The representation of emotion knowledge in hippocampal-prefrontal systems

Abstract

Emotional experiences involve more than bodily reactions and momentary feelings--they depend on knowledge about the world that spans contexts and time. Although it is well established that individuals conceptualize emotions using a low-dimensional space organized by valence and arousal, the neural mechanisms giving rise to this configuration remain unclear. Here, we examine whether hippocampal-prefrontal circuits--regions implicated in forming cognitive maps--also support the abstraction of emotional experiences. Using functional MRI data collected as participants viewed emotionally evocative film clips, we found that activity in hippocampal and prefrontal cortex predicted self-reported emotion across schematically distinct videos, consistent with a role in structural learning. Computational modeling of emotion transitions revealed that hippocampal responses to films and emotion self-reports could be predicted based on the statistical regularities of emotion transitions across different temporal scales. These findings demonstrate that hippocampal-prefrontal systems represent emotion concepts at multiple levels of abstraction, offering new insight into how the brain organizes emotion knowledge.