Previous research in multiple judgment domains has found that nonlinear functionsare typically processed less accurately than linear ones. This empiricalregularity has potential implications for consumer choice, given that nonlinearfunctions (e.g., diminishing returns) are commonplace. In two experimentalstudies we measured precision and bias in consumers’ ability to identifysurpluses when returns to product attributes were nonlinear. We hypothesizedthat nonlinear functions would reduce precision and induce bias towardlinearization of nonlinear relationships. Neither hypothesis was supported formonotonic nonlinearities. However, precision was greatly reduced for productswith nonmonotonic attributes. Moreover, assessments of surplus weresystematically and strongly biased, regardless of the shape of returns anddespite feedback and incentives. The findings imply that consumers use aflexible but coarse mechanism to compare attributes against prices, withimplications for the prevalence of costly mistakes.
