Visual Center Bias in Consumer Choices
Visual Center Bias in Consumer Choices
Consumer decision making often critically rely on online product ratings. In estimating the impact of online ratings on sales, researchers consider various descriptive statistics of rating distributions, such as the mean, variance, skewness, kurtosis, and even entropy and the Herfindahl-Hirschman Index. However, real-world consumer decisions are derived from visual perceptions about displayed rating distributions in the form of histograms. In this paper, we propose and identify a decision bias, the visual center bias (VCB), a consumer tendency to pay too much attention to salient features of distributions in visual decision making. In a series of experiments, we identify VCB's significant impact on consumer preference. We show that with VCB, consumers make choices that violate widely accepted decision rules. In our experiments, subjects are observed to prefer products with lower average ratings and higher rating variances. They may even prefer a first-order stochastically dominated distribution as a result of the VCB. Our study suggests how consumer choice models should be refined to accommodate such visual decision biases.
More details on Prof. Dr. Michael Zhang can be found at: http://mikezhang.com/#about.