It is hypothesized that there is a pervasive and fundamental bias in humans' understanding of physical causation: Once the roles of cause and effect are assigned to objects in interactions, people tend to overestimate the strength and importance of the causal object and underestimate that of the effect object in bringing about the outcome. This bias is termed the causal asymmetry. Evidence for this bias is reviewed in several domains, including visual impressions of causal relations, reasoning about Newton's third law in naive physics problems, concepts underlying linguistic expressions of causality, and research in causal judgment from contingency information. Although there might be an equivalent to the causal asymmetry in the domain of social causality, there are too many uncertainties in the evidence for conclusions to be drawn.