Depressed patients are often asked to take part in clinical research studies that carry risk. These patients are generally assumed to be mentally competent to consent to research, since depression often leaves a patient's cognitive abilities intact. In this article, it is argued that many severely depressed patients may not be competent to consent because they cannot be considered accountable for their decisions. The article presents 2 arguments: first, that it is unclear whether the decisions of some severely depressed patients are authentically theirs, and second, that some severely depressed patients may not have the appropriate minimal degree of concern for their own well-being. It is argued that assessments of competence must take account of emotional factors, and that, if severely depressed patients are incompetent to consent, research studies involving a poor risk-benefit ratio will be much more difficult to justify.