How sampling strategies shape experience-based risky choice.

Area of Science:

• Cognitive Psychology
• Behavioral Economics
• Computational Neuroscience

Background:

• Risky choice phenomena like risk aversion are often modeled using psychoeconomic curves.
• These models describe choices but not the underlying cognitive processes.
• Information search during decision-making is a crucial but often overlooked factor.

Purpose of the Study:

• To investigate how information search strategies, specifically switching and stopping rules, shape experience-based risky choices.
• To develop a computational framework for analyzing sampling strategies in decision-making.
• To link search processes to observed patterns in risky decision-making.

Main Methods:

• Developed a computational framework with three components: search rule (switching), comparison rule (roundwise vs. summary), and stopping rule.
• Conducted simulation analyses comparing various combinations of these sampling strategy components.
• Performed an empirical analysis to assess human decision-making strategies.

Main Results:

• Frequent switching can increase or decrease expected value (EV)-maximizing choices depending on the comparison rule used.
• Switching behavior influences the apparent weighting of rare events and risk attitudes.
• Empirical data suggest humans employ strategies that promote EV-maximizing choices.

Conclusions:

• Information search processes, including switching and stopping rules, play a significant role in generating patterns of risky choice.
• Computational modeling of search strategies provides a novel perspective on decision-making.
• Future research should further explore the interplay between information search and risky decision-making.