Sally Shin

Shin, Sally MJ, Nirajana Mishra, and Ravi Dhar, “From Lemons to LemonAIde: How Low Brand Trust Mitigates Aversion to Artificial Intelligence,” Revising for 2nd round review at Journal of Consumer Research

As brands increasingly introduce artificial intelligence (AI) into customer-facing interactions, understanding how key brand characteristics influence consumer preferences for AI relative to human agents becomes critical. This research investigates how a brand’s trustworthiness affects preferences for product recommendations curated by an AI relative to a human recommender. Across eight preregistered experiments, consumers are more willing to receive recommendations from (i.e., exhibit lower aversion to) an AI agent developed by a brand that is perceived as low-trust (vs. high-trust). This effect is robust across diverse industries, trust signals, populations, and measurement methods. The observed mitigation of AI aversion is driven by a perceived asymmetry in essence contagion: consumers believe that AI agents embody less of the brand’s essence—its (un)trustworthiness—than human agents do. Providing support for the process via moderation, we find that stronger beliefs in this differential essence contagion amplify the impact of trust on AI preference. Furthermore, we differentiate trust from other positive brand perception (i.e., liking). While trust, an essentialist attribute core to a brand’s identity, mitigates AI aversion, liking, a less essentialist attribute, does not. These findings offer theoretically novel contributions and actionable insights into how brand trust influences consumer reactions to AI-driven services in marketing.

Shin, Sally MJ, Paul Rozin, and Gal Zauberman, “When Cherished Memories are Threatened,” Under review at Psychological Science

People often hold cherished memories that are later threatened by unsettling revelations. How do individuals reconcile the joy they once felt with the knowledge they now hold? The fact that they once held favorable sentiments is part of their personal history, but can they continue to extract positive utility from revisiting their pasts in the wake of such revelations? This research focuses on one such instance: when positive memories of a creator (e.g., musician) and the experiences associated with their creations (e.g., listening to their music) are challenged by negative information about the creator. Six preregistered experiments reveal a robust pattern of selective memory preservation. When revisiting their pasts, individuals compartmentalize the creation from its creator, preserving the positivity of creation memories more than that of the creator memories. Unlike this dissociation in memory, the revelation diminishes how positively they currently evaluate both the creation and the creator.

Shin, Sally MJ, and Alexander G. Fulmer, “Single Farm-to-Table: Preference for Foods with a Single Source of Origin,” Under review at Journal of Marketing Research

While food and beverage companies frequently communicate origin-related messages about their products, their effectiveness remains largely unknown. The current research investigates the impact of a product’s origin numerosity—single versus multiple sources from which it is derived—on consumer preference. Eleven experiments reveal that consumers favor products sourced from a single origin over otherwise identical products sourced from multiple origins. This preference extends beyond categories traditionally favoring single origin (e.g., wine and whiskey) to those where such norms are less established (e.g., hot sauce, olive oil, cheese, strawberry jam, and peanut butter). Consumer preference for single origin products persists across purchase intentions, incentive-compatible choice, and even when a multi-origin product is rated as higher quality. This effect is driven by the perception that single origin products are purer, and consequently, healthier. Consistent with this framework, information that enhances the perceived purity or perceived healthiness of multi-origin products attenuates the preference for single origin numerosity, offering theory-driven and actionable solutions to benefit producers of multi-origin products. These findings offer theoretically rich and practically relevant insights into origin-related marketing communications.

Shin, Sally MJ, and Gal Zauberman, “Maximizing Return on Experiences (ROE): Effect of Memory Reminder Modes”

Whether through iPhone’s “You have a new memory” or Facebook’s “On This Day” notifications, marketers increasingly prompt consumers to revisit past experiences through old photos or ‘status updates.’ Yet little is known about how these different reminder modes influence consumers’ judgments and behavioral intentions. Evidence from longitudinal preregistered experiments suggests that photo-based reminders enhance remembered enjoyment more than text-based reminders or free recall. Notably, photo reminders increase enjoyment even when they are not perceived to fully capture the experience, suggesting that memory cues need not faithfully represent the original experience to boost well-being. Meaningful downstream consequences, such as willingness to recommend or anticipation of future similar experiences, align more with remembered enjoyment than with perceived comprehensiveness of representation. Using content analyses of open-ended responses and follow-up experiments, we examine underlying mechanisms, such as cue vividness and processing fluency. Our findings provide insights into how firms can design post-visit or post-purchase communications to improve consumer satisfaction and drive behavioral intentions.

Shin, Sally MJ, Mohin Banker, and Gal Zauberman, “Power of One: Preference for a Recommender’s Lived Experience Over the Aggregation of Others,” Data collection completed, Manuscript in preparation 

 When deciding whose recommendation to follow, consumers are often faced with a tradeoff between the number of information sources (one vs. many) and the degree of experiential proximity (direct vs. indirect experience). For example, consider a podcast host saying, “I spoke with three people who recommend Product A” (which we refer to as crowdsourcing) versus a single individual directly sharing their own recommendation with the listeners for Product B (which we refer to as livedsourcing). While multiple experiences may be more informative than a single experience, and indirect summaries may be more comprehensive and unbiased, we find, across seven preregistered experiments spanning various decision-making contexts, that consumers prefer recommendations based on a single individual’s lived experience. Content analyses of open-ended responses and follow-up experiments reveal that this preference is driven by the belief that a lived experience is more credible and more representative of their own tastes than crowdsourced experiences. This work highlights that consumers are surprisingly agnostic to the number of information sources but highly sensitive to the directness of the experience behind a recommendation. We conclude with implications for marketers and retailers on the use of consumer testimonials and product endorsements.

Mishra, Nirajana, Sally MJ Shin, and Ravi Dhar, “The More Information = More Trust Lay Belief,” Data collection completed, Manuscript in preparation 

Companies often ask consumers to provide information before using an app. While managerial intuition favors asking for relatively less information to reduce friction during onboarding to enhance downloading, this view overlooks the crucial role of trust in consumer decisions. Across ten experiments, we document a robust and counterintuitive phenomenon: consumers perceive apps that ask for more (vs. less) information as more trustworthy, a lay belief we term “more information = more trust.” This lay belief arises from a cognitive association between information-sharing and trust, whereby a greater information request serves as a trust signal that leads consumers to use an app that asks for more information over an otherwise similar app that asks for less. We explore the boundaries of this lay belief and find that it is attenuated when (a) consumers perceive the requested information as irrelevant to the app’s functionality, (b) a more explicit and diagnostic alternative cue is available to assess the app’s trustworthiness, and (c) consumers are made aware that the company intends to share their data with third parties through privacy notices. We discuss the implications of our findings on trust, information-sharing, privacy, and app design.