Temporal uncertainty in observational learning

Description: By observing others’ experiences, humans learn about threats while avoiding firsthand dangers. Yet, previous neuroscience research has focused on observational threats that are predictable. In firsthand learning, temporally predictable and unpredictable threats can be mechanistically discerned. In this study, we developed a novel observational paradigm in which participants learned from predictable (P) and unpredictable (U) observational threats, as well as a no-threat (N) condition. Participants encountered the same conditions during an expression phase based on the NPU paradigm to investigate how the brain encodes predictable and unpredictable threat cues observed in others. Participants in Experiment 1 (n = 20, 14 female, 6 male) and Experiment 2 (n = 23, 9 female, 1 diverse, 13 male) successfully learned threat contingencies, showing heightened threat expectations for predictable cues and unpredictable condition-onsets. Behavioral findings converged with neural (fMRI, Experiment 2) responses in the anterior insula during the expression phase. Reflecting the dynamic process of learning, amygdala activation in response to predictable threat cues with a linear decrease across trials. Interestingly, BOLD responses to others’ pain were enhanced within the amygdala, insula and hippocampus when participants could learn to predict threats compared to unpredictable conditions. We conclude that our findings suggest that humans can learn to resolve temporal uncertainty, through the observation of others. The present work thereby contributes to understanding the social aspects of fear and anxiety disorders. For a description of how fMRI data was preprocessed and analyzed and to understand contrasts: https://doi.org/10.1101/2025.10.21.682164