡片p,q和-p,q之间发生稳定变化的现象。视角效应对四卡问题的解决有重要影响，但其存在条件及发生机制仍存在争议。本研究通过3个实验，探讨四卡问题中的视角效应。实验1和实验2采用混合实验设计，结果表明，转换社会契约、因果规则的四卡问题中，发现视角效应，抽象规则中，提供反例比提供充分必要条件更有利于产生视角效应。实验3在不同视角下研究四卡问题与条件推理的关系。结果表明，四卡问题与条件推理之间不具有逻辑一致性，产生分离现象。总之，视角效应在抽象规则和特定规则中均存在，而提供反例更有利于视角效应的产生。领域一般性或领域特殊性均不能有效解释视角效应的产生，两种机制的综合调节为解释视角效应提供新的角度。

关键词: 四卡问题 视角效应 条件推理 领域一般性 领域特殊性