Event:

Abstract: Confidence judgments reflect people’s ability to monitor their own uncertainty, yet little is known about how this monitoring operates when individuals must generalize knowledge from learned risky options to unfamiliar ones. In this study, we examine how people learn the reward structure of risky options, how they generalize this knowledge to novel options, and how their confidence tracks this process. Participants first learn how shape and color correspond to the mean and variance of payoff distributions by repeatedly observing outcomes from nine exemplar distributions. They then make confidence-weighted preference judgements for novel stimuli that vary systematically in similarity to the learned exemplars. Using computational modelling, we estimate similarity-based generalisation functions for both mean and variance information and examine how these inferred beliefs feed into subjective utility and confidence. This approach allows us to test not only how people infer the value of new options from known ones, but also whether generalisation differs across the two statistical dimensions of risky choice. By comparing alternative generalisation functions, we also identify the specific strategies that people use to evaluate unfamiliar options. These findings show how people monitor their own uncertainty when judging novel risky prospects, offering new insights into the computational underpinnings of confidence and generalisation in experience-based choice.

Join our Research Seminar online:

Link: https://bbb.tum.de/rooms/sh4-fn5-pov-f3e/join


We are looking forward to a great discussion.


See full program: https://franziskaknolle.com/journal-club-tumnic/