Communicating probabilities through interactive computer graphics

描述（由申请人提供）：理解概率定量信息的能力是健康素养的重要组成部分。虽然事件的概率只是感知风险多维结构的一部分，但专家和医疗提供者经常使用概率来向公众传达风险。不幸的是，当概率以数字的形式呈现时，许多人无法理解或操纵它们。所有受教育程度的人在思考概率时会产生认知偏差，而低教育水平或计算能力会进一步损害理解能力。在计算能力和受教育程度较低的患者中，以概率或比例的形式呈现概率信息时，理解能力最差，而通过强调频率的图形（如图标网格）呈现概率信息时，理解能力最好。然而，即使是图形显示也不是对所有观众都成功，而且之前的许多研究都是理论上的。认知心理学的新研究表明，当概率是通过经验（比如从一副牌中抽牌）而不是用百分比（比如被告知23%的牌是脸牌）来学习时，概率推理中的认知偏差可以被最小化或改变。因此，我们预计，我们可以提高理解和改变对感知风险的影响，通过动画图形显示来创建虚拟的概率体验，并鼓励用户与显示器互动并操纵它们。本研究将(a)使用定性方法与非专业用户一起开发新的基于计算机的图形显示，将动画和用户交互结合起来，(b)在对大量异质用户样本的定量研究中评估这些显示的效果，以及(c)探索计算能力和健康素养对不同显示格式理解的影响。主要结果将是在不同风险情景、感知风险和定量理解之间的选择决策。用户满意度和界面可用性也将被评估。为了招募不同的样本，我们将在网上环境和哈莱姆健康促进中心招募，这是一个国家健康促进研究中心。本项目开发的交互式显示器有可能应用于更全面的健康素养教育、量身定制的健康和风险沟通、共享医疗决策和患者决策支持等项目。

DESCRIPTION (provided by applicant): The ability to understand quantitative information about probabilities is a critical component of health literacy. Although the probability of an event is only part of the multidimensional construct of perceived risk, probabilities are often used by experts and medical providers to convey risks to the public. Unfortunately, many people cannot understand or manipulate probabilities when they are presented as numbers. People of all educational levels are subject to cognitive biases when thinking about probabilities, and comprehension is further impaired by low educational level or numeracy. Among patients with low numeracy and educational level, comprehension is worst when probability information is presented in terms of probabilities or proportions, and best when it is presented through graphics that emphasize frequencies, such as grids of icons. However, not even graphical displays are successful with all audiences, and much previous research on them has been atheoretical. New research in cognitive psychology demonstrates that cognitive biases in probabilistic reasoning can be minimized or altered when probabilities are learned through experience (such as by drawing cards from a deck) rather than described as percentages (such as being told that 23 percent of the cards are face cards). Thus, we anticipate that we can improve comprehension and alter the effect on perceived risk by animating graphical displays to create virtual experiences of probabilities, and by encouraging users to interact with the displays and manipulate them. This study will (a) use qualitative methods with lay users to develop new computer-based graphical displays that incorporate animation and user interaction, (b) assess the effect of these displays in quantitative research on a large heterogeneous sample of users, and (c) explore the effects of numeracy and health literacy on comprehension of different display formats. The primary outcomes will be choice decisions between different risk scenarios, perceived risk, and quantitative comprehension. User satisfaction and interface usability will also be assessed. To enroll a diverse sample, we will recruit in both an on-line environment and from the Harlem Health Promotion Center, a national center for health promotion research. The interactive displays developed in this project have the potential to be applied to more comprehensive programs for health literacy education, tailored health and risk communication, shared medical decision-making, and patient decision support.