Despite the growing awareness surrounding the issue of fake online reviews, a new study conducted by the University of South Florida (USF) has uncovered a concerning trend: consumers continue to be significantly swayed by what they read online, even when there are ample reasons for scepticism. This work, published in the prestigious journal Information Systems Research under the title “The Illusion of Authenticity in Online Reviews: Truth Bias and the Role of Valence,” delves deep into why people remain so susceptible to falsehoods on the internet. It addresses a pivotal question at the heart of the modern digital marketplace: do consumers naturally adopt a suspicious stance when evaluating online reviews, or do they have an ingrained tendency to trust these narratives?

At the heart of the study lies the “truth bias,” a term from psychology that encapsulates the human tendency to accept information as truthful unless there is a compelling reason to doubt it. According to the study’s co-author, Dezhi Yin, an associate professor at the USF Muma College of Business, this tendency plays a critical role in how online reviews influence consumer choices. “Our research is among the first to examine how consumers make real or fake judgments of online reviews,” Yin explained. “Understanding the consumer mindset is crucial, as it is ultimately consumers who are the primary targets of review manipulation.”

The study, carried out in collaboration with Samuel D. Bond of the Georgia Institute of Technology and Han Zhang, affiliated with Georgia Tech and Hong Kong Baptist University, involved five experimental investigations conducted between 2018 and 2023. In these experiments, participants were presented with reviews and asked to determine whether each was genuine or fake. Intriguingly, even when participants were informed ahead of time that half of the reviews they would encounter were fabricated, they consistently judged most reviews to be authentic, illustrating the stubborn power of the truth bias.

One striking example from the research involved participants being shown twenty restaurant reviews while knowing only ten were real. All reviews were displayed together, enabling participants to navigate back and forth to re-evaluate and refine their judgments. Despite this, on average, participants identified 11.38 reviews as real, suggesting that cross-checking did little to overcome the instinctive assumption of honesty. Yin observed, “This illustrates the power of truth bias in this context,” underscoring how deeply ingrained the tendency to trust reviews can be, even in the face of explicit warnings.

Another dimension explored by the researchers was the role of a review’s tone or “valence” — whether a review was positive or negative — in shaping perceptions of authenticity. Real-world data across multiple platforms have consistently shown that negative reviews are more likely to be fake than positive ones. However, contrary to this reality, the study participants were found to be more inclined to believe negative reviews over positive ones. Yin remarked, “Our findings suggest a striking contrast between reality and perception,” highlighting a troubling gap that can leave consumers vulnerable to manipulation by those who exploit this imbalance.

The broader implications of these findings for online marketplaces and consumer platforms are significant. The researchers argue that relying on users to identify and report suspicious reviews is not a reliable method to weed out deceptive content. Instead, they advocate for more robust interventions, such as prioritising identifying and suppressing fake negative reviews and implementing interface changes that help consumers better evaluate the credibility of what they read. Yin and his colleagues hope this study will inspire further research at the intersection of deception, psychology, and consumer behaviour — and ultimately lead to a digital environment in which consumers can make better-informed choices without being misled by the ever-present spectre of fake reviews.

More information: Dezhi Yin et al, The Illusion of Authenticity in Online Reviews: Truth Bias and the Role of Valence, Information Systems Research. DOI: 10.1287/isre.2023.0339

Journal information: Information Systems Research Provided by University of South Florida