An apparatus, method, and computer program product are provided for the improved and automatic prediction of a relativistic, observer-specific perception and response to a potential event and, based at least in part on the predicted perception and response, generating and presenting observer-specific digital content items. Some example implementations employ predictive, machine-learning modeling to facilitate user-specific event perception and response prediction and the selection of particularized messages and other digital content items for presentation to the user.