ISSN 1671-3710
CN 11-4766/R
主办:中国科学院心理研究所
出版:科学出版社

心理科学进展, 2018, 26(10): 1794-1806 doi: 10.3724/SP.J.1042.2018.01794

研究前沿

推理判断中双重加工过程的协作与转换机制

艾炎, 胡竹菁,

江西师范大学心理学院, 南昌 330022

The cooperation and transformation mechanism of dual processing in reasoning and judgment

AI Yan, HU Zhujing,

School of Psychology, Jiangxi Normal University, Nanchang 330022, China

通讯作者: 胡竹菁, E-mail:huzjing@jxnu.edu.cn

收稿日期: 2017-11-27   网络出版日期: 2018-10-15

基金资助: *国家自然科学基金项目(31460252)

Received: 2017-11-27   Online: 2018-10-15

摘要

推理判断中双重加工理论的发展经历了不同发展阶段, 早期主要对两个加工过程的定义及特征的关注, 当前转向对两者间的协作及转换机制的研究。本研究梳理了双重加工过程协作及转换机制的代表性模型及其相关实验支持证据, 综合归纳为以下三类模型:序列加工模型(Serial processing model)、平行竞争模型(Parallel competitive model)以及混合模型(Hybrid model), 并比较和论述了三类模型在两个加工过程的转换和协作机制、冲突探查的加工机制、偏差反应的解释机制上的异同, 以及三类模型各自面临的问题。

关键词: 双重加工; 序列加工模型; 平行竞争加工模型; 混合加工模型

Abstract

Theories about dual processing in reasoning and judgment have gone through different stages of development, and the focus on the definition and characteristics of the two processing processes in the early stage has turned to the study of the cooperation and transformation mechanism between the two at present. This study sorts out the representative models of the collaboration and transformation mechanism and its related experimental support evidence in the dual processing process, and summarizes the following three models: serial processing model, parallel competitive model and hybrid model. This study proceeds to compare and discuss the problems faced by the three models respectively as well as the differences and similarities between the three models in their interpretations of the transformation and cooperation mechanism between the two processing processes, the processing mechanism of conflict detection, and bias response.

Keywords: dual processing; serial processing model; parallel competitive model; hybrid model

PDF (580KB) 元数据 多维度评价 相关文章 导出 EndNote| Ris| Bibtex  收藏本文

本文引用格式

艾炎, 胡竹菁. 推理判断中双重加工过程的协作与转换机制. 心理科学进展, 2018, 26(10): 1794-1806 doi:10.3724/SP.J.1042.2018.01794

AI Yan, HU Zhujing. The cooperation and transformation mechanism of dual processing in reasoning and judgment. Advances in Psychological Science, 2018, 26(10): 1794-1806 doi:10.3724/SP.J.1042.2018.01794

1 引言

双重加工理论研究者认为人类认知存在两个不同的加工过程:类型1 (Type1, 简称T1)和类型2 (Type2, 简称T2)。T1是一个快速的、自动的、无意识的、不需要认知资源的启发式加工系统, 它是人类古老进化的产物, 人类与动物共有; T2则是一个慢速的、控制的、有意识的、需要认知资源的分析式加工系统, 是人类独有的(Evans, 1984, 2003, 2006, 2007, 2008, 2009, 2010; Evans, Venn, & Feeney, 2002)。

当前双重加工理论被广泛地应用于推理(Stanovich & West, 2000; Scherer, Yates, Baker, & Valentine, 2017)、决策(Barbey & Sloman, 2007; Kahneman, 2011; Klein, 2015)、社会认知(Chaiken & Trope, 1999; Gubbins & Byrne, 2014)及认知发展(Barrouillet, 2011; Barr, Pennycook, Stolz, & Fugelsang, 2015)、认知神经科学(Goel, 2007; Lieberman, 2007)、临床心理学(Beevers, 2005; Pyszczynski, Greenberg, & Solomon, 1999)和认知诊断(Toplak, West, & Stanovich, 2014)等领域。随着双重加工理论越来越流行, 这一理论本身存在的问题也逐渐显现出来, 比如众多双重加工理论的研究者们所提出的双重加工理论, 不论命名还是描述双重加工的过程特征都存在较大差异, 详见表1。即便是同一个研究者如Evans, 他在不同的阶段对其双重加工理论的定义和命名也不一样(胡竹菁, 胡笑羽, 2012)。另外, 双重加工过程的定义太过于模糊、两个系统特征群上缺乏一致性和相容性、支持双重加工理论的证据与单一系统解释一致、以及对两个系统的命名多样、不统一等受到了越来越多研究者们的批评 (Gigerenzer & Regier, 1996; Osman, 2004, 2013; Keren & Schul, 2009; Keren, 2013; Kruglanski & Gigerenzer, 2011; Kruglanski, 2013)。

表1   不同双重加工理论者的双重加工过程命名及特征

双重加工理论提出者T1T2
Posner & Snyder (1975)自动激活系统(automatic activation system)意识加工系统(conscious processing system)
Shiffrin & Schneider (1977)自动加工过程(automatic processing)控制加工过程(controlled processing)
Johnson-Laird (1983)内隐推理(implicit inferences)外显推理(explicit inferences)
Evans (1984, 1989)启发式过程(heuristic processing)分析式过程(analytic processing)
Pollock (1991)快速和灵活模块(quick and inflexible modules)智力(intellection)
Reber (1993)内隐认知 (implicit cognition)外显学习(explicit learning)
Epstein (1973, 1994)经验系统(experiential system)理性系统(rational system)
Levinson (1995)交互智力(interactional intelligence)分析智力(analytic intelligence)
Sloman (1996)联想系统(associative system)规则系统(rule-based system)
Evans & Over (1996)内隐思维过程(tacit thought processes)外显思维过程(explicit thought processes)
Hammond (1996)直觉认知(intuitive cognition)分析认知(analytical cognition)
Klein (1998)预先认知决策(recognition-primed decisions)理性选择策略(rational choice strategy)
Stanovich & West (2000)系统1 (System 1)系统2 (System 2)
De Neys (2006)自动启发式加工(Automatic-heuristic processing)执行分析式加工(executive-analytic processing)
Evans (2011)直觉加工(Intuitive processing)反思性加工(Reflective processing)
特征联想的(associative)以规则为基础的(rule-based)
整体性(holistic)分析性(analytic)
自动的(automatic)控制性的(controlled)
不需要认知能力(relatively undemanding of cognitive capacity)需要认知能力(demanding of cognitive capacity)
快速的(relatively fast)缓慢的(relatively slow)
通过生理、暴露和个体经验获得(acquisition
by biology, exposure, and personal experience)
通过文化和形式指导获得(acquisition by cultural and formal tuition)

资料主要来源: Stanovich和West (2000)

新窗口打开| 下载CSV


Evans和Stanovich (2013a)在对上述批评的回应中指出, 并不是所有的双重加工理论都是一样的, 它们对T1和T2加工内涵的定义大部分只是相关特征, 而不是核心特征。Evans认为区分两类加工过程的核心特征是是否需要工作记忆资源的参与。这一回应使得双重加工理论者们对于双重加工过程的实质是什么基本有了共识, 但关于这两个加工过程是如何运行以及如何协作的看法尚有争议, 有的研究者认为这两个加工过程依次启动, 序列加工; 有的研究者则认为它们并行启动, 平行加工。下文中对这两类加工转换和协作机制的相关研究进行梳理, 将它们归纳总结为以下三类模型:序列加工模型(Serial processing model)、平行竞争模型(Parallel competitive model)以及混合模型(Hybrid model), 并对这三类模型中较有代表性模型的主要特征及其相关的实验证据进行论述。

2 三类模型的主要观点及其实验证据

2.1 序列加工模型(Serial processing model)

2.1.1 序列加工模型的主要观点

序列加工模型认为T1与T2加工依次先后进行, 关于两者的加工顺序又衍生出两个不同的模型:分别是信念第一和推理第一模型(Stupple & Ball, 2008)。前者认为当个体在做出推理判断时, 首先启动的是T1加工, 启发式反应是默认反应。信念第一模型最典型的代表是Evans (2011), Evans和Stanovich (2013b)的默认干预模型(Default- interventionist model, 简称DI模型)。而推理第一模型仅是一个假设, 此模型在研究中极少涉及, 下面主要介绍信念第一模型的代表性模型Evans的DI模型。

虽然Evans和Overs (1996)在他们的书《理性与推理》(第七章P133)中指出, 他的双重加工理论即不偏爱序列模型也不偏爱冲突模型, 而是倾向于交互作用模型, 但是从Evans (2011), Evans和Stanovich (2013b)关于两个加工过程的论述来看——DI模型(见图1), 两个加工更符合序列加工的特征。这一模型假设:推理过程中直觉加工(Intuitive processing)先行被激活, 首先产生一个默认反应, 如果反思性加工(Reflective processing)不进行干预, 人们将最终做出默认反应; 若反思性加工加以干预, 基于直觉加工的默认反应就可能被否定掉, 并由基于反思性加工生成的反应所取代。Evans认为人类推理判断中偏差的产生是因为人们大部分的行为受直觉加工所控制。只有当反思性加工介入干预才能修改默认的直觉反应, 而反思性加工是否介入干预取决于任务要求、时间压力以及个人认知倾向等多种因素。例如当完成困难、新奇的任务, 受到动机激发需要工作记忆资源时, 反思性加工就会介入。

图1

图1   默认干预模型

注:A1为直觉加工的初始反应, A2为反思性加工的替代性反应。资料来源:Evans (2011)


DI模型在Evans (2006)对其推理的启发式(heuristic processing)与分析式(analytic processing)理论的扩展与校正中已见雏形, 见图2。Evans指出启发式加工往往会先产生一个默认的偏差反应, 分析式加工可能会介入对这一加工的修改, 但往往要取决于个体的认知能力和时间的可获得性等因素。他用相关性原则(relevance principle)来描述启发式加工是如何工作的, 当个体完成某项任务, 启发式首先启动, 将任务信息分为“相关”与“不相关”, 只有被认为是相关的信息才会进入下一步的分析式加工。他用满意原则(satisficing principle)来描述分析式加工的工作, 满意是无休无止的分析加工的结束, 决策或判断的产生是基于个体的满意感。此外, 关于两者是如何协作的, Evans指出启发式加工为分析式加工传送加工资料, 而分析式加工则对启发式加工认为是相关的信息进行加工。

图2

图2   启发式与分析式理论的扩展与校正

资料来源:Evans (2006)


启发式与分析式加工模型是Evans (1984)对其认知加工的两阶段模型的修订, 尝试进一步明确两者内部加工机制的实质。认知加工两阶段模型, 见图3, 这一模型认为当我们要加工某个信息时, 启发式加工会把信息自动地分为相关与不相关两类, 只有相关的信息才会进入分析式加工。推理判断中很多偏差的产生是因为启发式加工将逻辑相关信息忽略了。

图3

图3   认知加工的两阶段模型

资料来源:Evans (1984)


以上三个Evans所建构的模型:认知加工的两阶段模型、启发式与分析式理论的扩展和校正模型以及默认干预模型, 这三个模型一脉相传, 虽然Evans并不强调两个加工过程的序列性, 但是三个模型无疑都呈现了这一点, 即:T1先于T2加工, 先行产生默认反应, T2加工介入后产生替代反应。

2.1.2 DI模型的验证研究

有许多研究结果可以验证DI模型假设。Evans和Curtis-Holmes (2005)的实验通过控制反应时间, 让被试完成四类三段论推理任务:有效可信、有效不可信、无效可信以及无效不可信, 比较被试在有时间限制和无时间限制两种条件下的表现, 结果发现相比无时间限制组, 有时间限制组信念偏差反应增加, 逻辑反应减少。根据DI假设:T1加工是快速的, 不需要努力的, 且不需要工作记忆资源; 而T2加工则是慢速的, 需要工作记忆资源的。被试在大部分的推理和判断任务中, T1阶段首先快速的产生一个有偏差反应, 而限制反应时间抑制了分析式加工的参与, 使被试没能成功修改偏差反应。De Neys (2006) 让被试完成次级任务后再完成三段论推理任务, 通过控制次级任务来控制工作记忆的负荷程度, 结果发现以信念为基础的判断没有受到工作记忆负荷的影响, 但以逻辑为基础的判断却显著下降。这一研究验证了DI模型对工作记忆的假设:工作记忆负荷不会影响T1的加工, 但会干预T2的加工。Kokis, Macpherson, Toplak, West和Stanovich (2002)请5、6、8年级的学生完成归纳推理、演绎推理以及概率推理任务, 结果发现5/6年级(10~11岁)与8年级(13岁)两组儿童在偏好分析式加工上的表现有差异, 随着年龄的增大分析式反应的比率增大。这验证了DI模型的假设:T1是人类与动物共有的特征, 是大脑古老进化的部分; 而T2却是人类所独有的, 是新大脑的部分, T1不受年龄发展的影响, 但T2会受年龄发展的影响。Goel和Dolan (2003)让被试完成逻辑与信念有或无冲突的三段论推理任务, 利用事件相关功能的磁共振成像技术探查了信念偏差的神经学基础, 结果发现当被试抑制先验知识经验, 给出符合逻辑结构的反应时, 右侧前额叶皮层((RLPFC)被激活; 而当做出以先验知识经验为基础的反应时, 腹侧前额叶皮层(VMPFC)被激活, 也就是说T1与T2来自不同的大脑加工。Markovits, Forgues和Brune (2012), Markovits, Brunet, Thompson和Brisson (2013), Markovits, Brisson和de Chantal (2015), 以及Markovits, Brisson, de Chantal和Thompson (2017)的研究指出双重推理加工策略:数理策略(statistical strategies)和反例策略(counterexample strategies), 对应T1和T2不同的加工过程。

2.2 平行竞争模型(Parallel competition model)

2.2.1 平行竞争模型的主要观点

平行竞争模型(Parallel competitive model)在Stupple和Ball (2008)的研究中首次提出, 其思想源于Epstein (1994)的“经验系统” (experiential system)与“理性系统” (rational system)和Sloman (1996, 2014)的“联想系统” (associative system)和“规则系统” (rule-based system)。Epstein通过搜集大量的日常生活事例, 例如给情绪过度紧张的人提出建议:“控制下自己, 你太情绪化, 无法思考, 一旦冷静下来, 你将会有不同的看法”, 这类事例表明人类存在两种基本的信息加工模式——经验系统与理性系统, 经验系统具有直觉性、自动性、自然性、非语言性和叙述性等特征; 理性系统则具有分析性、协商性、语言性和理性等特征。Sloman研究的核心是两个系统的加工原则, 他的研究中指出联想系统是以相似性和接近性为原则得出推论, 而规则系统的加工原则是生产性(productivity)和系统性(systematicity )。

不同于Evans提出的DI模型的观点, Epstein和Sloman认为两个加工同时被激活, 正如Sloman (1996)所说“人们可能同时相信两个矛盾的反应。”例如Muller-Lyer错觉, 知觉告诉我们一根线比另一根线长, 但即使我们对两根线进行测量, 测量结果告诉我们它们是一样长, 我们的知觉依然会与逻辑不一致, 感觉其中一根更长。人们经历着“感觉到了两种不同反应间的冲突”, 偏差的产生在于我们相信了自己的直觉, 或是以经验系统作为判断标准来做出反应。

Handley, Newstead和Trippas (2011)发展了Epstein和Sloman关于两类认知加工同时启动并行运作, 相互竞争做出反应这一假设。他指出一般来说T1的加工要快于T2的加工, 更快得出反应的加工要等更慢的加工做出反应后才能做出正确判断。如果两个过程最终得出相同的结果, 个体很快就能做出判断; 但是如果两者给出了不同的结果, 个体在潜在冲突没有被解决前, 就无法做出判断。Handley和Trippas (2015)进一步发展其理论, 提出了他们的平行竞争模型, 见图4。这一模型认为T1阶段人们基于知识信念和基于问题逻辑结构的加工是同时激活的。在复杂命题上以知识信念为基础的判断快于基于问题逻辑结构的加工; 而在简单问题上, 直觉逻辑加工快于基于知识信念的反应。因此, 当人们要对复杂问题做出正确反应时, T2加工要抑制基于知识信念的反应直到产生基于问题逻辑结构的反应; 而在简单问题上, T2加工要抑制基于直觉逻辑的反应直到产生基于知识信念的反应。图4中纵向横线处是T1和T2的转换, 转换处隐含着对两个加工结果的冲突探查。

图4

图4   平行竞争模型

资料来源:Handley和Trippas (2015)


2.2.2 平行竞争模型的验证研究

平行竞争模型也得到了实验证据的支持, Stupple和Ball (2008)的研究中让被试完成逻辑与信念有冲突的三段论问题, 通过控制问题的复杂度, 测量其对三段论问题前提与结论的探查时间。实验结果支持了平行竞争模型的假设:即当这两个系统产生了不同的结果时, 冲突产生, 相比那些逻辑与信念不冲突的问题, 对冲突的解决导致加工时间的增加, 而且问题复杂性使得反应朝向信念而偏离逻辑。

Handley等(2011)的研究中通过控制指导语(以信念判断即根据结论的可信性或是以逻辑判断即根据结论的有效性为判断标准), 让被试完成条件推理的MP式任务, 验证DI模型的三个假设:

(1)信念判断比逻辑判断的反应更快;

(2)信念判断比逻辑判断的错误率更少;

(3)逻辑信念冲突对逻辑判断的影响大于信念判断。

其研究中五个实验的结果均验证了假设1, 但是假设2和假设3都没有得到验证, 却得到了与假设相反的结果。结果表明:信念判断的错误率比逻辑判断更多(实验1至实验5均是如此), 而且当逻辑信念冲突时, 对信念判断的影响, 无论是在正确率或是反应时上均比逻辑判断的影响大。

这一研究间接的证明了平行竞争模型的假设。DI模型不能解释被试在信念与逻辑冲突项目上的这些结果, 平行竞争模型却可以很好地解释。结论的可信性受逻辑推理干扰, 或是逻辑有效性受结论可信性干扰, 它们在一定程度上依赖于完成这两个过程所需要的时间。平行竞争模型认为两个过程都会得出一个结果, 而产生什么结果是以两个过程的加工时间来决定, 通常来说T1的加工会快于T2的加工, 但并不总是会这样, 有时T2的加工过程会快于T1的加工过程。谁先产生反应, 就会输出哪个反应。

我国学者姚志强和李亚非(2016)对Handley等的研究进行了验证, 通过考察被试在信念判断和逻辑判断两种指导语下解决不同难度和不同类型推理问题的正确率和反应时, 对DI模型与PC模型的预测进行了检验。实验结果表明, 解决简单问题时, 问题类型对信念判断有显著影响, 对逻辑判断无显著影响; 解决复杂问题时, 问题类型对逻辑判断的影响显著大于对信念判断的影响。实验结果支持平行竞争模型的假设, 即两类加工在推理过程中同时启动, 共同竞争最终的反应。此外, 还有研究(Trippas, Handley, Verde, & Morsany, 2016; Trippas, Thompson, & Handley, 2017)也对平行竞争模型进行了验证。

虽然平行竞争模型得到了一些实验支持, 但它在各个研究领域的应用并不广泛。因而对其的验证研究比较少, 这说明这个模型本身的适用性有限, 对人类推理、判断和决策中行为的解释度不高。

2.3 混合模型(Hybrid model)

混合模型既认同序列加工模型的观点, 认为T1和T2加工是序列启动, 也认同平行竞争模型中提出的两种加工同时激活, 相互竞争做出反应, 所以有些研究者如Handley也把混合模型等同于平行竞争模型。混合模型不同于序列加工和平行竞争模型的最大之处是对冲突探查过程的看法, 持混合模型观点的研究者们都把冲突探查过程看作为一个重要且独立的过程, 并且认为冲突探查是T2介入的关键因素, 这一观点也得到了相应研究的支持, 如Bhatia (2017)的研究表明冲突会增加T2加工的需要, 通过控制论据的强度和字频, 增加错误的启发式论据使得被试偏差反应减少, 问题解决的正确率提高。

混合模型有两个主要的模型, 分别是De Neys的逻辑直觉模型(logical intuition model)和Pennycook的分析式参与的三阶段双重加工模型(Three-stage dual-process model of analytic engagement)。

2.3.1 逻辑直觉模型(logical intuition model)

(1)逻辑直觉模型的主要观点

逻辑直觉模型的构想源于De Neys对冲突探查所做的一系列研究。De Neys和Glumicic (2008)的研究中指出人们在进行推理、判断以及决策时, 并不是纯序列或平行加工过程, 而是一个浅分析监控加工过程(shallow analytic monitoring process), 这一初始探查过程会激活一些仅需较少认知资源的一般规则(general principle), De Neys将“自动地掌握一般规则”定义为逻辑直觉(logical intuition), 这些逻辑直觉会自动地激发被试的逻辑直觉反应, 这一反应与直觉启发式反应同时出现。

De Neys (2012, 2014)正式提出逻辑直觉模型, 见图5。逻辑直觉模型认为在T1加工阶段人们会产生两类直觉加工:以语义和刻板关联(semantic and stereotypical associations)为基础的启发式加工以及以激活逻辑惯例和标准概率规则(traditional logical and probabilistic normative principles)为基础的逻辑直觉加工, 这两个加工平行竞争产生反应。当两者出现矛盾冲突时, 个体会有唤醒感, 正是这个唤醒使得个体对直觉启发式反应产生疑问感。但是人们并不能明确标识这一经验, 所以当人们对他们的直觉启发式反应抱有疑问感时, 却说不清是怎么回事, 也解释不了他们为什么会对他们的答案有疑问, 而且也不足以让他们在冲突情境中做出判断, 对冲突情境进行判断需要T2加工的介入。

图5

图5   逻辑直觉模型

资料来源:De Neys (2012)


(2)逻辑直觉模型验证研究

De Neys和Glumicic (2008)的研究中让被试完成一致、不一致以及中性的基础比率问题, 结果发现被试虽然在外显的出声思考任务中没有报告他们经历了冲突, 但是在回顾基础比率信息的任务上, 即使那些在出声思考中从没有提及基础比率、在不一致问题上总是出现偏差的被试也依然能辨认哪一组人更多(基础比率信息), 也就是说被试没能正确地完成推理任务, 但还是注意到了那些人格描述与基础比率是不一致的。此外, 在不一致问题上的反应时间、对基础比率的回顾时间都长于一致问题, 这些反应时的结果也支持冲突探查过程。De Neys, Cromheeke和Osman (2011)的研究中, 让被试完成一致和不一致的基础比率问题、合取问题, 结果表明冲突问题的反应时显著长于非冲突问题, 反应自信率显著低于非冲突问题, 这些冲突与非冲突问题之间唯一的不同是启发式反应与逻辑和概率标准是否有冲突, 如果被试并没有把逻辑概率纳入思考, 仅仅是一个纯粹的启发式思考者, 那这两者间就不应该存在显著的不同。但是反应时和反应自信率的结果表明被试成功地探查到了两者的冲突。De Neys, Rossi和Houdé (2013)的研究中让被试完成一致与不一致版本的棒球问题, 在棒球问题上, 被试的偏差反应可能源于被试将复杂问题替换成了简单问题, 即将“the bat cost more than”替换成了“the bat cost”。那被试是否注意到或感知到他们的这一替换过程呢?如果没有感知到这一过程, 那么在一致和不一致版本问题上的自信率应该不会存在差异, 但结果表明不一致问题上的反应自信率显著低于一致问题, 这一结果说明被试感知了替换, 只是被试并不能明确标识这一经验, 也不足以让他们在冲突情境中做出判断。De Neys, Vartanian和Goel (2008)用fMRI神经影像学的研究方法记录被试完成一致、不一致、中性和启发式四类基础比率问题时的脑区活动, 结果表明在完成冲突问题时, 调节冲突的区域:大脑前扣带回(Anterior cingulate cortex)有明显地激活。之后还有大量神经影像学的实验证据也支持De Neys对冲突探查的观点(Banks & Hope, 2014; Liang, Goel, Jia, & Li, 2014; Stollstorff, Vartanian, & Goel, 2012; Prado, Kaliuzhna, Cheylus, & Noveck, 2008)。

此外, De Neys (2012)还总结了支持冲突探查的众多证据, 如反应潜伏期的证据表明, 人们在完成推理判断时, 对传统标准很敏感, 在有效不可信结论上, 被试飞快扫过大、小前提, 并重新检查这些信息, 这种审查在无冲突的问题上很少见; 注视和眼动轨迹的研究在基础率问题上也出现了类似的情况, 当解决冲突问题时, 被试阅读完关于人格的描述后, 增加了回顾基础比率信息所在段落的倾向; 皮电研究表明, 在解决冲突问题时, 皮电阻被明显激活。

2.3.2 分析式参与的三阶段双重加工模型(Three-stage dual-process model of analytic engagement)

(1) 分析式参与的三阶段双重加工模型的主要观点

Pennycook在众多研究(Pennycook, Cheyne, Barr, Koehler, & Fugelsang, 2014; Pennycook, Cheyne, Seli, Koehler, & Fugelsang, 2012; Pennycook, Fugelsang, & Koehler, 2012; Pennycook & Thompson, 2012; Pennycook, Trippas, Handley, & Thompson, 2014)中发现冲突的探查并不总是有效的; 尽管被试明显增加了T2的加工, 但依然给出了刻板反应。Pennycook对此提出了一系列疑问:如是什么使得个体放弃更为直觉和自动化的加工, 进入有意的、费力的加工?如果冲突探查并不总是成功的, 又是什么决定了冲突探查的可能?Pennycook, Fugelsang和Koehler (2015), Pennycook (2017)在探索这些疑问的基础上, 提出了分析式参与的三阶段双重加工模型, 见图6

图6

图6   分析式参与的三阶段双重加工模型

注:1R, T1 的初始反应; 1R1最突出和流畅的直觉反应, 例如信念偏差反应; 1Rn初始阶段可能的、潜在的竞争反应; AR替代性反应。资料来源:Pennycook et al. (2015)


分析式参与的三阶段双重加工模型与De Neys的逻辑直觉模型一样都认为冲突探查对分析式加工的参与起着决定性作用, 且明确指出探查的冲突并不是指T1与T2间的冲突, 而是T1加工阶段输出的多种反应间的冲突。但不同于De Neys的观点, Pennycook认为冲突探查并不总是成功的, 即使T1加工过程输出了两种加工, 冲突探查也可能是失败的。且逻辑直觉模型完全聚焦于成功探查到冲突的过程, 却没有包含T2加工过程质量的不同。

分析式参与的三阶段双重加工模型认为人类的认知加工包括了T1的初始反应、冲突探查以及T2的加工判断三个阶段。这一模型结合了不同学者的观点(De Neys, 2012; Epstein, 1994; Sloman, 1996; Evans, 2006; Evans & Stanovich, 2013(b); Handley & Trippas, 2015), 将推理过程分成不同阶段和成分, 调和了目前两种主要的关于推理偏差产生的观点: 1) T2加工参与的失败; 2)冲突探查成功后进入T2加工, 但是逆转偏差反应失败。Pennycook建立了一个综合不同研究者的看法来评估偏差反应的模型, 既不否认De Neys冲突探查的存在, 也不否认Evans分析性加工介入的失败。下面详细介绍这一模型三个阶段的主要特点:

第一阶段, T1的初始反应阶段, 自动加工的T1依赖于加工信息的特征平行加工产生多种“直觉”反应(1Rn), 这些反应不需要工作记忆或是执行功能。加工信息的特征可能会使得T1加工同时产生的这些反应彼此间存在着潜在的冲突和矛盾, 且这些不同的反应中, 有的反应可能会比另一些反应更快、更流畅。比如在基础比率问题上, 人格描述诱发被试产生的刻板印象与比率信息引发的反应就可能存在矛盾与冲突, 且由于两者信息强度的不同, 刻板印象反应就比逻辑直觉反应更快、更流畅。

第二阶段, 是对初始阶段T1反应(1Rn)间的冲突进行探查, 如果没有探查到冲突(可能不存在冲突, 或是冲突探查失败), 初始阶段的反应(1R1)继续进入第三阶段, T2加工粗略地对此分析并接受这一反应。这一结果被认为是偏差产生的典型路径:畅通无阻且几乎不需要认知努力。但如果探查到了冲突, 那么T2的分析性加工就会介入。

第三阶段, 有两种不同的T2加工过程, 分别是合理化(rationalize)和解耦(decouple)。合理化是T2加工的一种形式, 尽管成功地探查到了冲突, 但是推理者并没有仔细思考由刺激所引发的初始阶段的加工结果, 就致力于去证明或是论证初始加工(1R1), 并做出直觉反应。这也可以解释在很多的研究中尽管被试明显增加了T2的加工, 但依然给出了刻板反应。这很可能是因为被试花费了许多时间合理化刻板反应, 或是权衡刻板印象与基础比率概率, 导致了“需要努力的信念偏差”。

T2加工的第二种加工形式是解耦, 这是最典型的分析性加工, 在过去的研究中占了主导地位。解耦是对直觉反应进行抑制, 并逆转和修改它, 产生替代性的反应。解耦过程有三种可能:1) 1R1被抑制, 由1R2替代; 2)1R1被抑制, 由1Rn替代; 3)另一个新的反应产生。

(2)分析式参与的三阶段双重加工模型的验证

Pennycook等(2015)在研究中让被试完成快速反应的一致与不一致的基础比率问题, 其结果验证了其模型的两个假设, 1)冲突探查与解耦是T2加工可分离的两个来源; 2)冲突探查有时是失败的。三阶段模型也得到其它实验的支持, 如Thompson和Morsanyi (2012), Thompson等(2013)以及Thompson和Johnson (2014)的研究中对易引发偏差反应的基础比率问题、逻辑与信念存在冲突的三段论问题、分母忽视问题等问题进行研究发现, 有些直觉反应比另一些直觉反应能更快进入人的心灵, 这与三阶段模型中T1阶段的多种“直觉”反应(1Rn)相对应。Frey, Johnson和De Neys (2018)的研究指出已有关于冲突探查的研究大都是从组间水平的分析得出的, 这样的分析只是对推理者反应的“平均”或是“典型”的描绘, 很容易给人留下冲突探查是绝对的, 且无处不在这样的错误印象。组间水平的分析不能确保是否所有的被试都呈现出这种效应(冲突探查效应), 其研究中让被试完成一致和不一致的基础比率问题和合取问题, 结果发现通过个体水平的分析表明大部分被试探查到了冲突, 但也有一小部分的被试对冲突不敏感。Scherer等(2017)的研究表明计算能力高的被试其在解决合取问题时的正确率更高, 被试对其错误反应并没有意识, 这两种反应表明个体在T2加工质量上存在差异。

3 三类模型对双重加工模型的贡献和主要存在的问题

3.1 三类模型对双重加工模型的主要贡献

(1)明确和完善了两类加工的转换和协作机制

T1与T2加工间的转换机制困扰着双重加工理论家许久。这三类模型进一步明确和完善了两类加工的转换和协作机制, 将双重加工理论的视角从对双重加工过程的定义及内涵的争议转向更深入地内部加工机制的研究。

图7所示, 三类加工模型对两类加工机制都有其看法和假设。序列加工模型认为T1与T2加工依次优后进行, 以Evans (2011)提出的默认干预模型为代表; 平行竞争模型则认为两个加工同时启动并行运作, 相互竞争做出反应, 以Handley和Trippas (2015)提出的平行竞争模型为代表。混合模型继承和发展了序列加工和平行竞争加工模型, 且进一步地完善和拓展了T1与T2加工间的转换机制。混合模型认为T1与T2并不是严格意义上的序列或是平行加工, 这一模型以De Neys的逻辑直觉模型和Pennycook的分析式参与的三阶段双重加工模型为代表, 这两个模型都强调了对不同种反应间的冲突探查过程是T2加工介入的关键因素。

图7

图7   序列加工、平行竞争以及混合模型的时间进程

注:序列加工模型与平行竞争模型的时间进程参考资料为De Neys (2012)


(2) 探索了冲突探查的加工机制

混合模型的研究者们还对冲突探查过程的加工机制进行了探究, 冲突探查过程究竟是由哪个加工过程完成的呢?是T1还是T2?Johnson, Tubau和De Neys (2016)的研究结果表明认知负荷不会影响到替换感知(在棒球问题上, 被试的偏差反应, 可能源于被试将复杂问题替换成了简单问题, 即将“the bat cost more than”替换成了“the bat cost”), 因而可以推断将复杂问题替换成简单问题这一过程是由T1加工完成的; Franssens和De Neys (2009)的研究也表明冲突探查过程不受认知负荷的影响, 这一过程仅需要很小的认知资源, 且其操作相当的自动化。Pennycook, Trippas等(2014)的研究结果表明基础比率的加工进入到了T1阶段的加工, 且信念反应与逻辑反应之间是互为影响, 抑制刻板印象与抑制基础比率概率一样困难。这些研究都表明冲突探查并不一定是由T2加工完成的, 只是De Neys认为这一探查过程是由T1加工完成, 而且Pennycook强调这一探查过程是独立于T1和T2间的第三过程。

(3)扩展了对偏差反应的解释机制

对两类加工机制的研究也发展了对偏差反应的解释机制。序列加工模型认为偏差的产生发生于早期的默认直觉加工, 只有进入后期的加工才能输出分析式反应。平行竞争加工模型则认为偏差的产生在于引发直觉反应的刺激太过于强烈, 使得直觉反应的速度快于分析式反应。而混合模型认为偏差的产生发生在对多种直觉反应的冲突探查过程后, 探查的失败或是对直觉反应抑制的失败导致了偏差反应。对偏差反应的解释机制可以归结为以下两种:冲突探查失败和抑制失败, 它们对冲突探查的观点在加工性质(松散、马虎vs完美、无瑕)与加工过程都不一样(系列加工vs平行加工)。

Evans和Kahneman认为人们对两类加工间的冲突探查是松散和马虎的(Lax), 人们有时能探查到这两个加工间的冲突, 分析式加工介入并修改启发式加工结果, 但是大多数情况下冲突探查是失败的, 人们根本意识不到他们的反应不符合逻辑标准, 正如Kahneman和Frederick (2005)说到的“那些做出一个随意直觉判断的人, 他们几乎不知道他们的判断是怎么得出的, 更不知道它的逻辑蕴涵”。人们在推理任务上偏差的产生是因为他们忽略了基础比率信息和合取规则, 没有探查到直觉反应与标准逻辑规则反应间的冲突, 偏差的产生源于冲突探查的失败。不同于Evans和Kahneman的观点, De Neys则认为人们对冲突的探查过程是完美、无瑕的(Flawless)。逻辑直觉模型里关于T1与T2协作的蓝图:T1加工同时产生两种直觉加工, 两者间的冲突探查并不需要T2加工的介入就可以完成, 偏差反应不是因为冲突探查的失败, 而是因为没能抑制住占优势的直觉。De Neys对另一问题, 即:冲突探查过程为进入T2的加工提供了一个线索, 但是为什么T1加工中已有线索提示标准逻辑答案, 推理者却依然出现偏差反应呢?De Neys指出正如我们知道抽烟有害身体健康, 却依然阻止不了我们继续抽烟一样, 冲突探查让人们对直觉反应有疑问, 但却不足以让人们相信逻辑反应完全可信。这种抑制的失败可能是因为被试的动机不足或是没有足够多的认知资源去完成这一抑制过程。

3.2 三类模型面临的问题

序列加工模型认为冲突探查并不是一个独立的成分, 而是T2的一个部分, 如果没有T2加工的参与, 那人们是如何探查出直觉加工与反思性加工间是否存在冲突呢?按照DI的假设, 首先启动T1的加工, 如果探查到冲突, 则启动T2的加工, 但问题是T1是如何探查到直觉反应与标准逻辑反应间的冲突呢?平行竞争模型面临的问题是:当直觉反应与标准逻辑反应间不冲突时, 耗时而又费力的有意加工就显得多余, 这有悖于人类认知加工的经济、快捷原则。虽然混合模型能很好的解释这两者的缺陷, 但是混合模型将加工过程不断地细分, 试图整合目前两种对偏差反应的视角, 这固然能让我们更细致的了解推理判断的内部加工机制, 可无形中也将人类的认知加工越来越复杂化, 使得模型本身的验证变得困难。

此外, DI模型虽然对两个加工过程的协作给出了较为明确的描述:T1加工产生默认反应, T2加工可能介入并修改T1加工的默认反应, 也可能介入失败, 维持T1的默认反应。那究竟是什么因素决定了T2的介入呢?而T2成功介入为何偏差反应依然存在?序列加工模型并没有给出答案, 虽然平行竞争模型与混合模型从不同的角度回答了这些问题。但平行竞争模型绕开了关于T1和T2加工间的协作, 认为T1和T2加工是同时激活并共同竞争产生反应, 但两个加工如何输出反应, 其详细机制阐述不清晰。混合模型通过对T2加工质量进行区分来解释这一问题, 虽然很合理, 但这无疑会使得这一模型越来越庞大, 也会加剧对此模型验证的难度。

以上三个模型均是对人类推理判断的内部加工机制的假设, 虽然每个模型都有其局限性, 但每个模型都能在一定程度上解释人类的推理判断行为。这些关于推理判断内部加工机制的探索, 为我们展开了人类认知加工的视角, 也为我们揭开了人类思维浩瀚宇宙的面纱。正如Pennycook (2017)文中所说, 三阶段模型可能不一定正确的, 但也可能是正确且有用的, 因为它能帮助我们去了解“人类究竟是怎么样思考”。

参考文献

胡竹菁, 胡笑羽 . ( 2012).

Evans双重加工理论的发展过程简要述评

心理学探新, 32( 4), 310-316.

DOI:10.3969/j.issn.1003-5184.2012.04.004      URL     [本文引用: 1]

英国心理学家Evans在对他所提出的有关人类思维的"双重加工理论"进行30多年的研究过程中,其理论的名称和内涵等方面都有个不断发展的过程。就该理论的名称而言,通常称为双重加工理论,同时他还提出了双因素理论。主要与"双重加工理论"这一名称相联系的几组概念包括:类型1—类型2,启发式—分析式,不可明言加工—外显思维加工,系统1—系统2;主要与"双因素理论"这一名称相关联的两组概念是:理性1—理性2,逻辑加工—非逻辑加工。最近,他又以"两种心灵假设"为名概括他的理论内容。该文对双重加工理论在名称及相应的内涵的历史发展过程作了较为详尽的评述。

姚志强, 李亚非 . ( 2016).

逻辑-信念冲突与推理难度对逻辑和信念判断的影响

心理科学, 39( 1), 36-42.

[本文引用: 1]

Banks A. P., & Hope, C. ( 2014).

Heuristic and analytic processes in reasoning: An event-related potential study of belief bias

Psychophysiology, 51( 3), 290-297.

DOI:10.1111/psyp.12169      URL     PMID:25003167      [本文引用: 1]

Abstract Human reasoning involves both heuristic and analytic processes. This study of belief bias in relational reasoning investigated whether the two processes occur serially or in parallel. Participants evaluated the validity of problems in which the conclusions were either logically valid or invalid and either believable or unbelievable. Problems in which the conclusions presented a conflict between the logically valid response and the believable response elicited a more positive P3 than problems in which there was no conflict. This shows that P3 is influenced by the interaction of belief and logic rather than either of these factors on its own. These findings indicate that belief and logic influence reasoning at the same time, supporting models in which belief-based and logical evaluations occur in parallel but not theories in which belief-based heuristic evaluations precede logical analysis.

Barbey A. K., & Sloman, S. A . ( 2007).

Base-rate respect: From ecological rationality to dual processes

Behavioural and Brain Sciences, 30( 3), 241-254.

[本文引用: 1]

Barr N., Pennycook G., Stolz J. A., & Fugelsang J. A . ( 2015).

Reasoned connections: A dual-process perspective on creative thought

Thinking & Reasoning, 21( 1), 61-75.

DOI:10.1080/13546783.2014.895915      URL     [本文引用: 1]

A divide exists in the creativity literature as to whether relatively more or less executive processing is beneficial to creative thinking. To explore this issue, we employ an individual differences perspective informed by dual-process theories (DPTs) in which it is assumed that people vary in the extent to which they rely on autonomous (Type 1) or controlled processing (Type 2). We find that those more willing and/or able to engage Type 2 processing are more likely to successfully make creative connections in tasks requiring the unification of disparate elements and the novelty of generated items, but not in some other indices of creativity, namely, cognitive flexibility and fluency. Implications for the role of executive processing in creative thinking are discussed in the context of DPTs. We situate the ability to make remote connections alongside other advanced higher order thinking capabilities that are unique to humans.

Barrouillet P., ( 2011).

Dual-process theories and cognitive development: Advances and challenges

Developmental Review, 31( 2-3), 79-85.

DOI:10.1016/j.dr.2011.07.002      URL     [本文引用: 1]

Dual-process theories have gained increasing importance in psychology. The contrast that they describe between an old intuitive and a new deliberative mind seems to make these theories especially suited to account for development. Accordingly, this special issue aims at presenting the latest applications of dual-process theories to cognitive development. Dual-process theoreticians mainly focused on adult functioning have been invited to elaborate the developmental implications of their theory, while developmental psychologists who have proposed dual-process theories of development in the domain of thinking and reasoning present the most recent formulation of their theory and review supporting empirical evidence.

Beevers, C. G . ( 2005).

Cognitive vulnerability to depression: A dual process model

Clinical Psychology Review, 25( 7), 975-1002.

DOI:10.1016/j.cpr.2005.03.003      URL     [本文引用: 1]

Bhatia S., ( 2017).

Conflict and bias in heuristic judgment

Journal of Experimental Psychology: Learning, Memory, and Cognition, 43( 2), 319-325.

DOI:10.1037/xlm0000307      URL     PMID:27685023      [本文引用: 1]

Conflict has been hypothesized to play a key role in recruiting deliberative processing in reasoning and judgment tasks. This claim suggests that changing the task so as to add incorrect heuristic responses that conflict with existing heuristic responses can make individuals less likely to respond heuristically and can increase response accuracy. We tested this prediction in experiments involving judgments of argument strength and word frequency, and found that participants are more likely to avoid heuristic bias and respond correctly in settings with 2 incorrect heuristic response options compared with similar settings with only 1 heuristic response option. Our results provide strong evidence for conflict as a mechanism influencing the interaction between heuristic and deliberative thought, and illustrate how accuracy can be increased through simple changes to the response sets offered to participants. (PsycINFO Database Record(c) 2016 APA, all rights reserved).

Chaiken S. , & Trope, Y.( 1999) . Dual-process theories in social psychology. New York: Guilford Press.

[本文引用: 1]

De Neys W. , ( 2006).

Automatic-heuristic and executive- analytic processing during reasoning: Chronometric and dual-task considerations

Quarterly Journal of Experimental Psychology, 59( 6), 1070-1100.

DOI:10.1080/02724980543000123      URL     [本文引用: 2]

De Neys W. , ( 2012).

Bias and conflict: A case for logical intuitions

Perspectives on Psychological Science, 7( 1), 28-38.

DOI:10.1177/1745691611429354      URL     [本文引用: 5]

De Neys W. , ( 2014).

Conflict detection, dual processes, and logical intuitions: Some clarifications

Thinking & Reasoning, 20( 2), 169-187.

[本文引用: 1]

De Neys W., Cromheeke S., & Osman M . ( 2011).

Biased but in doubt: Conflict and decision confidence

PLoS One, 6( 1), e15954.

DOI:10.1371/journal.pone.0015954      URL     PMID:21283574      [本文引用: 1]

Human reasoning is often biased by intuitive heuristics. A central question is whether the bias results from a failure to detect that the intuitions conflict with traditional normative considerations or from a failure to discard the tempting intuitions. The present study addressed this unresolved debate by using people's decision confidence as a nonverbal index of conflict detection. Participants were asked to indicate how confident they were after solving classic base-rate (Experiment 1) and conjunction fallacy (Experiment 2) problems in which a cued intuitive response could be inconsistent or consistent with the traditional correct response. Results indicated that reasoners showed a clear confidence decrease when they gave an intuitive response that conflicted with the normative response. Contrary to popular belief, this establishes that people seem to acknowledge that their intuitive answers are not fully warranted. Experiment 3 established that younger reasoners did not yet show the confidence decrease, which points to the role of improved bias awareness in our reasoning development. Implications for the long standing debate on human rationality are discussed.

De Neys W., , & Glumicic, T. ( 2008).

Conflict monitoring in dual process theories of thinking

Cognition, 106( 3), 1248-1299.

DOI:10.1016/j.cognition.2007.06.002      URL     PMID:17631876      [本文引用: 2]

Popular dual process theories have characterized human thinking as an interplay between an intuitive-heuristic and demanding-analytic reasoning process. Although monitoring the output of the two systems for conflict is crucial to avoid decision making errors there are some widely different views on the efficiency of the process. Kahneman [Kahneman, D. (2002). Maps of bounded rationality: A perspective on intuitive judgement and choice . Nobel Prize Lecture. Retrieved January 11, 2006, from: http://nobelprize.org/nobel_prizes/economics/laureates/2002/kahnemann-lecture.pdf ] and Evans [Evans, J. St. B. T. (1984). Heuristic and analytic processing in reasoning. British Journal of Psychology, 75 , 451–468], for example, claim that the monitoring of the heuristic system is typically quite lax whereas others such as Sloman [Sloman, S. A. (1996). The empirical case for two systems of reasoning. Psychological Bulletin, 119 , 3–22] and Epstein [Epstein, S. (1994). Integration of the cognitive and psychodynamic unconscious. American Psychologists, 49 , 709–724] claim it is flawless and people typically experience a struggle between what they “know” and “feel” in case of a conflict. The present study contrasted these views. Participants solved classic base rate neglect problems while thinking aloud. In these problems a stereotypical description cues a response that conflicts with the response based on the analytic base rate information. Verbal protocols showed no direct evidence for an explicitly experienced conflict. As Kahneman and Evans predicted, participants hardly ever mentioned the base rates and seemed to base their judgment exclusively on heuristic reasoning. However, more implicit measures of conflict detection such as participants’ retrieval of the base rate information in an unannounced recall test, decision making latencies, and the tendency to review the base rates indicated that the base rates had been thoroughly processed. On control problems where base rates and description did not conflict this was not the case. Results suggest that whereas the popular characterization of conflict detection as an actively experienced struggle can be questioned there is nevertheless evidence for Sloman’s and Epstein’s basic claim about the flawless operation of the monitoring. Whenever the base rates and description disagree people will detect this conflict and consequently redirect attention towards a deeper processing of the base rates. Implications for the dual process framework and the rationality debate are discussed.

De Neys W., Rossi S., & Houdé O . ( 2013).

Bats, balls, and substitution sensitivity: Cognitive misers are no happy fools

Psychonomic Bulletin & Review, 20( 2), 269-273.

DOI:10.3758/s13423-013-0384-5      URL     PMID:23417270      [本文引用: 1]

Influential work on human thinking suggests that our judgment is often biased because we minimize cognitive effort and intuitively substitute hard questions by easier ones. A key question is whether or not people realize that they are doing this and notice their mistake. Here, we test this claim with one of the most publicized examples of the substitution bias, the bat-and-ball problem. We designed an isomorphic control version in which reasoners experience no intuitive pull to substitute. Results show that people are less confident in their substituted, erroneous bat-and-ball answer than in their answer on the control version that does not give rise to the substitution. Contrary to popular belief, this basic finding indicates that biased reasoners are not completely oblivious to the substitution and sense that their answer is questionable. This calls into question the characterization of the human reasoner as a happy fool who blindly answers erroneous questions without realizing it.

De Neys W., Vartanian O., & Goel V . ( 2008).

Smarter than we think: When our brains detect that we are biased

Psychological Science, 19( 5), 483-489.

DOI:10.1111/j.1467-9280.2008.02113.x      URL     [本文引用: 1]

Epstein S., ( 1973).

The self-concept revisited or a theory of a theory

. American Psychologist 28( 5), 404-416.

DOI:10.1037/h0034679      URL     PMID:4703058     

Presents a theory which attempts to incorporate phenomenological views on the self-concept within an objective framework: the self-concept is identified as a self-theory. (Author/JM)

Epstein S., ( 1994).

Integration of the cognitive and the psychodynamic unconscious

American Psychologist, 49( 8), 709-724.

DOI:10.1037/0003-066X.49.8.709      URL     [本文引用: 2]

Evans, J. St. B. T . ( 1984).

Heuristic and analytic processes in reasoning

British Journal of Psychology, 75( 4), 451-468.

DOI:10.1111/j.2044-8295.1984.tb01915.x      URL     [本文引用: 3]

A general two-stage theory of human inference is proposed. A distinction is drawn between heuristic processes which select items of task information as ‘relevant’, and analytic processes which operate on the selected items to generate inferences or judgements.These two stages are illustrated in a selective review of work on both deductive and statistical reasoning. Factors identified as contributing to heuristic selection include perceptual salience, linguistic suppositions and semantic associations. Analytic processes are considered to be context dependent: people reason from experience, not from inference rules. The paper includes discussion of the theory in comparison with other contemporary theories of human inference, and in relation to the current debate about human rationality.

Evans, J. St. B. T . ( 1989).

Bias in human reasoning: Causes and consequences.

Lawrence Erlbaum.

Evans, J. St. B. T . ( 2003).

In two minds: Dual-process accounts of reasoning

Trends in Cognitive Sciences, 7( 10), 454-459.

DOI:10.1016/j.tics.2003.08.012      URL     PMID:14550493      [本文引用: 1]

Abstract Researchers in thinking and reasoning have proposed recently that there are two distinct cognitive systems underlying reasoning. System 1 is old in evolutionary terms and shared with other animals: it comprises a set of autonomous subsystems that include both innate input modules and domain-specific knowledge acquired by a domain-general learning mechanism. System 2 is evolutionarily recent and distinctively human: it permits abstract reasoning and hypothetical thinking, but is constrained by working memory capacity and correlated with measures of general intelligence. These theories essentially posit two minds in one brain with a range of experimental psychological evidence showing that the two systems compete for control of our inferences and actions.

Evans, J. St. B. T . ( 2006).

The heuristic-analytic theory of reasoning: Extension and evaluation

Psychonomic Bulletin & Review, 13( 3), 378-395.

DOI:10.3758/BF03193858      URL     PMID:17048720      [本文引用: 4]

An extensively revised heuristic-analytic theory of reasoning is presented incorporating three principles of hypothetical thinking. The theory assumes that reasoning and judgment are facilitated by the formation of epistemic mental models that are generated one at a time (singularity principle) by preconscious heuristic processes that contextualize problems in such a way as to maximize relevance to current goals (relevance principle). Analytic processes evaluate these models but tend to accept them unless there is good reason to reject them (satisficing principle). At a minimum, analytic processing of models is required so as to generate inferences or judgments relevant to the task instructions, but more active intervention may result in modification or replacement of default models generated by the heuristic system. Evidence for this theory is provided by a review of a wide range of literature on thinking and reasoning.

Evans, J. St. B. T . ( 2007).

On the resolution of conflict in dual process theories of reasoning

Thinking & Reasoning, 13( 4), 321-339.

DOI:10.1080/13546780601008825      URL     [本文引用: 1]

In this paper, I show that the question of how dual process theories of reasoning and judgement account for conflict between System 1 (heuristic) and System 2 (analytic) processes needs to be explicated and addressed in future research work. I demonstrate that a simple additive probability model that describes such conflict can be mapped on to three different cognitive models. The pre-emptive conflict resolution model assumes that a decision is made at the outset as to whether a heuristic or analytic process will control the response. The parallel-competitive model assumes that each system operates in parallel to deliver a putative response, resulting sometimes in conflict that then needs to be resolved. Finally, the default-interventionist model involves the cueing of default responses by the heuristic system that may or may not be altered by subsequent intervention of the analytic system. A second, independent issue also emerges from this discussion. The superior performance of higher-ability participants on reasoning tasks may be due to the fact that they engage in more analytic reasoning (quantity hypothesis) or alternatively to the fact that the analytic reasoning they apply is more effective (quality hypothesis).

Evans, J. St. B. T . ( 2008).

Dual-processing accounts of reasoning, judgment, and social cognition

Annual Review of Psychology, 59, 255-278.

DOI:10.1146/annurev.psych.59.103006.093629      URL     PMID:18154502      [本文引用: 1]

Abstract This article reviews a diverse set of proposals for dual processing in higher cognition within largely disconnected literatures in cognitive and social psychology. All these theories have in common the distinction between cognitive processes that are fast, automatic, and unconscious and those that are slow, deliberative, and conscious. A number of authors have recently suggested that there may be two architecturally (and evolutionarily) distinct cognitive systems underlying these dual-process accounts. However, it emerges that (a) there are multiple kinds of implicit processes described by different theorists and (b) not all of the proposed attributes of the two kinds of processing can be sensibly mapped on to two systems as currently conceived. It is suggested that while some dual-process theories are concerned with parallel competing processes involving explicit and implicit knowledge systems, others are concerned with the influence of preconscious processes that contextualize and shape deliberative reasoning and decision-making.

Evans, J. St B. T.. ( 2009) .

How many dual-process theories do we need? One, two, or many?

In J. St. B. T. Evans & K. Frankish (Eds.), In two minds: Dual processes and beyond (pp. 33-54). Oxford: Oxford University Press.

DOI:10.1093/acprof:oso/9780199230167.003.0002      URL     [本文引用: 1]

Abstract Dual-process theories of cognition are to be found everywhere in psychology although the literatures concerned may contain little or no cross referencing to each other. These theories come under many labels, but at least superficially all seem to be making a similar distinction (see Evans 2008; and Frankish and Evans, this volume). One question addressed in this chapter is that of whether we need to have this great multiplicity of theories, or whether there is one grand unifying dual-process theory that can incorporate them all. The literature already contains dual-system theories which purport to integrate many if not all accounts in this way (see Evans 2003; Stanovich, this volume; Evans and Over 1996; Smith and DeCoster 2000; Stanovich 1999) and one objective is here is to assess the adequacy of such accounts. However, I shall argue that such theories fall into two distinct groups from the viewpoint of the cognitive architecture they imply. There is also a third notion (cognitive styles) which can all too readily be confused with such two-process accounts. In previous reviews of this topic (e.g. Evans 2003), I have followed the fashion started by Stanovich (1999) for referring to System 1 and System 2 processes. I shall not do so in this chapter, except when referring specifically to dual-system theories. Instead I will talk of type 1 and 2 processes, a terminology first used over 30 years ago in the literature on reasoning (Wason and Evans 1975). (PsycINFO Database Record (c) 2012 APA, all rights reserved)

Evans, J. St. B. T . ( 2010).

Intuition and reasoning: A dual- process perspective

Psychological Inquiry, 21( 4), 313-326.

DOI:10.1080/1047840X.2010.521057      URL     [本文引用: 1]

A lay definition of intuition holds that it involves immediate apprehension in the absence of reasoning. From a more technical point of view, I argue also that intuition should be seen as the contrastive of reasoning, corresponding roughly to the distinction between Type 1 (intuitive) and Type 2 (reflective) processes in contemporary dual process theories of thinking. From this perspective, we already know a great deal about intuition: It is quick, provides feelings of confidence, can reflect large amounts of information processing, and is most likely to provide accurate judgments when based on relevant experiential learning. Unlike reasoning, intuition is low effort and does not compete for central working memory resources. It provides default responses which may—or often may not—be intervened upon with high effort, reflective reasoning. Intuition has, however, been blamed for a range of cognitive biases in the psychological literatures on reasoning and decision making. The evidence indicates that with novel and abstract problems, not easily linked to previous experience, intervention with effortful reasoning is often required to avoid such biases. Hence, although it seems that intuition dominates reasoning most of the time—both in the laboratory and the real world—it can indeed be a false friend.

Evans, J. St. B. T . ( 2011).

Dual-process theories of reasoning: Contemporary issues and developmental applications

Developmental Review, 31( 2-3), 86-102.

DOI:10.1016/j.dr.2011.07.007      URL     [本文引用: 4]

In this paper, I discuss the current state of theorising about dual processes in adult performance on reasoning and decision making tasks, in which Type 1 intuitive processing is distinguished from Type 2 reflective thinking. I show that there are many types of theory some of which distinguish modes rather than types of thinking and that assumptions about underlying cognitive architecture vary. I show that some dual-system theories have been replaced recently by the idea that we have two or more distinct inds with different evolutionary histories. I also present the most recent formulation of my own account of dual processing within hypothetical thinking theory, at a level more easily applied to performance on specific tasks. I then consider implications for cognitive development, pointing out that while Type 2 thinking is clearly linked to the development of cognitive ability, it combines and competes with multiple Type 1 processing systems which persist in adult cognition, each of which could have their own developmental time course. Hence, while dual-process theories can and should inspire much research and theory in cognitive development, the derivation of predictions for cognitive development is far from straightforward.

Evans J. St. B. T., & Curtis-Holmes, J. ( 2005).

Rapid responding increases belief bias: Evidence for the dual- process theory of reasoning

Thinking & Reasoning, 11( 4), 382-389.

DOI:10.1080/13546780542000005      URL     [本文引用: 1]

In this study, we examine the belief bias effect in syllogistic reasoning under both standard presentation and in a condition where participants are required to respond within 10 seconds. As predicted, the requirement for rapid responding increased the amount of belief bias observed on the task and reduced the number of logically correct decisions, both effects being substantial and statistically significant. These findings were predicted by the dual-process account of reasoning, which posits that fast heuristic processes, responsible for belief bias, compete with slower analytic processes that can lead to correct logical decisions. Requiring rapid responding thus differentially inhibits the operation of analytic reasoning processes, leading to the results observed.

Evans, J. St. B. T ., & Over, D. E.( 1996).

Rationality and reasoning

Hove, England: Psychology Press.

[本文引用: 1]

Evans J. St. B. T., Venn S., & Feeney A . ( 2002).

Implicit and explicit processes in a hypothesis testing task

British Journal of Psychology, 93, 31-46.

DOI:10.1348/000712602162436      URL     PMID:11839100      [本文引用: 1]

We present the results of two experiments investigating the factors that determine responding on the pseudo-diagnosticity task. In Expt 1 we manipulated people's beliefs about the degree to which an initial piece of evidence supported a focal hypothesis and found decreased pseudo-diagnostic (PD) responding when the evidence offered low support for the focal hypothesis. In Expt 2 we manipulated the instructions given to participants. We found that instructions to select evidence to help decide between the focal and the complementary hypotheses produced fewer PD responses than both instructions to decide whether the focal hypothesis was the case and instructions to decide whether its complement was the case. The results are interpreted within the framework of recent dual process theories of reasoning.

Evans J. St. B. T., & Stanovich, K. E . ( 2013

a). Dual-process theories of higher cognition

Perspectives on Psychological Science, 8( 3), 223-241.

DOI:10.1177/1745691612460685      URL     [本文引用: 1]

Evans J. St. B. T., & Stanovich, K. E . ( 2013

b). Theory and metatheory in the study of dual processing: Reply to comments

Perspectives on Psychological Science, 8( 3), 263-271.

DOI:10.1177/1745691613483774      URL     [本文引用: 3]

Franssens S., & De Neys, W. (2009).

The effortless nature of conflict detection during thinking

Thinking & Reasoning, 15( 2), 105-128.

DOI:10.1080/13546780802711185      URL     [本文引用: 1]

Dual process theories conceive human thinking as an interplay between heuristic processes that operate automatically and analytic processes that demand cognitive effort. The interaction between these two types of processes is poorly understood. De Neys and Glumicic (2008) recently found that most of the time heuristic processes are successfully monitored. This monitoring, however, would not demand as many cognitive resources as the analytic thinking that is needed to solve reasoning problems. In the present study we tested the crucial assumption about the effortless nature of the monitoring process directly. Participants solved base-rate neglect problems in which heuristic and analytic processes cued a conflicting response or not. Half of the participants reasoned under a secondary task load. A surprise recall task was used as an implicit measure of whether the participants detected the conflict in the problems. Results showed that, even under load, base-rate recall performance was better for conflict problems than for no-conflict problems. Although participants made more reasoning errors under load, recall of the conflict problems was not affected by the working memory load. These findings support the claim about the successful and undemanding nature of the conflict detection process during thinking.

Frey D., Johnson E. D., & De Neys W . ( 2018).

Individual differences in conflict detection during reasoning

The Quarterly Journal of Experimental Psychology, 71( 5), 1188-1208.

DOI:10.1080/17470218.2017.1313283      URL     [本文引用: 1]

Abstract Decades of reasoning and decision-making research have established that human judgment is often biased by intuitive heuristics. Recent "error" or bias detection studies have focused on reasoners' abilities to detect whether their heuristic answer conflicts with logical or probabilistic principles. A key open question is whether there are individual differences in this bias detection efficiency. Here we present three studies in which co-registration of different error detection measures (confidence, response time, and confidence response time) allowed us to assess bias detection sensitivity at the individual participant level in a range of reasoning tasks. Results indicate that although most individuals show robust bias detection, as indexed by increased latencies and decreased confidence, there is a subgroup of reasoners who consistently fails to do so. We discuss theoretical and practical implications for the field.

Gigerenzer G., & Regier, T. ( 1996).

How do we tell an association from a rule? Comment on Sloman (1996)

Psychological Bulletin, 119( 1), 23-26.

DOI:10.1037/0033-2909.119.1.23      URL     [本文引用: 1]

ABSTRACT S. A. Sloman's (see PA, Vol 83:21925) intriguing argument for separate associative and rule-based reasoning systems is unfortunately damaged by a certain amount of slack in the distinction he makes between these two posited mental mechanisms. The authors suggest that the distinction could be sharpened by overt reference to explicit models of associative and rule-based processing. They also point out that "simultaneous contradictory belief," which Sloman takes as evidence for separate associative and rule-based systems, need not be interpreted in this fashion. It may also signal a number of other things, including the presence of linguistic ambiguity (as in the Linda problem), competing lines of formal reasoning (as in the Wason selection task), and unclarified assumptions (as in the 3 doors problem). (PsycINFO Database Record (c) 2012 APA, all rights reserved)

Goel V., ( 2007).

Anatomy of deductive reasoning

Trends in Cognitive Sciences, 11( 10), 435-441.

DOI:10.1016/j.tics.2007.09.003      URL     [本文引用: 1]

Goel V., & Dolan, R. J . ( 2003).

Explaining modulation of reasoning by belief

Cognition, 87( 1), B11-B22.

DOI:10.1016/S0010-0277(02)00185-3      URL     PMID:12499108      [本文引用: 1]

Abstract Although deductive reasoning is a closed system, one's beliefs about the world can influence validity judgements. To understand the associated functional neuroanatomy of this belief-bias we studied 14 volunteers using event-related fMRI, as they performed reasoning tasks under neutral, facilitatory and inhibitory belief conditions. We found evidence for the engagement of a left temporal lobe system during belief-based reasoning and a bilateral parietal lobe system during belief-neutral reasoning. Activation of right lateral prefrontal cortex was evident when subjects inhibited a prepotent response associated with belief-bias and correctly completed a logical task, a finding consistent with its putative role in cognitive monitoring. By contrast, when logical reasoning was overcome by belief-bias, there was engagement of ventral medial prefrontal cortex, a region implicated in affective processing. This latter involvement suggests that belief-bias effects in reasoning may be mediated through an influence of emotional processes on reasoning. Copyright 2002 Elsevier Science B.V.

Gubbins E., & Byrne, R. M. J . ( 2014).

Dual processes of emotion and reason in judgments about moral dilemmas

Thinking & Reasoning, 20( 2), 245-268.

DOI:10.1080/13546783.2013.877400      URL     [本文引用: 1]

We report the results of two experiments that show that participants rely on both emotion and reason in moral judgments. Experiment 1 showed that when participants were primed to communicate feelings, they provided emotive justifications not only for personal dilemmas, e.g., pushing a man from a bridge that will result in his death but save the lives of five others, but also for impersonal dilemmas, e.g., hitting a switch on a runaway train that will result in the death of one man but save the lives of five others; when they were primed to communicate thoughts, they provided non-emotive justifications for both personal and impersonal dilemmas. Experiment 2 showed that participants read about a protagonist's emotions more quickly when the protagonist was faced with a personal dilemma than an impersonal one, but they read about the protagonist's decision to act or not act equally quickly for personal and impersonal dilemmas.

Hammond K. R. (1996). Human judgment and social policy. Oxford University Press..

Handley S. J., Newstead S. E., & Trippas D . ( 2011).

Logic, beliefs, and instruction: A test of the default interventionist account of belief bias

Journal of Experimental Psychology: Learning, Memory, and Cognition, 37( 1), 28-43.

DOI:10.1037/a0021098      URL     PMID:21058879      [本文引用: 2]

According to dual-process accounts of thinking, belief-based responses on reasoning tasks are generated as default but can be intervened upon in favor of logical responding, given sufficient time, effort, or cognitive resource. In this article, we present the results of 5 experiments in which participants were instructed to evaluate the conclusions of logical arguments on the basis of either their logical validity or their believability. Contrary to the predictions arising from these accounts, the logical status of the presented conclusion had a greater impact on judgments concerning its believability than did the believability of the conclusion on judgments about whether it followed logically. This finding was observed when instructional set was presented as a between-participants factor (Experiment 1), when instruction was indicated prior to problem presentation by a cue (Experiment 2), and when the cue appeared simultaneously with conclusion presentation (Experiments 3 and 4). The finding also extended to a range of simple and more complex argument forms (Experiment 5). In these latter experiments, belief-based judgments took significantly longer than those made under logical instructions. We discuss the implications of these findings for default interventionist accounts of belief bias.

Handley S. J., & Trippas, D. ( 2015).

Dual processes and the interplay between knowledge and structure: A new parallel processing model

Psychology of Learning and Motivation, 62, 33-58.

DOI:10.1016/bs.plm.2014.09.002      URL     [本文引用: 3]

Johnson E. D., Tubau E., & De Neys W . (2016).

The doubting system 1: Evidence for automatic substitution sensitivity

Acta Psychogica, 164, 56-64.

DOI:10.1016/j.actpsy.2015.12.008      URL     PMID:26722837      [本文引用: 1]

A long prevailing view of human reasoning suggests severe limits on our ability to adhere to simple logical or mathematical prescriptions. A key position assumes these failures arise from insufficient monitoring of rapidly produced intuitions. These faulty intuitions are thought to arise from a proposed substitution process, by which reasoners unknowingly interpret more difficult questions as easier ones. Recent work, however, suggests that reasoners are not blind to this substitution process, but in fact detect that their erroneous responses are not warranted. Using the popular bat-and-ball problem, we investigated whether this substitution sensitivity arises out of an automatic System 1 process or whether it depends on the operation of an executive resource demanding System 2 process. Results showed that accuracy on the bat-and-ball problem clearly declined under cognitive load. However, both reduced response confidence and increased response latencies indicated that biased reasoners remained sensitive to their faulty responses under load. Results suggest that a crucial substitution monitoring process is not only successfully engaged, but that it automatically operates as an autonomous System 1 process. By signaling its doubt along with a biased intuition, it appears System 1 is marter than traditionally assumed.

Johnson-Laird, P. N.( 1983).

Mental models: Towards a cognitive science of language, inference and consciousness

Harvard University Press.

Kahneman D., ( 2011).

Thinking, fast and slow

New York: Farrar, Straus and Giroux.

[本文引用: 1]

Kahneman D. , & Frederick, S.( 2005) .

A model of heuristic judgment

In K. J. Holyoak & R. G. Morrison (Eds.), The Cambridge handbook of thinking and reasoning (pp. 267-293). Cambridge: Cambridge University Press.

URL     [本文引用: 1]

The program of research now known as the heuristics and biases approach began with a study of the statistical intuitions of experts, who were found to be excessively confident in the replicability of results from small samples. The persistence of such systematic errors in the intuitions of experts implied that their intuitive judgments may be governed by fundamentally different processes than the slower, more deliberate computations they had been trained to execute. The ancient idea that cognitive processes can be partitioned into two main families--traditionally called intuition and reason--is now widely embraced under the general label of dual-process theories. Dual-process models come in many flavors, but all distinguish cognitive operations that are quick and associative from others that are slow and governed by rules. To represent intuitive and deliberate reasoning, we borrow the terms "system 1" and "system 2" from Stanovich and West. In the following section, we present an attribute-substitution model of heuristic judgment, which assumes that difficult questions are often answered by substituting an answer to an easier one. Following sections introduce a research design for studying attribute substitution, as well as discuss the controversy over the representativeness heuristic in the context of a dual-system view that we endorse. The final section situates representativeness within a broad family of prototype heuristics, in which properties of a prototypical exemplar dominate global judgments concerning an entire set

Keren G., ( 2013).

A tale of two systems: A scientific advance or a theoretical stone soup? Commentary on evans stanovich

Perspectives on Psychological Science, 8( 3), 257-262.

DOI:10.1177/1745691613483474      URL     [本文引用: 1]

Keren G., & Schul, Y. ( 2009).

Two is not always better than one: A critical evaluation of two-system theories

Perspectives on Psychological Science, 4( 6), 533-550.

DOI:10.1111/j.1745-6924.2009.01164.x      URL     [本文引用: 1]

Since moving to Berlin in January for my semester abroad, I have witnessed many fantastic scenes while riding the U-Bahn. Ranging from a suit-clad man hurdle-jumping into the train to women in hijabs gossiping about another passenger in a burqa, my glimpses of Berlin never cease to amaze me, but last week, as I watched a family say goodbye, the children tearfully waving long after the train left, I shot back to my own tearful goodbyes. The exchange I saw was so loving, the exact opposite of all those years I transferred from one parent to the other: every Monday, every other Thursday, every holiday, and every birthday. [excerpt]

Klein G., ( 1998).

Sources of power: How people make decisions

MIT Press.

URL    

Klein G., ( 2015).

A naturalistic decision making perspective on studying intuitive decision making

Journal of Applied Research in Memory and Cognition, 4( 3), 164-168.

DOI:10.1016/j.jarmac.2015.07.001      URL     [本文引用: 1]

The Naturalistic Decision Making (NDM) community defines intuition as based on large numbers of patterns gained through experience, resulting in different forms of tacit knowledge. This view contrasts with Fast and Frugal Heuristics (FFH) researchers, who view intuition in terms of general purpose heuristics. The NDM view also differs from the Heuristics and Biases (HB) community, which sees intuitions as a source of bias and error. Seven suggestions are offered to assist the FFH and H&B communities in improving intuitive decision making and in conducting research that has greater potential for application. Rather than trying to help people analyze which option to choose, the NDM community recommends that intuitions be strengthened by providing a broader experience base that lets people build better tacit knowledge, such as perceptual skills and richer mental models, as a means of achieving better decisions.

Kokis J. V., Macpherson R., Toplak M. E., West R. F., & Stanovich K. E . ( 2002).

Heuristic and analytic processing: Age trends and associations with cognitive ability and cognitive styles

Joural of Experimental Child Psychology, 83( 1), 26-52.

DOI:10.1016/S0022-0965(02)00121-2      URL     [本文引用: 1]

Kruglanski, A. W . ( 2013).

Only one? The default interventionist perspective as a unimodel—Commentary on Evans & Stanovich (2013)

Perspectives on Psychological Science, 8( 3), 242-247.

DOI:10.1177/1745691613483477      URL     [本文引用: 1]

Kruglanski A. W., & Gigerenzer, G. ( 2011).

Intuitive and deliberate judgments are based on common principles

Psychological Review, 118( 1), 97-109.

DOI:10.1037/a0020762      URL     [本文引用: 1]

Levinson S. C. ( 1995).

Interactional biases in human thinking

In E. Goody (Ed. ), Social intelligence and interaction (pp. 221-260). Cambridge University Press.

DOI:10.1017/CBO9780511621710.014      URL    

There is a growing view that intelligence evolved as a product of social independence and that intelligence was linked to the use of spoken language. Taking as their starting-point the social production of intelligence and of language, scholars from a range of disciplines are beginning to rethink fundamental questions about human evolution, language and social institutions. In this volume, anthropologists, linguists, primatologists and pychologists come together to work on this new frontier of research.

Liang P. P., Goel V., Jia X. Q., & Li K. C . ( 2014).

Different neural systems contribute to semantic bias and conflict detection in the inclusion fallacy task

Frontiers in Human Neuroscience, 8, 797.

[本文引用: 1]

Lieberman, M. D . ( 2007).

Social cognitive neuroscience: A review of core processes

Annual Review of Psychology, 58, 259-289.

DOI:10.1146/annurev.psych.58.110405.085654      URL     [本文引用: 1]

Markovits H., Brisson J ., & de Chantal, P. L.( 2015).

Additional evidence for a dual-strategy model of reasoning: Probabilistic reasoning is more invariant than reasoning about logical validity

Memory & Cognition, 43( 8), 1208-1215.

[本文引用: 1]

Markovits H., Brisson J., de Chantal P. L., & Thompson V. A . ( 2017).

Interactions between inferential strategies and belief bias

Memory & Cognition, 45( 7), 1182-1192.

DOI:10.3758/s13421-017-0723-2      URL     PMID:28608194      [本文引用: 1]

The dual strategy model of reasoning proposed by Verschueren, Schaeken, and d’Ydewalle (Thinking & Reasoning, 11(3), 239–278, 2005a; Memory & Cognition, 33(1), 107–119, 2005b) suggests that people can

Markovits H., Brunet M.-L., Thompson V., & Brisson J . ( 2013).

Direct evidence for a dual process model of deductive inference

Journal of Experimental Psychology: Learning, Memory, and Cognition, 39( 4), 1213-1222.

DOI:10.1037/a0030906      URL     PMID:23206167      [本文引用: 1]

In 2 experiments, we tested a strong version of a dual process theory of conditional inference (cf. Verschueren et al., 2005a, 2005b) that assumes that most reasoners have 2 strategies available, the choice of which is determined by situational variables, cognitive capacity, and metacognitive control. The statistical strategy evaluates inferences probabilistically, accepting those with high conditional probability. The counterexample strategy rejects inferences when a counterexample shows the inference to be invalid. To discriminate strategy use, we presented reasoners with conditional statements (if p, then q) and explicit statistical information about the relative frequency of the probability of plq (50% vs. 90%). A statistical strategy would accept the more probable inferences more frequently, whereas the counterexample one would reject both. In Experiment 1, reasoners under time pressure used the statistical strategy more, but switched to the counterexample strategy when time constraints were removed; the former took less time than the latter. These data are consistent with the hypothesis that the statistical strategy is the default heuristic. Under a free-time condition, reasoners preferred the counterexample strategy and kept it when put under time pressure. Thus, it is not simply a lack of capacity that produces a statistical strategy; instead, it seems that time pressure disrupts the ability to make good metacognitive choices. In line with this conclusion, in a 2nd experiment, we measured reasoners' confidence in their performance; those under time pressure were less confident in the statistical than the counterexample strategy and more likely to switch strategies under free-time conditions.

Markovits H., Forgues H. L., & Brunet M. L . ( 2012).

More evidence for a dual-process model of conditional reasoning

Memory & Cognition, 40( 5), 736-747.

DOI:10.3758/s13421-012-0186-4      URL     PMID:22287219      [本文引用: 1]

Abstract Many studies have shown that the deductive inferences that people make have global properties that reflect the statistical information implicit in the premises. This suggests that such reasoning can be explained by a single, underlying probabilistic model. In contrast, the dual process model of conditional reasoning (Verschueren, Schaeken, & d'Ydewalle, 2005b) proposes that people can use either a logical, counterexample-based strategy or a probabilistic one. In two studies, we presented reasoners with sequences of affirmation-of-the-consequent inferences that differed with respect to the statistical properties of the premises, either explicitly or implicitly. As predicted by the dual-process model, an analysis of individual response patterns showed the presence of two distinct strategies, with use of the counterexample strategy being associated with higher levels of abstract-reasoning competence. Use of the counterexample strategy was facilitated by the explicit presentation of counterexample information. In a further study, we then examined explicitly probabilistic inferences. This study showed that although most reasoners made statistically appropriate inferences, the ability to make more-accurate inferences was associated with higher levels of abstract reasoning competence. These results show that deductive inferential reasoning cannot be explained by a single, unitary process and that any analysis of reasoning must consider individual differences in strategy use.

Osman M., ( 2004).

An evaluation of dual-process theories of reasoning

Psychonomic Bulletin & Review, 11( 6), 988-1010.

DOI:10.3758/BF03196730      URL     PMID:15875969      [本文引用: 1]

Abstract Current theories propose that reasoning comprises two underlying systems (Evans & Over, 1996; Sloman, 1996; Stanovich & West, 2000). The systems are identified as having functionally distinct roles, differ according to the type of information encoded, vary according to the level of expressible knowledge, and result in different responses. This article evaluates the arguments and the evidence from a select number of key tasks that have been supportive of dual-reasoning theorists' proposals. The review contrasts the dualist approach with a single-system framework that conjectures that different types of reasoning arise through the graded properties of the representations that are utilized while reasoning, and the different functional roles that consciousness has in cognition. The article concludes by arguing in favor of the alternative framework, which attempts to unify thedifferent forms of reasoning identified by dual-process theorists under a single system.

Osman M., ( 2013).

A case study: Dual-process theories of higher cognition—Commentary on Evans & Stanovich (2013)

Perspectives on Psychological Science, 8( 3), 248-252.

DOI:10.1177/1745691613483475      URL     [本文引用: 1]

Pennycook G., ( 2017).

A perspective on the theoretical foundation of dual-process models

In W. De Neys (Ed.), Dual process theory 2.0 . New York, NY: Psychology Press.

URL     [本文引用: 2]

Abstract Dual-process theories formalize a salient feature of human cognition: We have the capacity to rapidly formulate answers to questions, but we sometimes engage in deliberate reasoning processes before responding. It does not require deliberative thought to respond to the question “what is your name”. It did, however, require some thinking to write this paragraph (perhaps not enough). We have, in other words, two minds that might influence what we decide to do (Evans, 2003; Evans & Frankish, 2009). Although this distinction is acceptable (and, as I’ll argue, essentially irrefutable), it poses serious challenges for our understanding of cognitive architecture. In this chapter, I will outline what I view to be important theoretical groundwork for future dual-process models. I will start with two core premises that I take to be foundational: 1) dual-process theory is irrefutable but falsifiable, and 2) analytic thought has to be triggered by something. I will then use these premises to outline my perspective on what I consider the most substantial challenge for dual-process theorists: We don’t (yet) know what makes us think.

Pennycook G., Cheyne J. A., Barr N., Koehler D. J., & Fugelsang J. A . ( 2014).

Cognitive style and religiosity: The role of conflict detection

Memory and Cognition, 42( 1), 1-10.

DOI:10.3758/s13421-013-0340-7      URL     PMID:23784742      [本文引用: 1]

Abstract Recent research has indicated a negative relation between the propensity for analytic reasoning and religious beliefs and practices. Here, we propose conflict detection as a mechanism underlying this relation, on the basis of the hypothesis that more-analytic people are less religious, in part, because they are more sensitive to conflicts between immaterial religious beliefs and beliefs about the material world. To examine cognitive conflict sensitivity, we presented problems containing stereotypes that conflicted with base-rate probabilities in a task with no religious content. In three studies, we found evidence that religiosity is negatively related to conflict detection during reasoning. Independent measures of analytic cognitive style also positively predicted conflict detection. The present findings provide evidence for a mechanism potentially contributing to the negative association between analytic thinking and religiosity, and more generally, they illustrate the insights to be gained from integrating individual-difference factors and contextual factors to investigate analytic reasoning.

Pennycook G., Cheyne J. A., Seli P., Koehler D. J., & Fugelsang J. A . ( 2012).

Analytic cognitive style predicts religious and paranormal belief

Cognition, 123( 3), 335-346.

DOI:10.1016/j.cognition.2012.03.003      URL     PMID:22481051      [本文引用: 3]

An analytic cognitive style denotes a propensity to set aside highly salient intuitions when engaging in problem solving. We assess the hypothesis that an analytic cognitive style is associated with a history of questioning, altering, and rejecting (i.e., unbelieving) supernatural claims, both religious and paranormal. In two studies, we examined associations of God beliefs, religious engagement (attendance at religious services, praying, etc.), conventional religious beliefs (heaven, miracles, etc.) and paranormal beliefs (extrasensory perception, levitation, etc.) with performance measures of cognitive ability and analytic cognitive style. An analytic cognitive style negatively predicted both religious and paranormal beliefs when controlling for cognitive ability as well as religious engagement, sex, age, political ideology, and education. Participants more willing to engage in analytic reasoning were less likely to endorse supernatural beliefs. Further, an association between analytic cognitive style and religious engagement was mediated by religious beliefs, suggesting that an analytic cognitive style negatively affects religious engagement via lower acceptance of conventional religious beliefs. Results for types of God belief indicate that the association between an analytic cognitive style and God beliefs is more nuanced than mere acceptance and rejection, but also includes adopting less conventional God beliefs, such as Pantheism or Deism. Our data are consistent with the idea that two people who share the same cognitive ability, education, political ideology, sex, age and level of religious engagement can acquire very different sets of beliefs about the world if they differ in their propensity to think analytically.

Pennycook G., Fugelsang J. A., & Koehler D. J . ( 2012).

Are we good at detecting conflict during reasoning?

Cognition, 124( 1), 101-106.

DOI:10.1016/j.cognition.2012.04.004      URL     PMID:22575046      [本文引用: 1]

Recent evidence suggests that people are highly efficient at detecting conflicting outputs produced by competing intuitive and analytic reasoning processes. Specifically, De Neys and Glumicic (2008) demonstrated that participants reason longer about problems that are characterized by conflict (as opposed to agreement) between stereotypical personality descriptions and base-rate probabilities of group membership. However, this finding comes from problems involving probabilities much more extreme than those used in traditional studies of base-rate neglect. To test the degree to which these findings depend on such extreme probabilities, we varied base-rate probabilities over five experiments and compared participants response time for conflict problems with non-conflict problems. Longer response times for stereotypical responses to conflict versus non-conflict problems were found only in the presence of extreme probabilities. Our results suggest that humans may not be consistently efficient at detecting conflicts during reasoning.

Pennycook G., Fugelsang J. A., & Koehler D. J . ( 2015).

What makes us think? A three-stage dual-process model of analytic engagement

Cognitive Psychology, 80, 34-72.

DOI:10.1016/j.cogpsych.2015.05.001      URL     PMID:26091582      [本文引用: 2]

The distinction between intuitive and analytic thinking is common in psychology. However, while often being quite clear on the characteristics of the two processes (‘Type 1’ processes are fast, autonomous, intuitive, etc. and ‘Type 2’ processes are slow, deliberative, analytic, etc.), dual-process theorists have been heavily criticized for being unclear on the factors that determine when an individual will think analytically or rely on their intuition. We address this issue by introducing a three-stage model that elucidates the bottom-up factors that cause individuals to engage Type 2 processing. According to the model, multiple Type 1 processes may be cued by a stimulus (Stage 1), leading to the potential for conflict detection (Stage 2). If successful, conflict detection leads to Type 2 processing (Stage 3), which may take the form of rationalization (i.e., the Type 1 output is verified post hoc) or decoupling (i.e., the Type 1 output is falsified). We tested key aspects of the model using a novel base-rate task where stereotypes and base-rate probabilities cued the same (non-conflict problems) or different (conflict problems) responses about group membership. Our results support two key predictions derived from the model: (1) conflict detection and decoupling are dissociable sources of Type 2 processing and (2) conflict detection sometimes fails. We argue that considering the potential stages of reasoning allows us to distinguish early (conflict detection) and late (decoupling) sources of analytic thought. Errors may occur at both stages and, as a consequence, bias arises from both conflict monitoring and decoupling failures.

Pennycook G., & Thompson, V. A . ( 2012).

Reasoning with base rates is routine, relatively effortless, and context dependent

Psychonomic Bulletin & Review, 19( 3), 528-534.

DOI:10.3758/s13423-012-0249-3      URL     PMID:22427266      [本文引用: 1]

We tested models of base rate “neglect” using a novel paradigm. Participants ( N = 62) judged the probability that a hypothetical person belonged to one of two categories (e.g., nurse/doctor) on the basis of either a personality description alone (NoBR) or the personality description and a base rate probability (BR). When base rates and descriptions were congruent, judgments in the BR condition were higher and more uniform than those in the NoBR condition. In contrast, base rates had a polarizing effect on judgments when they were incongruent with the descriptions, such that estimates were either consistent with the base rates or discrepant with them. These data suggest that the form of base rate use (i.e., whether base rates will be integrated with diagnostic information) is context dependent. In addition, judgments made under instructions to respond intuitively were influenced by the base rates and took the same length of time in the two conditions. These data suggest that the use of base rates is routine and effortless and that base rate “neglect” is really a mixture of two strategies, one that is informed primarily by the base rate and the other by the personality description.

Pennycook G., Trippas D., Handley S. J., & Thompson V. A . ( 2014).

Base rates: Both neglected and intuitive

Journal of Experimental Psychology: Learning, Memory, and Cognition, 40( 2), 544-554.

DOI:10.1037/a0034887      URL     PMID:24219086      [本文引用: 2]

Base-rate neglect refers to the tendency for people to underweight base-rate probabilities in favor of diagnostic information. It is commonly held that base-rate neglect occurs because effortful (Type 2) reasoning is required to process base-rate information, whereas diagnostic information is accessible to fast, intuitive (Type 1) processing (e.g., Kahneman & Frederick, 2002). To test this account, we instructed participants to respond to base-rate problems on the basis of "beliefs" or "statistics," both in free time (Experiments 1 and 3) and under a time limit (Experiment 2). Participants were given problems with salient stereotypes (e.g., "Jake lives in a beautiful home in a posh suburb") that either conflicted or coincided with base-rate probabilities (e.g., "Jake was randomly selected from a sample of 5 doctors and 995 nurses for conflict; 995 doctors and 5 nurses for nonconflict"). If utilizing base-rates requires Type 2 processing, they should not interfere with the processing of the presumably faster belief-based judgments, whereas belief-based judgments should always interfere with statistics judgments. However, base-rates interfered with belief judgments to the same extent as the stereotypes interfered with statistical judgments, as indexed by increased response time and decreased confidence for conflict problems relative to nonconflict. These data suggest that base-rates, while typically underweighted or neglected, do not require Type 2 processing and may, in fact, be accessible to Type 1 processing.

Pollock J. L. ( 1991).

OSCAR: A general theory of rationality

In R. Cummins & J. L. Pollock (Eds.), Philosophy and AI: Essays at the interface (pp. 189-213). Cambridge, MA, US: The MIT Press.

DOI:10.1080/09528138908953702      URL    

The enterprise is the construction of a general theory of rationality, and its implementation in an automated reasoning system named OSCAR. The paper describes a general architecture for rational thought. This includes both theoretical reasoning and practical reasoning, and builds in important interconnections between them. It is urged that a sophisticated reasoner must be an introspective reasoner, capable of monitoring its own reasoning and reasoning about it. An introspective reasoner is built on top of a non-introspective reasoner that represents the system''s default reasoning strategies. The introspective reasoner engages in practical reasoning about reasoning in order to override these default strategies. The paper concludes with a discussion of some aspects of the default reasoner, including the manner in which reasoning is interest driven, and the structure of defeasible reasoning.

Posner, M. I. & Snyder, C. R. R . ( 1975).

Attention and cognitive control

In Robert L. Solso (Ed.),Information Processing and Cognition: The Loyola Symposium. Lawrence Erlbaum.

Prado J., Kaliuzhna M., Cheylus A., & Noveck I. A . ( 2008).

Overcoming perceptual features in logical reasoning: An event-related potentials study

Neuropsychologia, 46( 11), 2629-2637.

DOI:10.1016/j.neuropsychologia.2008.04.017      URL     PMID:18541277      [本文引用: 1]

Abstract It is more difficult for reasoners to detect that the letter-number pair H7 verifies the conditional rule If there is not a T then there is not a 4 than to detect that it verifies the rule If there is an H then there is a 7. In prior work [Prado, J., & Noveck, I. A. (2007). Overcoming perceptual features in logical reasoning: a parametric functional magnetic resonance imaging study. Journal of Cognitive Neuroscience 19(4), 642-657], we argued that this difficulty was due to mismatching effects, i.e. perceptual mismatches that arise when the items mentioned in the rule (e.g. T and 4) mismatch those presented in the test-pair (H and 7). The present study aimed to test this claim directly by recording ERPs while participants evaluated conditional rules in the presence or absence of mismatches. We found that mismatches, not only trigger a frontocentral N2 (an ERP known to be related to perceptual mismatch) but that they, parametrically modulate its amplitude (e.g. two mismatches prompt a greater N2 than one). Our results indicate that the main role of negations in conditional rules is to focus attention on the negated constituent but also suggest that there is some inter-individual differences in the way participants apprehend such negations, as indicated by a correlation between N2 amplitude and participants' reaction times. Overall, these findings emphasize how overcoming perceptual features plays a role in the mismatching effect and extend the mismatch-related effects of the N2 into a reasoning task.

Pyszczynski T., Greenberg J., & Solomon S . ( 1999).

A dual-process model of defense against conscious and unconscious death-related thoughts: An extension of terror management theory

Psychological Review, 106( 4), 835-845.

DOI:10.1037/0033-295X.106.4.835      URL     [本文引用: 2]

Scherer L. D., Yates J. F., Baker S. G., & Valentine K. D . ( 2017).

The influence of effortful thought and cognitive proficiencies on the conjunction fallacy: Implications for dual-process theories of reasoning and judgment

Personality and Social Psychology Bulletin, 43( 6), 874-887.

DOI:10.1177/0146167217700607      URL     [本文引用: 2]

Shiffrin, R. M. & Schneider W. , (1977).

Controlled and automatic human information processing: II. Perceptual learning, automatic attending, and a general theory

Psychological Review 84( 2), 127-90.

Sloman, S. A . ( 1996).

The empirical case for two systems of reasoning

Psychological Bulletin, 119( 1), 3-22.

DOI:10.1037/0033-2909.119.1.3      URL     [本文引用: 3]

Abstract Distinctions have been proposed between systems of reasoning for centuries. This article distills properties shared by many of these distinctions and characterizes the resulting systems in light of recent findings and theoretical developments. One system is associative because its computations reflect similarity structure and relations of temporal contiguity. The other is "rule based" because it operates on symbolic structures that have logical content and variables and because its computations have the properties that are normally assigned to rules. The systems serve complementary functions and can simultaneously generate different solutions to a reasoning problem. The rule-based system can suppress the associative system but not completely inhibit it. The article reviews evidence in favor of the distinction and its characterization.

Sloman, S. A . ( 2014).

Two systems of reasoning: An update

In J. W. Sherman, B. Gawronski, & Y. Trope (Eds.),Dual-process theories of the social mind (pp. 69-79). New York: Guilford Press

URL     [本文引用: 1]

Stanovich K. E., & West, R. F . ( 2000).

Individual differences in reasoning: Implications for the rationality debate?

Behavioral and Brain Sciences, 23( 5), 645-665.

DOI:10.1017/S0140525X00003435      URL     [本文引用: 2]

Stollstorff M., Vartanian O., & Goel V . ( 2012).

Levels of conflict in reasoning modulate right lateral prefrontal cortex

Brain Research, 1428, 24-32.

DOI:10.1016/j.brainres.2011.05.045      URL     PMID:21684531      [本文引用: 1]

78 Right lateral PFC (rlPFC) was activated during multiple forms of conflict in deductive reasoning (transitive inference problems). 78 Belief-logic conflict (4 types) was related to decreased logical reasoning performance and increased rlPFC activity. 78 Belief-content conflict (presence of counterfactual statements) modulated performance and rlPFC activity independently of belief-logic conflict. 78 Both belief-logic and belief-content conflict contribute to the belief-bias effect and associated rlPFC engagement. 78 Various forms of conflict in deductive reasoning consistently activate a specific region within rlPFC.

Stupple E. J. N., & Ball, L. J . ( 2008).

Belief-logic conflict resolution in syllogistic reasoning: Inspection-time evidence for a parallel-process model

Thinking & Reasoning, 14( 2), 168-181.

[本文引用: 3]

Toplak M. E., West R. F., & Stanovich K. E . ( 2014).

Assessing miserly information processing: An expansion of the Cognitive Reflection Test

Thinking & Reasoning, 20( 2), 147-168.

DOI:10.1080/13546783.2013.844729      URL     [本文引用: 1]

The Cognitive Reflection Test (CRT; Frederick, 2005) is designed to measure the tendency to override a prepotent response alternative that is incorrect and to engage in further reflection that leads to the correct response. It is a prime measure of the miserly information processing posited by most dual process theories. The original three-item test may be becoming known to potential participants, however. We examined a four-item version that could serve as a substitute for the original. Our data show that it displays a .58 correlation with the original version and that it has very similar relationships with cognitive ability, various thinking dispositions, and with several other rational thinking tasks. Combining the two versions into a seven-item test resulted in a measure of miserly processing with substantial reliability (.72). The seven-item version was a strong independent predictor of performance on rational thinking tasks after the variance accounted for by cognitive ability and thinking dispositions had been partialled out.

Trippas D., Handley S. J., Verde M. F., & Morsanyi K . ( 2016).

Logic brightens my day: Evidence for implicit sensitivity to logical validity

Journal of Experimental Psychology: Learning, Memory, and Cognition, 42( 9), 1448-1457.

DOI:10.1037/xlm0000248      URL     PMID:26889685      [本文引用: 1]

Abstract A key assumption of dual process theory is that reasoning is an explicit, effortful, deliberative process. The present study offers evidence for an implicit, possibly intuitive component of reasoning. Participants were shown sentences embedded in logically valid or invalid arguments. Participants were not asked to reason but instead rated the sentences for liking (Experiment 1) and physical brightness (Experiments 2-3). Sentences that followed logically from preceding sentences were judged to be more likable and brighter. Two other factors thought to be linked to implicit processing-sentence believability and facial expression-had similar effects on liking and brightness ratings. The authors conclude that sensitivity to logical structure was implicit, occurring potentially automatically and outside of awareness. They discuss the results within a fluency misattribution framework and make reference to the literature on discourse comprehension. (PsycINFO Database Record (c) 2016 APA, all rights reserved).

Trippas D., Thompson V. A., & Handley S. J . ( 2017).

When fast logic meets slow belief: Evidence for a parallel- processing model of belief bias

Memory & Cognition, 45( 4), 539-552.

DOI:10.3758/s13421-016-0680-1      URL     PMID:28028779      [本文引用: 1]

Two experiments pitted the default-interventionist account of belief bias against a parallel-processing model. According to the former, belief bias occurs because a fast, belief-based evaluation of the conclusion pre-empts a working-memory demanding logical analysis. In contrast, according to the latter both belief-based and logic-based responding occur in parallel. Participants were given deductive reasoning problems of variable complexity and instructed to decide whether the conclusion was valid on half the trials or to decide whether the conclusion was believable on the other half. When belief and logic conflict, the default-interventionist view predicts that it should take less time to respond on the basis of belief than logic, and that the believability of a conclusion should interfere with judgments of validity, but not the reverse. The parallel-processing view predicts that beliefs should interfere with logic judgments only if the processing required to evaluate the logical structure exceeds that required to evaluate the knowledge necessary to make a belief-based judgment, and vice versa otherwise. Consistent with this latter view, for the simplest reasoning problems (modus ponens), judgments of belief resulted in lower accuracy than judgments of validity, and believability interfered more with judgments of validity than the converse. For problems of moderate complexity (modus tollens and single-model syllogisms), the interference was symmetrical, in that validity interfered with belief judgments to the same degree that believability interfered with validity judgments. For the most complex (three-term multiple-model syllogisms), conclusion believability interfered more with judgments of validity than vice versa, in spite of the significant interference from conclusion validity on judgments of belief. The online version of this article (doi:10.3758/s13421-016-0680-1) contains supplementary material, which is available to authorized users.

Thompson V. A., & Johnson, S. C . ( 2014).

Conflict, metacognition, and analytic thinking

Thinking & Reasoning, 20( 2), 215-244.

DOI:10.1080/13546783.2013.869763      URL     [本文引用: 1]

One hundred and three participants solved conflict and non-conflict versions of four reasoning tasks using a two-response procedure: a base rate task, a causal reasoning task, a denominator neglect task, and a categorical syllogisms task. Participants were asked to give their first, intuitive answer, to make a Feeling of Rightness (FOR) judgment, and then were given as much time as needed to rethink their answer. They also completed a standardized measure of IQ and the actively open-minded thinking questionnaire. The FORs of both high- and low-capacity reasoners were responsive to conflict, such that FORs were lower for conflict relative to non-conflict problems. Consistent with the quantity hypothesis, high-capacity reasoners made a greater distinction between conflict and non-conflict items on measures of Type 2 thinking, namely, rethinking time and probability of changing answers. In contrast to the quality hypothesis, however, this rethinking time did not advantage the ability of the high-capacity group to produce normative answers, except for the base rate task. Indeed, we observed that the correlation between capacity and the probability of normative answers emerged at the initial response, rather than after rethinking.

Thompson V. A., & Morsany, K. ( 2012).

Analytic thinking: Do you feel like it?

Mind & Society, 11( 1), 93-105.

DOI:10.1007/s11299-012-0100-6      URL     [本文引用: 1]

A major challenge for Dual Process Theories of reasoning is to predict the circumstances under which intuitive answers reached on the basis of Type 1 processing are kept or discarded in favour of analytic, Type 2 processing (Thompson 2009 ). We propose that a key determinant of the probability that Type 2 processes intervene is the affective response that accompanies Type 1 processing. This affective response arises from the fluency with which the initial answer is produced, such that fluently produced answers give rise to a strong feeling of rightness. This feeling of rightness, in turn, determines the extent and probability with which Type 2 processes will be engaged. Because many of the intuitions produced by Type 1 processes are fluent, it is common for them to be accompanied by a strong sense of rightness. However, because fluency is poorly calibrated to objective difficulty, confidently held intuitions may form the basis of poor quality decisions.

Thompson V. A., Turner J. A. P., Pennycook G., Ball L. J., Brack H., Ophir Y., & Ackerman R . ( 2013).

The role of answer fluency and perceptual fluency as metacognitive cues for initiating analytic thinking

Cognition, 128( 2), 237-251.

DOI:10.1016/j.cognition.2012.09.012      URL    

/


版权所有 © 《心理科学进展》编辑部
地址:北京市朝阳区林萃路16号院 
邮编:100101 
电话:010-64850861 
E-mail:jinzhan@psych.ac.cn
备案编号:京ICP备10049795号-1 京公网安备110402500018号

本系统由北京玛格泰克科技发展有限公司设计开发