【2026-2期】

Regulating AI: Lessons From Scientific Computing

https://doi.org/10.25300/MISQ/2025/19111

Motivated by the perceived importance of explainability to AI regulation, this paper examines the complex relationship between theory, as a proxy for explainability, and computation in scientific practice. Drawing from both historical and contemporary examples of scientific computing, we make three key arguments: First, we show that theory-computation relationships exist in diverse configurations, ranging from theory-foregrounded and theory-backgrounded to more nuanced arrangements where theory is selectively integrated. Second, scientific fields strategically manage the inherent opacity of some computational methods by injecting theory at critical junctures, showing that complete “explainability” is not always a prerequisite for scientific progress or utility. Third, we argue that these lessons from scientific computing have important implications for AI regulation anchored in explainable AI (XAI). We suggest that general approaches to AI oversight should be complemented with a context-specific focus on how theory and computation interact in different domains. Implications for AI regulatory regimes and Information Systems (IS) research are discussed.

Keywords: Regulation, Explainable AI (XAI), Science, Scientific Computing

[!小结]
完全的可解释性并非科学进步的先决条件。

Tripartite Reactions to Emotions (TRE) Theory: The Impact of Emotional Expressions in User-Generated Content

https://doi.org/10.25300/MISQ/2025/18570

Although emotional expression is commonplace in online user-generated content (UGC), theoretical understanding of its consequences is limited and fragmented. We advance a theory of tripartite reactions to emotions (TRE) in UGC that addresses the following question: When and how do content creators’ emotional expressions about a target influence observers’ behaviors toward the target and the content itself? Integrating and extending existing theories, TRE theory distinguishes between two kinds of observer behavior, proposes three distinct cognitive and affective pathways, and highlights the moderating influence of UGC context. To guide future research, we develop an extended, dynamic model that allows for interactions among pathways, and we illustrate the application of TRE in an online word-of-mouth setting. TRE represents a novel, emotion-focused theory that researchers in information systems and other disciplines can leverage to explore the interpersonal impacts of emotional expressions on UGC platforms.

Keywords: User-generated content, emotional expression, affect, inference, behavior, content evaluation

[!小结]
问题：内容创作者的情绪表达如何影响用户对目标对象和内容本身的行为？

Health Information Technology Redeployment From Targets to Acquirers in the Healthcare Industry: The Role of Co-Specialized Human Capital in Value Creation

https://doi.org/10.25300/MISQ/2025/18693

In view of the increasing number of hospital mergers and acquisitions (M&As) and the significant role of healthcare services in the economy, a variety of recent studies have shed light on how health information technology (HIT) contributes to value creation from hospital M&As. However, little attention has been paid to value creation through the redeployment of the target hospital’s digital resources to the acquiring hospital and the conditions needed to create value from such resource redeployment. This paper highlights how post-M&A HIT redeployment from the target to the acquirer influences the quality of care of the acquirer. First, the results demonstrate the importance of retaining elements of the co-specialized human capital of the target hospital post-M&A as central to creating value from HIT redeployment. Second, HIT redeployment from the target to the acquirer is more valuable when the diagnostic context of the acquiring hospital is complementary to but also distinct from the diagnostic context of the target hospital. Third, the electronic health record system compatibility between the target and the acquirer also plays an essential role in facilitating value creation for the acquiring hospital from the transfer of knowledge-based capabilities from the target.

Keywords: Hospital M&A, HIT redeployment, adherence to recommended care, average length of stay, average inpatient cost, co-specialized human capital, diagnostic complementarity, EHR compatibility

[!小结]
问题：医院并购后，通过重新部署目标医院的数字资源（HIT）来为收购方创造价值（提升护理质量）的机制和条件是什么？

Overcoming Breakdowns in Customer-Chatbot Interaction: Design and Impact of Collaborative Repair Strategies

https://doi.org/10.25300/MISQ/2025/18742

When chatbots are deployed to automate customer service, it is nearly inevitable that situations will arise in which they struggle to understand customer requests. Unfortunately, the onus of resolving such conversational breakdowns tends to fall on either the customer or the chatbot alone, turning customer-chatbot interaction into a frustrating and often unsuccessful guessing game. Despite indications that customers would be open to collaboration, we know little about repair strategies that involve the customer and chatbot working together to resolve breakdowns. Our research addresses this gap by investigating the design and impact of collaborative repair strategies in customer-chatbot interaction. Drawing upon an integration of the theory of least collaborative effort with research on human-machine communication and customer service chatbots, we propose a novel repair strategy design; we instantiated it in the chatbot of a large insurance company and conducted a naturalistic summative evaluation through a randomized field experiment. Overall, our results suggest that a collaborative repair strategy can lead to more breakdowns being resolved and mitigate the negative impacts of breakdowns on key customer outcomes. Our research offers a new way of thinking about customer-AI service interactions by shifting the narrative from confrontation to collaboration, extends the theory of least collaborative effort by integrating the perspective of customer-chatbot interaction, and provides in-depth insights into breakdown and repair in real-world conversations between customers and chatbots.

Keywords: Customer service chatbot, conversational breakdown, repair strategy, service failure, human-AI interaction, theory of least collaborative effort, design science research, field experiment

[!小结]
问题：在客户与聊天机器人的交互中，如何设计一种协作式的修复策略，使双方共同解决对话中断问题，并检验其对服务结果的有效性？

Automatically Detecting Voice Phishing: A Large Audio Model Approach

https://doi.org/10.25300/MISQ/2025/19532

Phishing attacks remain one of the most prevalent and pervasive cybersecurity concerns. Voice phishing (i.e., vishing) is an emerging type of phishing attack where malicious actors use audio channels to steal sensitive information from victims. However, vishing detection is a challenging task due to its real-time nature and the limited availability of datasets. To help address the concern of vishing detection, this study proposes the vishing generative pretrained transformer (VishGPT). VishGPT adopts the computational design paradigm and incorporates novel reinforcement learning-based large language model fine-tuning and synthetic data model pretraining to automatically detect vishing attempts in real time. We evaluated VishGPT using a series of benchmark experiments, where we empirically demonstrated its improvement over state-of-the-art vishing detection and audio classification models. The results suggest that our proposed VishGPT achieved state-of-the-art performance in terms of accuracy (86.18%), precision (90.63%), recall (85.02%), and F1-score (87.74%). VishGPT offers practical value to cybersecurity professionals, end users, and academia. Additionally, VishGPT provides important design principles in the form of a custom proximal policy optimization (PPO) reward function and synthetic pretraining to the information systems knowledge base.

Keywords: Deep learning, voice phishing, computational design science, information systems, large language models, reinforcement learning, audio classification

[!小结]
问题：如何设计一个能够实时、准确地自动检测语音钓鱼攻击的深度学习模型（VishGPT），以应对其检测挑战？

Consumer Evaluation of Digital Device Innovations: Disentangling Effects of Novelty and Familiarity

https://doi.org/10.25300/MISQ/2025/18750

As digital devices increasingly integrate hardware and software features, firms must adopt innovation strategies that effectively balance novelty and familiarity to enhance consumer evaluation. While novel hardware components can introduce unique functionalities that attract consumers, excessive novelty may impede consumer acceptance. This study investigates how hardware component innovation strategies must navigate the delicate interplay between novelty and familiarity by examining two critical dimensions: the timing of hardware innovations (early vs. late) and the role of software-supported interaction (with vs. without related software support). By distinguishing between dominant design components familiar to consumers and non-dominant design components that are inherently unfamiliar, we uncover nuanced strategic insights. Our findings reveal that early introduction of dominant design innovations is crucial, and enhancing consumer interactions through software support significantly improves consumer satisfaction. Conversely, for non-dominant component innovations, a later market introduction proves more advantageous. Notably, software-supported interactions are less effective for these non-dominant innovations, as such support may inadvertently accentuate their unfamiliarity. These findings provide strategic guidance for smartphone manufacturers to leverage software-supported interactions and optimize the timing of hardware innovations to achieve an optimal balance between novelty and familiarity.

Keywords: Digital devices, component innovation, novelty, familiarity, dominant design, consumer evaluation

[!小结]
本研究聚焦数字设备硬件创新中“新颖性-熟悉度”的平衡难题，通过区分主导设计组件（消费者熟悉）与非主导设计组件（不熟悉），考察了创新时机（早/晚）和软件交互支持（有/无）的影响。研究发现，主导组件的创新宜早并辅以软件支持以提升满意度，而非主导组件则更适合晚期推出，且软件支持反而会凸显其陌生性而降低效果，为智能手机制造商提供了优化创新组合的策略指导。

Enhancing AI-Assisted Purchase Decisions: The Role of the Sense of Autonomy

https://doi.org/10.25300/MISQ/2025/17607

Empowered by large-scale consumer data, artificial intelligence (AI) systems act as shopping gurus, serving up highly personalized and expertly curated product recommendations for consumers. Despite their proficiency at inferring what could be the commonly “optimal” choices for the average consumer, AI systems have a limited ability to account for the idiosyncratic factors that uniquely shape each consumer’s purchase decisions, such as one’s hidden motives or complex situations. This inherent limitation, termed AI’s uniqueness neglect, poses a key challenge to the value of AI-advised decisions. Our research identifies a heightened sense of autonomy as a crucial means to improve the quality of AI-advised decisions through compensating for AI’s uniqueness neglect. Across five laboratory experiments and one field experiment in the context of apparel purchases, we show that fostering the sense of autonomy enhances purchase intention and the quality of purchase decisions. We also delineate the underlying mechanism and demonstrate a managerially actionable solution to effectively promote the sense of autonomy. Our work contributes to the literature by uncovering a distinct mechanism by which the sense of autonomy improves the quality of AI-advised decisions and by offering a feasible design strategy that leverages personal smartphones to address the uniqueness neglect of AI recommendation systems. Field data on actual product purchases and returns confirm the efficacy of our AI design in a real-world setting. Our findings offer both theoretical and managerial implications for a wide range of AI-aided decision-making contexts where idiosyncratic factors are important to individual human decision makers yet are commonly overlooked by today’s AI systems.

Keywords: Artificial intelligence, uniqueness neglect of AI, sense of autonomy, AI-assisted decision-making, decision quality, purchase decision, product return

[!小结]
问题：如何通过增强消费者的自主感，来弥补AI推荐系统对个人独特性因素的忽视，进而提高AI辅助决策的质量？
本文揭示了人工智能推荐系统因无法考虑消费者隐藏动机或复杂情境等独特因素而存在“独特性忽视”问题，并通过五项实验室实验和一项服装购买实地实验证明，增强消费者的自主感能有效补偿这一缺陷，从而提升购买意愿和决策质量。研究还提出了利用个人智能手机来促进自主感的可行设计策略，并在实际产品购买和退货数据中验证了其有效性。

The Myth of Good Data: An Ecosystem Perspective on Data Integrity Breakdown and Rehabilitation

https://doi.org/10.25300/MISQ/2025/17587

In recent years, a huge amount of excitement has surrounded the potential to transform organizations and institutions through evidence-based or data-driven decision-making. However, such promises are often premised on a stable view of data quality. In this paper, we take a practice perspective on data integrity, asking how attempts to leverage data at one point in a data ecosystem can ripple unexpectedly, triggering a breakdown in the integrity of data resources at another point in the ecosystem. We further outline the intensive efforts involved in rehabilitating data when integrity is undermined. Through a multi-sited ethnographic study in multiple healthcare organizations, we trace four empirical examples of situated data integrity breakdown (negotiated category breakdown, category expansion breakdown, data mismatch breakdown, and data disaggregation breakdown), linking the triggers, responses, and rehabilitation efforts in each case. We argue that organizations need to take the ongoing work of retraining personnel, modifying IT systems, and/or shifting data collection practices into account to rehabilitate data whose integrity will be inevitably undermined as a result of shifting needs, goals, expectations, and/or IT systems across a connected data ecosystem. In so doing, we offer insights into the often unpredictable impacts of attempts to leverage data, thereby enriching the discourse on data-driven decision-making and its implications for organizational strategy and practice.

Keywords: Data-driven decision-making, data analytics, data management, data integrity, data quality, data work, data rehabilitation, ethnography

[!小结]
问题：在互联的数据生态系统中，利用数据的尝试如何意外引发其他环节数据完整性的崩溃，而组织又该如何修复这种动态变化中必然受损的数据完整性？

Leveraging Network Effects for Connected Products: Strategies and Implications for the Value Chain

https://doi.org/10.25300/MISQ/2025/18441

This paper examines how a value chain can strategically harness network effects in connected products. We consider a representative two-tier value chain, in which a manufacturer sells a connected product through a retailer. Consumers derive both a stand-alone value and a network benefit, which increases with the size of the user base and the strength of the network effect. We investigate two strategies to effectively leverage these network effects. First, we analyze network expansion through seeding, which involves providing the product for free to a targeted group of consumers. Our analysis shows that when the production cost is sufficiently low, seeding arises in equilibrium and exhibits a “bang-bang” pattern, with either the manufacturer or the retailer exclusively undertaking seeding. Under strong network effects, the manufacturer can induce retailer seeding through a quantity-forcing contract with a wholesale discount, thereby replicating the centralized outcome in which both firms are integrated and act as a central planner. Next, we explore engineering network effects through investment. The decentralized value chain results in less efficient investment than the centralized one, leading to lower social welfare. However, when both strategies are jointly employed, we uncover a striking result: The decentralized value chain can yield higher social welfare than the centralized one. This occurs because the central planner underinvests in network effects, as investment generates a positive externality for seeded users, whereas the manufacturer may overinvest to reduce the wholesale discount, potentially resulting in greater social welfare.

Keywords: Connected product, value chain, seeding, engineering network effects, pricing

[!小结]
问题：在两级价值链中，制造商和零售商应如何策略性地结合网络扩张（播种）与网络效应投资，以最优利用联网产品的网络效应，并比较分散与集中决策下的社会效益？
本文研究一个由制造商和零售商组成的两级价值链如何策略性地利用联网产品中的网络效应，分析了网络扩张（通过向目标用户免费“播种”）和网络效应投资两种策略。研究发现，在低成本条件下，播种会呈现制造商或零售商“二选一”的均衡模式，且强网络效应下制造商可通过数量强制合约与批发折扣诱导零售商播种；而单独投资时分散价值链效率更低，但两种策略联合时，分散价值链反而可能因集中计划者对网络效应投资不足而产生更高的社会福利。

Rewards or Upgrades? Incentive Designs in Referral Programs

https://doi.org/10.25300/MISQ/2025/19540

Referral programs are widely used for customer acquisition. Traditionally, these programs adopt a referral-reward approach, offering monetary incentives for successful referrals. However, many firms, especially in digital industries, have embraced an alternative—referral-upgrade programs that reward successful referrers with product upgrades such as premium features or enhanced services. Despite their growing use, little is known about when such programs can outperform the traditional referral-reward approach. This paper develops a stylized model to compare these two referral mechanisms. We address two questions: (1) when referral-upgrade programs are more (or less) profitable than referral-reward programs, and (2) how key factors such as referral costs and referral reachability, captured by multiple referrals and the degree of overlap among referred friends, affect firms’ decisions and customer behaviors. We find that referral-upgrade programs are generally more profitable than referral-reward programs, and this profit dominance remains robust to various extensions such as marginal costs or boundedly rational customers but may reverse when referral costs are correlated with customer valuations or when referral reachability expands via multiple referrals. Our findings offer managerial insights into when firms should adopt upgrade-based incentives to manage their referral programs.

Keywords: Referral-upgrade, referral-reward, product line design, versioning, pricing

[!小结]
问题：在什么条件下，推荐升级计划比传统的推荐奖励计划能为企业带来更高利润，以及推荐成本、多次推荐和好友圈重叠等关键因素如何影响这一比较结果？
通过构建理论模型，系统比较了传统推荐奖励（现金激励）与推荐升级（产品功能或服务升级）两种客户推荐机制的盈利能力，发现升级计划通常更优，且该优势在多种扩展条件下稳健，但在推荐成本与客户估值相关或推荐可及性（如多次推荐、好友圈重叠）扩大时可能逆转，为企业选择激励类型提供了管理启示。

Autistic Traits, Theory of Mind, and Implications for IT Professionals’ Emotional Labor and Mental Well-Being

https://doi.org/10.25300/MISQ/2025/19395

Departing from the situation-centric approach common in emotional labor research, we adopt a trait-based perspective to examine how autistic traits shape IT professionals’ experiences of social demands and mental health. Across three quantitative studies—two with IT professionals and one with college students preparing for IT careers—we found that autistic traits are consistently central to IT identities. IT professionals as a group are likely underdiagnosed and, on average, exhibit elevated autistic traits, which, in turn, can heighten their emotional labor and erode mental well-being. Compared with the general population, IT professionals, on average, report substantially higher levels of anxiety and depression, with prevalence rates more than 2.3 times the norm, underscoring the potential mental health costs of sustained emotional labor. These findings suggest that emotional labor is not merely situational but anchored in dispositional factors that are widespread within the IT profession. Results from the Reading the Mind in the Eyes Test further show that theory-of-mind processing is more laborious for IT students than for business management students, thus providing a socio-cognitive explanation for many IT professionals’ social challenges. Taken together, these findings advance a disposition-centric paradigm that expands the nomological net of emotional labor and calls for a reexamination of long-standing assumptions about IT work, IT worker identities, and the broader discourse on neurodiversity in technical professions. This work challenges scholars and practitioners alike to embrace autistic traits as an integral and foundational element of IT professional identity and to reimagine IT management as managing neurodiversity.

Keywords: Autism, autistic traits, neurodiversity, theory of mind, emotional labor, well-being, mental health, empathizing-systemizing theory, Reading the Mind in the Eyes Test, AQ

[!小结]
问题：IT专业人士的自闭特质如何影响其情绪劳动与心理健康，以及这一特质视角对理解IT工作者身份和管理神经多样性有何意义？
本文采用特质视角而非传统的情境中心视角，通过三项定量研究发现，自闭特质在IT专业人士身份中居于核心地位，该群体可能普遍存在未被充分诊断的自闭特质，这些特质会加剧其情绪劳动并严重损害心理健康（焦虑抑郁患病率超常模2.3倍）。研究还通过心理理论测试提供了社会认知解释，最终倡导将神经多样性视为IT专业身份的基础要素，并重新构想IT管理。

Latent Similarity-Enhanced Credit Risk Prediction

https://doi.org/10.25300/MISQ/2025/18080

Given the sheer size of the consumer credit market and the huge number of consumer credit users, credit risk prediction, or predicting the probability of consumer credit delinquency (or default), has become a critical problem in the consumer credit industry. Effective credit risk prediction aids financial institutions in granting and managing extensions of credit and can help secure the availability of credit for worthy applicants. While it is desirable to employ both users’ intrinsic characteristics and similarities among them for effective credit risk prediction, existing studies rely solely on similarities derived from their observed characteristics and fail to account for unobserved similarities among them. To address this challenge, we propose a latent similarity-enhanced credit risk prediction model, which operationalizes the similarity between a pair of users as a combination of the observed and latent similarities between them. We then present a new design for a new method that estimates the model parameters, learns latent similarities among users, and integrates both observed and latent similarities among users with their intrinsic characteristics for credit risk prediction. We further extend our method to the multiclass and numerical credit risk prediction problems. Extensive empirical evaluations with real-world data demonstrate the superior predictive power of our method over benchmark methods for a broad spectrum of credit risk prediction problems. We also show substantial economic value generated from the superiority of our method through a case study.

Keywords: Credit risk prediction, latent similarity, machine learning, information systems: enabling technologies

[!小结]
问题：如何在信用风险预测中，结合用户的显性特征与用户间未观测到的潜在相似性，以提升对消费信贷违约概率的预测准确性？
针对消费信贷风险预测中现有方法仅依赖用户可观测特征而忽略其潜在相似性的问题，提出了一种“潜在相似性增强”的预测模型，该模型将用户间相似性表示为观测相似性与潜在相似性的组合，并设计了相应的参数估计与学习算法。该方法进一步拓展至多分类和数值型预测场景，在真实数据集上的实证评估表明其预测能力显著优于基准方法，并通过案例研究证实了其巨大的经济价值。

Learning by Phishing via Post-Simulation Feedback: From Embedded to Non-Embedded Training

https://doi.org/10.25300/MISQ/2025/19354

Given the frequent occurrence of phishing attacks and their devastating consequences, organizations are increasingly deploying phishing simulation emails to quantify employee susceptibility (e.g., clicking within-email links) and investing in training programs to reduce such susceptibility. Interestingly, phishing simulations can also be turned into a training opportunity in themselves. A best practice in industry is “embedded training”—providing immediate feedback on landing pages to employees who fail the simulations. This intervention is intuitively appealing given its “just-in-time” nature. Although laboratory studies from the literature have offered broad support for its effectiveness in reducing employee susceptibility, studies conducted in field settings have observed weaker evidence or even a reversed effect that increased susceptibility. In this research, we recognize an inherent shortcoming of the real-world implementation of embedded training: limited reach. To address this practical challenge, we propose an alternative, novel intervention—“non-embedded training”—that decouples feedback from the failure action and sends delayed feedback to all the employees. Following an “empirics-first” approach, we conducted three randomized field experiments using a leading phishing simulation platform to explore the respective and combined effects of embedded and non-embedded training in reducing user vulnerability over time. This research contributes to the practice and literature on phishing and cybersecurity by challenging the assumed effectiveness of embedded training in practice and revealing how non-embedded training could be a more promising intervention.

Keywords: Phishing attacks, social engineering, cybersecurity, phishing simulations, embedded training, non-embedded training, timing, exposure

[!小结]
问题：在真实工作环境中，如何设计更有效的网络钓鱼模拟培训干预措施，以降低员工对钓鱼攻击的长期脆弱性，特别是非嵌入式培训是否比广泛使用的嵌入式培训效果更好？
针对组织网络安全培训中广泛采用的“嵌入式培训”（对点击模拟钓鱼邮件失败者即时反馈）在实际应用中的局限性（覆盖率不足），提出了“非嵌入式培训”这一新干预措施（将反馈与失败行为解耦并向所有员工发送延迟反馈）。通过基于领先模拟平台的三项随机实地实验，研究发现非嵌入式培训在降低员工长期脆弱性方面可能比嵌入式培训更有效，从而挑战了后者的实践有效性假设。

https://doi.org/10.25300/MISQ/2025/18022

Engaging in side projects outside of regular employment has become a growing trend among knowledge workers, particularly information technology (IT) professionals. Side projects offer valuable opportunities for employees to learn new skills and foster creativity. However, the legal ownership of side projects remains uncertain, raising questions about how this affects employee innovation at their primary jobs, which we refer to as “employee innovation at work.” In this study, we leverage an exogenous change in the legal arrangement of side project ownership—the Alcatel v. Brown case—to investigate how firms’ enhanced control over side projects influences employees’ innovation performance at work. We find that in states where firms gained greater contractual authority to claim ownership of employees’ side projects, the number of IT patents owned by firms decreased. However, paradoxically, the quality of these patents, as measured by the number of forward citations, improved following the legal change. Further analyses of the underlying mechanisms suggest that these contrasting findings likely stem from shifts in both employee innovation behaviors and firms’ innovation strategies, post-Alcatel v. Brown. Our findings contribute to the information systems literature by highlighting the nuanced effects of side project ownership on IT innovation.

Keywords: Employee innovation at work, side project, IT innovation, intellectual property, Alcatel v. Brown

[!小结]
问题：企业加强对员工副项目所有权的控制，如何影响IT员工在主职工作中的创新绩效（数量与质量）？
本研究利用Alcatel v. Brown案这一外生法律冲击，考察了企业加强对员工副项目控制权后对IT创新的影响。研究发现，在赋予企业更大合同权力主张副项目所有权的州，企业拥有的IT专利数量减少，但专利质量（以正向引用数衡量）反而提高。这种看似矛盾的结果可能源于法律变化后员工创新行为与企业创新策略的共同调整。

The Digital Privacy Paradox and Choice Architecture: Evidence from an Experiment in Fintech

https://doi.org/10.25300/MISQ/2025/19206

“Notice and choice” is a mainstay of policies designed to safeguard consumer privacy. This paper investigates distortions in consumer behavior when faced with notice and choice, which may limit the ability of consumers to safeguard their privacy. We used data derived from a field experiment at MIT that distributed a new product, Bitcoin, to all 4,494 undergraduate students. There are two primary findings. First, small navigation costs have a tangible effect on how privacy-protective consumers’ choices are, often in sharp contrast with individual stated preferences about privacy. Second, the introduction of irrelevant but reassuring information about privacy protection makes consumers less likely to avoid surveillance, regardless of their stated preferences toward privacy.

Keywords: Privacy, privacy paradox, digital currency, digital wallets, field experiment

[!小结]
问题：在现有的“通知与选择”隐私政策框架下，消费者的实际隐私保护行为是如何被环境因素（如导航成本和信息呈现）所系统性扭曲的？
基于麻省理工学院向所有本科生分发比特币的实地实验数据，揭示了在“通知与选择”隐私保护框架下，消费者行为的两种显著扭曲：第一，微小的导航成本会切实改变消费者出于隐私保护的选择，并可能与其事先陈述的隐私偏好相悖；第二，引入不相关但看似安心的隐私保护信息，会降低消费者回避监控的可能性，无论其隐私偏好如何。