Preference reversal in multiattribute choice.

Area of Science:

• Cognitive Psychology
• Decision Science
• Computational Neuroscience

Background:

• Decision-making theories face challenges explaining preference reversals caused by decoy or irrelevant options.
• The attraction, compromise, and similarity effects are key examples of such preference reversals.
• Existing theoretical proposals struggle to simultaneously account for all three decoy effects.

Purpose of the Study:

• To review proposed mechanisms for decoy effects in choice.
• To analyze and compare two computational models: decision field theory (DFT) and leaky competing accumulators (LCA).
• To determine which model provides a more integrated and robust account of decoy effects.

Main Methods:

• Literature review of mechanisms explaining decoy effects.
• Computational modeling and simulation of DFT and LCA.
• Comparative analysis of model predictions for attraction, compromise, and similarity effects.

Main Results:

• Both DFT and LCA were simulated to examine their predictions regarding decoy effects.
• Differences in how the models predict these effects were identified.
• The LCA framework demonstrated a more robust ability to account for the observed decoy effects.

Conclusions:

• The leaky competing accumulators (LCA) framework provides a more comprehensive explanation for decoy effects.
• LCA's success suggests shared mechanisms between high-level decision-making and perceptual choice.
• This aligns with Tversky's relational evaluation with loss aversion theory.