Disentangling Uncertainty Signals in Neural Reward Dynamics

Most decisions are made under uncertainty, which manifests in either risk (know probability) or ambiguity (unknown probability). However, it is not well understood how the two forms of uncertainty signal differ in value-based decision-making. This ERP study aimed to disentangle them by examining how their neural reward dynamics are influenced by individual attitudes and their derived subjective value (SV). Thirty-six participants performed a wheel-of-fortune task while EEG activity was recorded, followed by a lottery task where their attitudes toward risk and ambiguity were assessed. These attitudes were then used to calculate the trial-level SV for gambling wheels in the wheel-of-fortune task. Results showed that risk was associated with a greater cue-P3, whereas ambiguity was linked to a greater cue-N2 and stronger valence effects on the both reward positivity and feedback-P3. Importantly, SV was represented by all ERP components under risk but not under ambiguity. Finally, less risk-averse attitudes were associated with reduced valence and probability effects on the feedback-P3, whereas less ambiguity-averse attitudes were linked to greater ambiguity-level effect on the cue-P3. Our findings reveal a dissociable relationship between risk and ambiguity, supporting the crucial role of individual attitudes and SV in disentangling them in uncertainty neural signals.