In the contemporary digital marketplace, consumer ratings play a pivotal role in shaping purchase decisions, serving as one of the most influential indicators alongside price and brand reputation. A groundbreaking study from the Cornell SC Johnson College of Business now reveals that the format in which these ratings are displayed—whether through familiar star icons or traditional numerical values—can drastically alter consumer perception. This discrepancy not only highlights significant cognitive biases but also raises pressing concerns about how rating systems can mislead consumers and businesses alike.
The researchers embarked on a comprehensive inquiry involving six controlled experiments designed to examine how fractional ratings are interpreted when presented as stars compared to Arabic numerals. Their findings indicate a consistent tendency among consumers to perceive fractional star ratings as higher than their actual numerical value, whereas fractional numerals are routinely undervalued. For instance, a rating of 3.5 stars is often cognitively rounded up to a 4 in the consumer’s mind, while a numerical 3.5 is mentally anchored at 3. This phenomenon underscores a vital difference in how the human brain processes graphical imagery versus numeric symbols.
This divergence stems from the differing cognitive pathways activated during visual and numerical processing. According to Deepak Sirwani, the first author of the study and now an assistant professor at the University of British Columbia, the brain’s interpretation of stars involves a gestalt completion mechanism. When consumers see three full stars followed by a half star, their neural circuits naturally fill in the incomplete visual element, giving rise to an inflated representation of the rating’s magnitude. Conversely, when confronted with the text-based notation ‘3.5,’ the brain tends to emphasize the integer component (‘3’), causing the fractional element to be psychologically diminished.
The implications of these findings extend deeply into the fields of marketing and consumer psychology. Manoj Thomas, a management professor at Cornell University and co-author of this research, elaborates that this difference in mental representation constitutes what they describe as an “a-ha” moment. He explains that the striking contrast in cognitive activation between processing images—specifically stars—and Arabic numerals is an overlooked but critical factor in understanding consumer behavior. This insight invites a reevaluation of how rating information should be standardized and presented to avoid consumer misinterpretations.
A key experiment illustrating these effects involved 616 participants who were shown a series of ratings spanning from 1 to 5, increasing by increments of 0.25. Participants were randomly assigned to conditions wherein they viewed either star-based symbols, numeral-based ratings, or a combination of both simultaneously. Subsequently, they were asked to estimate the position of each rating on a hypothetical, unlabeled continuum between 1 and 5. Results demonstrated a systematic overestimation of star ratings and underestimation of numerals, confirming the robustness of this cognitive bias across a broad sample population.
This discrepancy has material consequences for commercial enterprises. When fractional star ratings are perceived as higher than they truly are, companies may inadvertently set consumer expectations beyond what their products can reliably deliver. On the other hand, products rated with fractional numbers may suffer from suppressed consumer enthusiasm due to underestimation of their quality. Such misalignments between perception and reality can result in either customer dissatisfaction or lost sales, motivating the researchers to call for the development of new industry standards that better calibrate rating presentations to actual consumer cognition.
Moreover, the economic reverberations of this perceptual bias are profound. Previous studies have established that slight increments in ratings—even as small as 0.2 points—can trigger sales increases up to 300%. Considering this, the current research posits that modifying only the format—from numerals to stars—might amplify sales by an order of magnitude. This raises ethical and strategic questions about the deployment of rating formats, as businesses may exploit star ratings to inflate perceived value, potentially leading to market distortions and consumer distrust.
From a neuroscience perspective, these findings open the door to exploring how distinct areas of the brain respond differently to symbolic visual stimuli versus numeric representations. The concept of ‘completion’ in visual cognition suggests that our neural processing favors holistic image interpretation, facilitating the upward adjustment of star ratings. In contrast, the parsing of Arabic numerals engages analytical processes that prioritize literal and sequential interpretation. This bifurcation reflects broader themes in cognitive science regarding dual processing modes—intuitive versus analytical—and their real-world applications.
The study contributes meaningfully to the domain of behavioral economics, particularly relating to decision-making heuristics and biases. It exemplifies how subtle variations in information presentation can significantly influence consumer judgments and, ultimately, economic outcomes. Understanding these dynamics is essential for policymakers, regulators, and businesses aiming to design fair and transparent rating systems that align better with objective product performance and user satisfaction.
As digital platforms proliferate and customer feedback becomes increasingly central to the reputation economy, the stakes for accuracy and clarity in ratings are escalating. The Cornell researchers’ advocacy for standardized rating formats is not merely a technical recommendation but a call to recognize that human perception is anything but neutral or uniform. Accounting for cognitive biases inherent in processing visual versus numerical information is crucial for maintaining trust and efficacy in market signaling mechanisms.
In conclusion, this investigation underscores a fundamental, yet overlooked aspect of consumer psychology: the medium of information delivery profoundly impacts how that information is cognitively processed and valued. In the age of e-commerce and online reviews, where split-second decisions can hinge on rating interpretations, appreciating this phenomenon offers new avenues for enhancing consumer experience and business integrity. As markets evolve, adapting rating systems to human cognitive architecture will be essential in ensuring that ratings fulfill their intended role as accurate and reliable predictors of product quality.
Subject of Research: Consumer perception of product ratings; cognitive processing differences between star versus numeric rating formats.
Article Title: Overestimating Stars, Underestimating Numbers: The Hidden Impact of Rating Formats
News Publication Date: 15-May-2025
Web References:
- Journal article: https://journals.sagepub.com/doi/10.1177/00222437251322425
- Cornell Chronicle story: https://news.cornell.edu/stories/2025/05/stars-or-numerals-how-rating-formats-change-consumer-behavior
References:
Sirwani, D., Thomas, M., et al. (2025). Overestimating Stars, Underestimating Numbers: The Hidden Impact of Rating Formats. Journal of Marketing Research. DOI: 10.1177/00222437251322425
Keywords: Behavioral economics, consumer psychology, rating formats, e-commerce, cognitive bias, marketing research, decision making, neuroscience, product evaluation, digital marketplaces, visual cognition, numerical cognition
