The prefrontal cortex is a core limbic region crucial for emotional memory formation and adaptive decision-making. Accumulating evidence across species highlights the distinct roles of prefrontal subregions in various behaviors and memory processes. Despite this understanding, little is known about how distinct features of associative memories are encoded in neuronal ensembles within and across distinct prefrontal subregions and the role of these representations in the encoding and retrieval of emotional memories. In this study, we aim to elucidate the interplay between prefrontal subregions and investigate how distinct neuronal populations within and across these regions collectively encode and integrate emotional information. We aim to uncover the latent structure underlying the representation of emotional features within prefrontal ensembles and to decipher the emergent patterns of neural activity associated with specific emotional states, which form the neural substrate of adaptive decision-making and emotional regulation. To address these questions, we use a combination of large-scale electrophysiological recordings, paired with computational modeling, to dissect the spatiotemporal dynamics of the prefrontal ensemble representation of emotionally salient cues during associative learning. Leveraging the simultaneous recordings from multiple prefrontal subregions, we characterize the neuronal dynamics and oscillatory landscape within and across regions and we establish their differential but complementary involvement in the representation of emotional features along the valence and saliency axes, as well as the encoding of ongoing internal and behavioral states within specific prefrontal ensembles.