This project empirically addresses how people come to detect and represent supernatural agents (SNA) as acting within their physical environments. Within the cognitive science of religion (CSR), two cornerstone cognitive mechanisms have been hypothesised to facilitate belief in SNA. One is the ability to reason about minimally counterintuitive concepts (Boyer, 2001). The other is the ability to infer the presence of agency within the environment, suggested by Barrett (2004) to be the product of a hypersensitive agency detection device (HADD). While an empirical research project investigating minimally counterintuitive concepts exists, there is currently a paucity of work addressing the psychological mechanisms through which the HADD gives rise to belief in SNAs. Given that Gibson and Barrett (2008) have called for more psychological attention to be paid to CSR, and, given that theories of supernatural agency detection represent an important theoretical research program both within (e.g., Barrett, 2004; Guthrie, 1993) and outside of CSR (e.g., Dennett, 2006), we have identified this as an area in particular need of empirical attention. The output of a proposal focusing on this topic would not only inform theory within the field but would also set the foundations for future work in this area. Building on earlier work by Barnes and Gibson (under review), this project will address the gap in the literature in the following ways: 1) conducting research using cognitive and physiological methods to examine the situational and dispositional factors hypothesized to affect the deployment of the hypersensitive agency detection device and its subsequent interaction with religious beliefs; 2) producing a conceptual article that synthesizes CSR theory with data and theory in other relevant areas of psychology (e.g., agency detection, attribution) to inform a detailed model for the workings of HADD; 3) disseminating findings in scientific journals and conferences.