Allocating resources across time in a cumulative allocation game

In human Cumulative Allocation Games, participants’ proportion bet fell by almost a factor of two over time, cutting expected log growth rate by roughly 25% versus normative play.

• Why it matters: The study addresses a gap in how people perform in cumulative allocation tasks and whether their behavior matches optimal decision-under-risk strategies.
• What they did: Across two experiments, participants completed four CAGs and their choices were compared with normative play that maximizes log rate of growth.
• The result: The systematic drift away from the normative strategy suggests loss-aversion–like behavior, potentially driven by an asymmetry between losses and gains specific to CAGs.

AI summary of the abstract · may contain errors · not medical advice · read the paper.