这是用户在 2025-6-27 13:04 为 https://app.immersivetranslate.com/pdf-pro/6c29e391-2ec9-442b-9a98-cc516f3dd58b/ 保存的双语快照页面,由 沉浸式翻译 提供双语支持。了解如何保存?

Is a picture worth a thousand words? The interaction of visual display and attribute representation in attenuating framing bias
一张图片胜过千言万语吗?视觉显示和属性表示在减弱框架偏差中的交互作用

Eyal Gamliel* Hamutal Kreiner ^(†){ }^{\dagger}

Abstract  抽象

The attribute framing bias is a well-established phenomenon, in which an object or an event is evaluated more favorably when presented in a positive frame such as “the half full glass” than when presented in the complementary negative framing. Given that previous research showed that visual aids can attenuate this bias, the current research explores the factors underlying the attenuating effect of visual aids. In a series of three experiments, we examined how attribute framing bias is affected by two factors: (a) The display mode-verbal versus visual; and (b) the representation of the critical attribute-whether one outcome, either the positive or the negative, is represented or both outcomes are represented. In Experiment 1 a marginal attenuation of attribute framing bias was obtained when verbal description of either positive or negative information was accompanied by corresponding visual representation. In Experiment 2 similar marginal attenuation was obtained when both positive and negative outcomes were verbally represented. In Experiment 3, where the verbal description represented both positive and negative outcomes, significant attenuation was obtained when it was accompanied by a visual display that represented a single outcome, and complete attenuation, totally eliminating the framing bias, was obtained when it was accompanied by a visual display that represented both outcomes. Thus, our findings showed that interaction between the display mode and the representation of the critical attribute attenuated the framing bias. Theoretical and practical implications of the interaction between verbal description, visual aids and representation of the critical attribute are discussed, and future research is suggested.
属性框架偏差是一种公认的现象,其中,当对象或事件以正框架(如“半满玻璃杯”)呈现时,比以互补消极框架呈现时更有利。鉴于先前的研究表明视觉辅助设备可以减轻这种偏差,目前的研究探讨了视觉辅助设备衰减效果的潜在因素。在一系列的三个实验中,我们研究了属性框架偏差如何受到两个因素的影响:(a) 显示模式——语言与视觉;(b) 关键属性的表示 - 是否表示一个结果,无论是积极的还是消极的,还是两个结果都被表示。在实验 1 中,当正面或负面信息的口头描述伴随着相应的视觉表示时,获得了属性框架偏差的边际衰减。在实验 2 中,当口头表示积极和消极结果时,获得了类似的边际衰减。在实验 3 中,口头描述同时代表积极和消极结果,当它伴随着代表单个结果的视觉显示时,获得了显着的衰减,而当它伴随着代表两种结果的视觉显示时,获得了完全衰减,完全消除了框架偏倚。因此,我们的研究结果表明,显示模式和关键属性的表示之间的交互减弱了框架偏差。讨论了口头描述、视觉辅助和关键属性表示之间相互作用的理论和实践意义,并提出了未来的研究建议。

Keywords: attribute framing, visual display, framing bias, attribute representation.
关键词:属性框架、视觉展示、框架偏差、属性表示。

1 Introduction  1 引言

Various writers have suggested that biases in judgments and decisions may be reduced by concrete and visual presentation of data (e.g., Gigerenzer et al., 2007). In this paper we focus on the attribute-framing bias and examine whether it is moderated by displaying the critical attribute visually in a graph alongside the verbal description. While previous studies have already showed attenuation of the framing bias using visual aids (e.g., GarciaRetamero & Cokely, 2011; Garcia-Retamero & Galesic, 2010) the aim of the current research was to explore the factors underlying this attenuation. Based on the literature, we propose that two factors are involved in visual attenuation. The first is the display mode-whether the problem is presented verbally or visually. The second factor is the representation of the critical attributewhether both the positive and the negative outcomes or only one outcome are represented. Systematic examination of the interaction of these two factors may shed light
多位作者建议,通过具体和可视化的数据呈现可以减少判断和决定中的偏见(例如,Gigerenzer 等人,2007 年)。在本文中,我们专注于属性框架偏差,并通过在图表中与口头描述一起直观地显示关键属性来检查它是否受到调节。虽然之前的研究已经显示使用视觉辅助工具可以减弱框架偏差(例如,GarciaRetamero & Cokely,2011 年;Garcia-Retamero & Galesic, 2010)当前研究的目的是探索这种衰减背后的因素。根据文献,我们提出视觉衰减涉及两个因素。首先是显示模式 - 问题是以口头方式还是视觉方式呈现。第二个因素是关键属性的表示,无论是正面和负面结果,还是只表示一个结果。对这两个因素相互作用的系统研究可能会有所启发
on the cognitive processes underlying attribute-framing bias. Moreover, it may have practical implications, given that attribute framings influence important judgments and decisions in health, financial decisions, consumer behavior, and many other domains (see review in Levin et al., 1998).
关于属性框架偏见的认知过程。此外,鉴于属性框架会影响健康、财务决策、消费者行为和许多其他领域的重要判断和决策,它可能具有实际意义(参见 Levin 等人,1998 年的评论)。

1.1 Framing effects  1.1 取景效果

Research on framing effects took off after Tversky and Kahneman (1981) introduced the systematic reversals of preferences that people exhibit in alternate framings of problems, contingencies or outcomes. Following Tversky and Kahneman (1981), many studies examined the valence framing effect in various contexts, demonstrating that the mere presentation of the same glass as halffull or half empty affects judgment and decision making (Frisch, 1993; Keren, 2010).
在 Tversky 和 Kahneman (1981) 引入了人们在问题、偶发事件或结果的交替框架中表现出的偏好的系统性逆转之后,对框架效应的研究开始了。继 Tversky 和 Kahneman (1981) 之后,许多研究考察了各种情况下的价框架效应,表明仅仅将同一个杯子呈现为半满或半空会影响判断和决策(Frisch,1993;Keren,2010 年)。
Reviewing the abundant research on framing effects, Levin et al. (1998) discerned three types of framing: risky choice framing, goal framing, and attribute framing. Tversky and Kahneman’s (1981) famous “Asian disease” is the most well-known typical example of the first type, risky choice framing. They demonstrated that people show reversed preferences between two options when these were presented in either a positive (“lives saved”)
回顾了对框架效应的大量研究,Levin et al. (1998) 辨别出了三种类型的框架:风险选择框架、目标框架和属性框架。Tversky 和 Kahneman (1981) 著名的“亚洲病”是第一种类型、风险选择框架最著名的典型例子。他们证明,当两个选项以正面(“挽救生命”)表示时,人们会在这两个选项之间表现出相反的偏好

or negative (“lives lost”) frame. The second type, goal framing, relates to the persuasive power of presenting the same message in either gain terms when performing an action or loss terms when not performing it. A typical context of goal framing is health related behavior, in which both society and each individual would benefit if a particular health promoting or disease preventing behavior is adopted. Rothman and Salovey (1997) proposed that the perceived risk of the health behavior moderates the effectiveness of the message frame: for relatively safe behaviors (e.g., prevention behavior) positive framing is more effective, while for risky behaviors (e.g., detection of an illness) negative framing is more effective. Although this moderating effect was reported upon in the literature, recent meta-analysis has raised some doubt about its empirical veracity (Gallagher & Updergraff, 2012).
或负数(“lives lost”)框架。第二种类型,目标框架,与在执行某项作时以收益项或不执行某项时以损失项呈现相同信息的说服力有关。目标框架的一个典型背景是与健康相关的行为,如果采用特定的健康促进或疾病预防行为,社会和每个人都会受益。Rothman 和 Salovey (1997) 提出,健康行为的感知风险会调节信息框架的有效性:对于相对安全的行为(例如,预防行为),积极的框架更有效,而对于危险的行为(例如,检测到疾病),消极的框架更有效。尽管这种调节效应在文献中有所报道,但最近的元分析对其实证真实性提出了一些怀疑(Gallagher & Updergraff,2012)。
In this paper we focus on the third type, attribute framing. Specifically, we examine the situation in which one can refer to an object or event by presenting either the positive aspect of a central attribute (e.g., the success rate in an academic course) or the complementary negative aspect of the same attribute (e.g., the failure rate in the same course). Such framing was shown to result in a bias in evaluation. For example, Levin and Gaeth (1988) showed that people evaluated the quality of ground beef as better when it was labeled as “75% lean” (positive framing) relative to “25% fat” (negative framing). Attribute framing typically shows a robust effect with an advantage for the positive versus negative framing when evaluating events or objects. Levin et al. (1998) reviewed dozens of studies and concluded that, in most of them, positive framing resulted in more favorable evaluations of objects or events relative to negative framing. In a meta-analysis of 30 studies, Pinon and Gambara (2005) reported an average attribute framing effect size of 0.26 , but they suspected that this relatively small effect size reflected an underestimate due to the statistical methods used for the meta-analysis.
在本文中,我们重点介绍第三种类型,属性框架。具体来说,我们通过呈现中心属性的积极方面(例如,学术课程的成功率)或相同属性的互补消极方面(例如,同一课程的失败率)来研究一个人可以指代对象或事件的情况。这种框架被证明会导致评估中的偏倚。例如,Levin 和 Gaeth (1988) 表明,当碎牛肉被标记为“75% 瘦肉”(正面框架)相对于“25% 脂肪”(负面框架)时,人们评估碎牛肉的质量要好。属性框架通常显示稳健的效果,在评估事件或对象时,正框架优于负框架。Levin et al. (1998) 回顾了数十项研究,并得出结论,在大多数研究中,相对于消极框架,积极框架导致对对象或事件的评价更有利。在对 30 项研究的荟萃分析中,Pinon 和 Gambara (2005) 报告了平均属性框架效应大小为 0.26 ,但他们怀疑这个相对较小的效应量反映了由于用于荟萃分析的统计方法而被低估。
Levin et al. (1998) proposed a theoretical explanation for the cognitive mechanism through which attribute framing affects people’s judgments and evaluations. They suggested that presenting an object or an event in positive or negative framing affects information processing in a way similar to priming. Positive or negative framing activates corresponding favorable or unfavorable associations that bias people’s judgments and evaluations (Levin et al., 1998). This theoretical explanation attributed the framing bias to information encoding and processing and was not restricted to verbal presentation. Although most studies examined verbal attribute framing, we argue that examining the effect of visual display mode on attribute-framing bias might help understand the cognitive processes underlying framing bias.
Levin et al. (1998) 对属性框架影响人们判断和评价的认知机制提出了一种理论解释。他们认为,以积极或消极的框架呈现对象或事件以类似于启动的方式影响信息处理。积极或消极的框架会激活相应的有利或不利的联想,从而使人们的判断和评价产生偏见(Levin et al., 1998)。这种理论解释将框架偏差归因于信息编码和处理,并不局限于口头表达。尽管大多数研究都检查了语言属性框架,但我们认为,检查视觉显示模式对属性框架偏差的影响可能有助于理解框架偏差背后的认知过程。

1.2 Visual framing versus verbal framing
1.2 视觉框架与语言框架

In discussing people’s errors in the evaluation of quantitative data, Gigerenzer et al. (2007) argued that one reason for errors and biased evaluations may be that the presentation of numerical data is too difficult and that presenting data visually might offer a more concrete representation that would be easier to process and thereby attenuate biases relative to verbal presentation of numerical data.
在讨论人们在评估定量数据时的错误时,Gigerenzer et al. (2007) 认为,错误和有偏见的评估的一个原因可能是数字数据的呈现太困难,而直观地呈现数据可能会提供更具体的表示,更容易处理,从而减轻相对于数字数据的口头表示的偏见。
Several studies examined the possible effect of visual and graphical presentation on framing bias. For example, Sun et al. (2012) altered the spatial characteristics of a graphical display presenting two attributes of MP3 products, namely reliability and capability: Product A had superior reliability while product B had superior capability. The products were presented visually in a figure where the X axis represented capability and the Y axis represented reliability. The relative length of the axes was manipulated such that in one figure the axis representing capability was longer than that representing reliability, and in the other this was reversed. Participants gave higher evaluations to the product (either A or B) when its superior aspect was represented by the longer axis. Thus, the authors concluded that the aspect that presents the differences in a more salient fashion had a stronger effect on participants’ evaluations (Sun et al., 2012) and suggested that evaluations are affected by the spatial characteristic of the visual display. Sun et al. examined framing problems in the broader or loose definition of framing. They did not examine the effect of visual display on valence framing.
几项研究检查了视觉和图形呈现对框架偏倚的可能影响。例如,Sun 等人 (2012) 改变了图形显示器的空间特性,呈现了 MP3 产品的两个属性,即可靠性和能力:产品 A 具有卓越的可靠性,而产品 B 具有卓越的能力。产品以图形形式直观地呈现,其中 X 轴代表能力,Y 轴代表可靠性。轴的相对长度纵,使得在一个图中代表能力的轴比代表可靠性的轴长,而在另一个图中,这是相反的。当产品的优越性由较长的轴表示时,参与者对产品(A 或 B)给予更高的评价。因此,作者得出结论,以更突出的方式呈现差异的方面对参与者的评价有更强的影响(Sun et al., 2012),并认为评价受到视觉显示的空间特征的影响。Sun 等人研究了更广泛或松散的框架定义中的框架问题。他们没有检查视觉显示对价框架的影响。
More relevant to the current paper are papers that showed that visual aids facilitated the processing of statistical information (e.g., Kurz-Milcke et al., 2008) in the context of applied psychology. For example, recent studies examined the efficacy of visual aids in preventing framing biases related to goal framing (Garcia-Retamero & Cokely, 2011) and attribute framing (Garcia-Retamero & Galesic, 2010) in the context of health information. In these studies participants were presented with a verbal message framed either in a positive or negative manner accompanied by an identical visual display. These findings clearly indicate that visual presentation of the statistical information can reduce the framing bias. These studies, however, did not explore what factors in the visual display contribute to the attenuation of the framing bias. Possibly, the visual attenuation may be due to the fact that an identical graph was presented in both positive and negative framing conditions. Thus, while the GarciaRetamero and Galesic (2010) study manipulated framing by presenting either the positive or the negative outcome in the verbal description, it did not manipulate framing in the visual display because both positive and negative outcomes were equally represented in the graphs. The
与当前论文更相关的是表明视觉辅助工具在应用心理学背景下促进了统计信息的处理(例如,Kurz-Milcke 等人,2008 年)。例如,最近的研究检查了视觉辅助工具在防止与目标框架相关的框架偏见(Garcia-Retamero & Cokely,2011)和属性框架(Garcia-Retamero & Galesic,2010)在健康信息背景下的效果。在这些研究中,参与者以积极或消极的方式呈现了口头信息,并伴有相同的视觉显示。这些发现清楚地表明,统计信息的视觉呈现可以减少框架偏倚。然而,这些研究并没有探讨视觉显示中的哪些因素导致框架偏差的减弱。视觉衰减可能是由于在正负取景条件下呈现了相同的图形。因此,虽然 GarciaRetamero 和 Galesic (2010) 研究通过在口头描述中呈现积极或消极的结果来纵框架,但它并没有在视觉显示中纵框架,因为积极和消极的结果在图表中都同等表示。这
Table 1: The text and/or figures presented in the four conditions of Experiment 1.
表 1:实验 1 的四个条件中呈现的文本和/或图表。
Positive framing  正向框架 Negative framing  否定框架
Verbal display  口头展示 You know a driving instructor 85% of whose students pass their driving test the first time they take it.
您认识一位驾驶教练,他 85% 的学生第一次参加驾驶考试就通过了考试。
You know a driving instructor 15% of whose students fail their driving test the first time they take it.
您认识一位驾驶教练,他 15% 的学生第一次参加驾驶考试时未通过。
Visual display  视觉显示 You know a driving instructor some of whose students pass their driving test the first time they take it, as described in the adjacent figure:
您认识一位驾驶教练,其中一些学生第一次参加驾驶考试就通过了考试,如相邻的图所示:
You know a driving instructor some of whose students fail their driving test the first time they take it, as described in the adjacent figure:
您认识一位驾驶教练,其中一些学生第一次参加驾驶考试时未通过考试,如相邻图所示:
Positive framing Negative framing Verbal display You know a driving instructor 85% of whose students pass their driving test the first time they take it. You know a driving instructor 15% of whose students fail their driving test the first time they take it. Visual display You know a driving instructor some of whose students pass their driving test the first time they take it, as described in the adjacent figure: You know a driving instructor some of whose students fail their driving test the first time they take it, as described in the adjacent figure: https://cdn.mathpix.com/cropped/2025_06_27_76305d376a4fe8e1add7g-3.jpg?height=358&width=382&top_left_y=738&top_left_x=509 https://cdn.mathpix.com/cropped/2025_06_27_76305d376a4fe8e1add7g-3.jpg?height=354&width=382&top_left_y=740&top_left_x=1179| | Positive framing | Negative framing | | :--- | :--- | :--- | | Verbal display | You know a driving instructor 85% of whose students pass their driving test the first time they take it. | You know a driving instructor 15% of whose students fail their driving test the first time they take it. | | Visual display | You know a driving instructor some of whose students pass their driving test the first time they take it, as described in the adjacent figure: | You know a driving instructor some of whose students fail their driving test the first time they take it, as described in the adjacent figure: | | | ![](https://cdn.mathpix.com/cropped/2025_06_27_76305d376a4fe8e1add7g-3.jpg?height=358&width=382&top_left_y=738&top_left_x=509) | ![](https://cdn.mathpix.com/cropped/2025_06_27_76305d376a4fe8e1add7g-3.jpg?height=354&width=382&top_left_y=740&top_left_x=1179) |
current research aims to extend the investigation about the factors affecting the attenuation of framing bias by systematically manipulating two factors: display modewhether the problem is presented verbally or visuallyand the representation of the critical attribute-whether both the positive and the negative outcomes or only one outcome are represented.
目前的研究旨在通过系统地纵两个因素来扩展对影响框架偏差衰减的因素的调查:显示模式问题是口头还是视觉呈现以及关键属性的表示——是同时表示积极和消极结果还是只表示一个结果。

1.3 Review of the experiments
1.3 实验回顾

In a series of three experiments, we systematically manipulated two factors, namely display mode and representation of the critical attribute, in order to examine their interactive contribution to the attenuation of the attribute framing bias. In Experiment 1 we represented only one outcome of the critical attribute, either the positive or the negative, and examined whether and to what extent adding a visual display of a particular outcome attenuates framing bias compared to verbal description without visual aids. Experiments 2 and 3 examined to what extent the representation of the critical attribute-whether both the positive and the negative outcomes or only one outcome are represented-affects attenuation of the framing bias. Experiment 2 focused on verbal descriptions and examined to what extent the mixed representation of both the positive and the negative outcomes attenuates framing bias compared to a single-outcome representation. Experiment 3 focused on visual display to similarly examine to what extent visual display of mixed versus single representation of the outcomes attenuate framing bias compared to a single outcome visual display.
在一系列的三个实验中,我们系统地纵了两个因素,即显示模式和关键属性的表示,以检查它们对属性框架偏差衰减的交互贡献。在实验 1 中,我们只代表了关键属性的一个结果,无论是积极的还是消极的,并研究了与没有视觉辅助的口头描述相比,添加特定结果的视觉显示是否以及在多大程度上减轻了框架偏差。实验 2 和 3 检查了关键属性的表示(无论是正面和负面结果都表示,还是只表示一个结果)在多大程度上影响了框架偏倚的衰减。实验 2 侧重于口头描述,并检查了与单一结果表示相比,积极和消极结果的混合表示在多大程度上减轻了框架偏倚。实验 3 侧重于视觉显示,以类似地检查与单一结果视觉显示相比,结果的混合表示与单一表示的视觉显示相比,在多大程度上减轻了框架偏倚。

2 Experiment 1  2 实验 1

The aim of Experiment 1 was to examine whether and to what extent visual display attenuates framing bias when different visual displays are used to represent the positive and negative outcomes. Whereas Garcia-Retamero and Galesic (2010) presented the same graphs for the negative and positive framing conditions, in this experiment we used different graphs to present the positive and negative numerical information (see Table 1). If visual aids generally attenuate the verbal framing bias, we would expect to replicate the attenuating effect found by GarciaRetamero and Galesic (2010). If, however, the attenuation of the framing bias actually results from the presentation of identical graphs for both the positive and negative framing conditions, we would expect only marginal attenuation of the framing bias, if any.
实验 1 的目的是检查当使用不同的视觉显示来表示积极和消极结果时,视觉显示是否以及在多大程度上减轻了框架偏差。虽然 Garcia-Retamero 和 Galesic (2010) 为负和正框架条件提供了相同的图表,但在这个实验中,我们使用不同的图表来表示正和负数值信息(见表 1)。如果视觉辅助工具通常减弱了语言框架偏差,我们有望复制 GarciaRetamero 和 Galesic (2010) 发现的衰减效果。但是,如果成帧偏差的衰减实际上是由于在正和负成帧条件下呈现相同的图形而导致的,那么我们预计成帧偏差只会有边际衰减(如果有的话)。

2.1 Method  2.1 方法

2.1.1 Participants  2.1.1 参与者

Participants were 141 undergraduate students in an undergraduate program in an Israeli higher education institution ( 85 % 85 % 85%85 \% female; mean age 23.5 with a standard deviation of 1.8 ) who participated in the study in partial fulfillment of their course requirement.
参与者是以色列高等教育机构本科课程的 141 名本科生( 85 % 85 % 85%85 \% 女性;平均年龄 23.5 岁,标准差为 1.8),她们参加了该研究,部分完成了他们的课程要求。

2.1.2 Design and Procedure
2.1.2 设计与程序

Experiment 1 comprised a between-participants 2 × 2 2 × 2 2xx22 \times 2 factorial design that manipulated the display mode (verbal versus visual) and the framing (positive versus negative).
实验 1 包括参与者 2 × 2 2 × 2 2xx22 \times 2 间因子设计,该设计纵显示模式(语言与视觉)和框架(积极与消极)。

  1. The authors wish to thank Lior Ginat and Itai Ziv for data collection.
    作者要感谢 Lior Ginat 和 Itai Ziv 的数据收集。
    Copyright: © 2013. The authors license this article under the terms of the Creative Commons Attribution 3.0 License.
    版权所有:© 2013 年。作者根据 Creative Commons Attribution 3.0 License 的条款对本文进行许可。

    *Behavioral Sciences Department, Ruppin Academic Center, Emek Hefer 40250, Israel. E-mail: eyalg @ ruppin.ac.il.
    *以色列 Emek Hefer 40250 Ruppin 学术中心行为科学系。电子邮件: eyalg @ ruppin.ac.il.

    ^(†){ }^{\dagger} Ruppin Academic Center
    ^(†){ }^{\dagger} 鲁平学术中心