Agency detection

The concept comes from cognitive anthropologist Stewart Guthrie's Faces in the Clouds (1993) and was developed by Justin Barrett, Pascal Boyer, and others into the Hyperactive Agency Detection Device (HADD) hypothesis. The idea is that natural selection favored cognitive systems that erred heavily on the side of detecting agents — a rustle in the grass that turns out to be the wind costs little, but a rustle that turns out to be a predator and is missed costs everything. The result is a brain that sees agency everywhere, including where there is none.

If the hypothesis is correct, belief in invisible intentional agents — ancestors, spirits, gods — is not a strange cultural accident but a predictable byproduct of an evolved cognitive system doing exactly what it was selected to do. Belief in agents we cannot see, who notice us, and who care about our behavior would emerge spontaneously across cultures, which matches the ethnographic record well.

The implications cut multiple ways. For the secular thinker, agency detection is a parsimonious naturalistic explanation for religion that does not require any of the religion's claims to be true. For the religious thinker, the same data are compatible with God designing the cognitive faculty to track real divine agency — a sensus divinitatis in Calvin's terms, or a properly basic perceptual capacity in Plantinga's. Cognitive science of religion is a description of the mechanism, not a verdict on what the mechanism is responding to.

The hypothesis has been criticized for being too neat — not every culture's god-concepts fit the predicted profile — and for its limited evidential support, which leans heavily on cross-cultural surveys and lab experiments rather than direct neuroimaging. But the broader research program of cognitive science of religion (CSR), of which HADD is one part, is one of the most productive recent developments in the secular study of belief.