Why contextual preference reversals maximize expected value.

Abstract

Contextual preference reversals occur when a preference for one option over another is reversed by the addition of further options. It has been argued that the occurrence of preference reversals in human behavior shows that people violate the axioms of rational choice and that people are not, therefore, expected value maximizers. In contrast, we demonstrate that if a person is only able to make noisy calculations of expected value and noisy observations of the ordinal relations among option features, then the expected value maximizing choice is influenced by the addition of new options and does give rise to apparent preference reversals. We explore the implications of expected value maximizing choice, conditioned on noisy observations, for a range of contextual preference reversal types—including attraction, compromise, similarity, and phantom effects. These preference reversal types have played a key role in the development of models of human choice. We conclude that experiments demonstrating contextual preference reversals are not evidence for irrationality. They are, however, a consequence of expected value maximization given noisy observations.

Optimal policy for multi-alternative decisions

“…Some studies suggest that apparently irrational human behavior could be accounted for by heuristic weighting rules for features that integrate feature valences through feedforward 26,37,38 or recurrent 39,40 neural interactions. Interestingly, a recent study reported that a contextdependent feature weighting can increase the robustness of value encoding to neural noise in later processing stages 38,49 , whereas another recent study provided a unified adaptive gain-control model that produces context-dependent behavioral biases 50 . However, to our knowledge, the optimal policy for these more complex models where the value function is computed by combining multiple features, presented sequentially, remains unknown.…”

The Effect of Expected Value on Attraction Effect Preference Reversals

“…In contrast, an alternative explanation is that preference reversals are rational (Bordley, 1992;Howes et al, 2016;McNamara, Trimmer, & Houston, 2014;Shenoy, & Yu, 2013;Trimmer, 2013). Rather than taking axiomatic violations as evidence that people's cognitive bounds prevent them from achieving normatively correct decisions, we can assume the cognitive bounds and ask what the optimal decision should be given those limitations (Howes, Lewis, & Vera, 2009;Lewis, Howes, & Singh, 2014).…”

“…We discuss the implications of these results for theories of human choice. In particular, we argue that while contextual preference reversals appear locally irrational in the sense of violating axioms of decision models, they may in fact be globally rational when the decision environment and cognitive constraints are taken into account (Howes, Warren, Farmer, El‐Deredy, & Lewis, ).…”

Optimal policy for multi-alternative decisions

“…Some studies suggest that apparently irrational human behavior could be accounted for by heuristic weighting rules for features, which integrate feature valences through feedforward 42,43,46 or recurrent 12,44,45 neural interactions. Interestingly, a recent study reports that a context-dependent feature weighting can increase the robustness of value encoding to neural noise in later processing stages 43,54 . However, to our knowledge, the optimal policy for these more complex models in which the value function is computed by combining multiple features, presented sequentially, remains unknown.…”