The present article examines how people's belief in an afterlife, as well as closely related supernatural beliefs, may open an empirical backdoor to our understanding of the evolution of human social cognition. Recent findings and logic from the cognitive sciences contribute to a novel theory of existential psychology, one that is grounded in the tenets of Darwinian natural selection. Many of the predominant questions of existential psychology strike at the heart of cognitive science. They involve: causal attribution (why is mortal behavior represented as being causally related to one's afterlife? how are dead agents envisaged as communicating messages to the living?), moral judgment (why are certain social behaviors, i.e., transgressions, believed to have ultimate repercussions after death or to reap the punishment of disgruntled ancestors?), theory of mind (how can we know what it is "like" to be dead? what social-cognitive strategies do people use to reason about the minds of the dead?), concept acquisition (how does a common-sense dualism interact with a formalized socio-religious indoctrination in childhood? how are supernatural properties of the dead conceptualized by young minds?), and teleological reasoning (why do people so often see their lives as being designed for a purpose that must be accomplished before they perish? how do various life events affect people's interpretation of this purpose?), among others. The central thesis of the present article is that an organized cognitive "system" dedicated to forming illusory representations of (1) psychological immortality, (2) the intelligent design of the self, and (3) the symbolic meaning of natural events evolved in response to the unique selective pressures of the human social environment.

Participants were interviewed about the biological and psychological functioning of a dead agent. In Experiment 1, even 4- to 6-year-olds stated that biological processes ceased at death, although this trend was more apparent among 6- to 8-year-olds. In Experiment 2, 4- to 12-year-olds were asked about psychological functioning. The youngest children were equally likely to state that both cognitive and psychobiological states continued at death, whereas the oldest children were more likely to state that cognitive states continued. In Experiment 3, children and adults were asked about an array of psychological states. With the exception of preschoolers, who did not differentiate most of the psychological states, older children and adults were likely to attribute epistemic, emotional, and desire states to dead agents. These findings suggest that developmental mechanisms underlie intuitive accounts of dead agents' minds.

Little is known about how the minds of dead agents are represented. In the current experiment, individuals with different types of explicit afterlife beliefs were asked in an implicit interview task whether various mental state types, as well as pure biological imperatives, continue after death. The results suggest that, regardless of one's explicit reports about personal consciousness after death, those who believe in some form of life after death (and, to a certain extent, even those who do not) implicitly represent dead agents' minds in the same way: psychobiological and perceptual states cease while emotional, desire, and epistemic states continue. The ndings are interpreted according to simulation constraints -because it is epistemologically impossible to know what it is like to be dead, individuals will be most likely to attribute to dead agents those types of mental states that they cannot imagine being without. Such a model argues that it is natural to believe in life after death and social transmission serves principally to conceptually enrich (or degrade) intuitive conceptions of the afterlife.

The evolution of human cooperation remains a puzzle because cooperation persists even in conditions that rule out mainstream explanations. We present a novel solution that links two recent theories. First, Johnson & Kruger (2004) suggested that ancestral cooperation was promoted because norm violations were deterred by the threat of supernatural punishment. However, this only works if individuals attribute negative life events (or a prospective afterlife) as intentionally caused by supernatural agents. A complementary cognitive mechanism is therefore required. Recently, Bering and Shackelford (2004) suggested precisely this. The evolution of "theory of mind" and, specifically, the "intentionality system" (a cognitive system devoted to making inferences about the epistemic contents and intentions of other minds), strongly favoured: (1) the selection of human psychological traits for monitoring and controlling the flow of social information within groups; and (2) attributions of life events to supernatural agency. We argue that natural selection favoured such attributions because, in a cognitively sophisticated social environment, a fear of supernatural punishment steered individuals away from costly social transgressions resulting from unrestrained, evolutionarily ancestral, selfish interest (acts which would rapidly become known to others, and thereby incur an increased probability and severity of punishment by group members). As long as the net costs of selfish actions from real-world punishment by group members exceeded the net costs of lost opportunities from self-imposed norm abiding, then god-fearing individuals would outcompete non-believers.

In the present study, a modified dictator game was used to test the hypothesis that the threat of gossip would encourage prosocial decision making. All participants were asked to distribute an endowment between themselves and an anonymous second party. Half of the participants were told that the second party would be discussing their economic decision with a third party. For some participants, this third party was someone to whom they had first disclosed personally identifying information. Participants who received the threat of gossip manipulation were more generous than control participants, but only when the third party could personally identify them was this difference significant. These data reveal that at least some prosocial decisions are motivated by actor's reputational concerns-concerns that are directly mediated by language.