Prediction-based neural mechanisms for shielding the self from existential threat

Abstract

The human mind has an automatic tendency to avoid awareness of its mortality. How this protective mechanism is implemented at the neuronal level is unknown. Here we test the hypothesis that prediction-based mechanisms mediate death-denial by shielding the self from existential threat. We provide evidence that self-specific predictive processes are downregulated during the perception of death-related linguistic stimuli and that this mechanism can predict fear-of-death. Using a magnetoencephalography visual mismatch paradigm, we show that the brain’s automatic prediction response to deviancy is eliminated when death words and self-face representations are coupled, but remains present when coupled to other-face or to negative words. We further demonstrate a functional link between how death impacts self-image vs. Other-image, and show that it predicts fear-of-death. Finally, we confirm this effect in a behavioral active inference experiment showing that death-related words bias perceptual judgment on facial self and other morphed video clips. Together these results lay out, for the first time, a plausible neural-based mechanism of death-denial.