Selected Abstracts
How Consumers Photograph Products for Positive versus Negative Reviews
Jiani Xue and Shiri Melumad
Recent years have witnessed a rapid growth in the inclusion of user-generated pictures inonline reviews. While the presence (vs. absence) of found to increase the helpfulness of reviews, less is known about the process images has been images that underlies the creation of such or its downstream consequences. This paper explores how an important factor—the valence of reviewers’ a product—affects the aesthetic taken of it and, ultimately, how reviewers quality of the photos are to observers. A theory is proposed arguing that put more effort into crafting well-composed photos of products they like (vs. dislike), resulting in images of higher aesthetic quality. These higher-quality images, in turn, are seen as more helpful in conveying information about the product, both because they are more fluent to process, and because observers infer that the reviewer put more work into creating the images, thereby engendering trust in that reviewer. Results from five field, experimental, and observational lab studies (N = 380,999) provide convergent support for these predictions.
Wisdom of the Algorithmic Crowd: Encouraging the Adoption of Ensemble Models by Leveraging Intuitions of Crowd Wisdom
Jiani Xue, Stefano Puntoni and Barbara A. Mellers
Ensemble models, a class of machine learning algorithms that combine the predictions of multiple algorithms to form more accurate predictions, are widely used in marketing applications. This research explores ways to enhance the perceived accuracy of these models. Thirteen experimental studies (ten in the main paper and three in the Web Appendix) collectively demonstrate that consumers have an intuitive grasp of the “wisdom of the crowd,” and they perceive machine ensembles as more accurate than single algorithms. Framing algorithms as a collection of models rather than a single algorithm boosts perceived accuracy, purchase intent, and consumer preferences across different domains. Results also indicate that preferences are driven by a mistaken belief that ensembles primarily reduce bias rather than noise. Consumers seem to recognize that the diversity and relevance of the datasets on which ensembles are trained influence bias reduction. Taken together, these experiments shed light on a novel psychological process shaping AI adoption and reveal practical ways to promote the adoption of ensemble methods.
If There is a Fly in My Soup, Does Anyone Care? The Emotional Perception Bias in Online Communications
Jiani Xue and Maurice E. Schweitzer
Consumer reviews profoundly guide purchase decisions. How consumers actually react to reviews, however, can be very different from how reviewers expect readers to react. We focus on the communication of emotional experiences, a very common feature of reviews (Pilot Study of Yelp reviews). Across six preregistered studies, we document a robust Emotion Perception Bias: Reviewers consistently overestimate the emotional responses of readers who read their reviews. This is true when reviewers communicate disgust, anger, and happiness, and it is true for both restaurant and hotel reviews (Study 1A – 3) as well as reviews of a movie clip (Study 4). This bias is not moderated by the intensity of the emotions expressed (Study 2). Importantly, communicating emotional content can harm perceptions of the reviewer; reviewers who express anger are perceived to be less credible (Study 2) and to have overreacted (Studies 3 and 4) compared to how reviewers perceive themselves. We show that the Emotion Perception Bias reflects a difference in attributions; reviewers attribute the emotional experience to the situation, but readers attribute part of the emotional experience to the reviewer themselves (Study 3). Finally, we show that experience can mitigate this bias (Study 4). Taken together, our findings underscore the challenge of communicating emotional experiences.