Counterfactual thinking — considering what could have come of choosing the other path — can facilitate inference. Previous studies have demonstrated that memory and decision making interact to shape the inferred value of unchosen options. Indeed, when an individual chooses an option and it turns out to be good, the alternative is often inferred to have lower value, even though the outcome associated with the unchosen option is never experienced. The current study investigated whether these inferred values followed some of the same memory-mediated rules as values formed through direct experience. For example, much is known about how directly experienced reward value spreads to related items in memory. Is the same true for inferred values? Our findings demonstrate that yes, participants generalize inferred values of unchosen options to novel items within the same semantic category. Specifically, participants were more likely to choose or avoid novel items from the same category as an unchosen option that had been paired with a valuable or non-valuable chosen item, respectively. These findings advance our understanding of how counterfactual thinking aids inference formation and interacts with memory to guide adaptive behavior in novel contexts. • The value of an unchosen option is inferred to be positive if the outcome of a choice is poor. • The inferred value of an unchosen option spreads to items semantically similar to it. • Memory interacts with counterfactual reasoning to allow inferences and guide choice.
Wang et al. (Sat,) studied this question.