Discounting causes of behavior: Are two reasons better than one?

Several theories propose that people discount a cause of an action when other plausible causes are present. This view has recently been challenged, but the relevant research has not been reviewed. In this article, the author reviews research on factors that affect discounting and the use of conjunctive explanations. Some studies are inconclusive because inappropriate measures are used. Other studies fail to distinguish between the probability, necessity, sufficiency, and relevance of causes. When these distinctions are recognized, patterns of discounting are predictable. When causes are presented sequentially, people may underdiscount the prior cause, suggesting that an anchoring process may limit discounting, In other cases, discounting is absent because people perceive multiple causal influences on actions or because they judge that certain causes are necessary or sufficient. Theory has assumed that causes are negatively associated, but causes may be independent or positively associated. This conclusion challenges analogies between discounting and scientific and legal reasoning.