I propose that the ways people respond to their own symptoms of depression influence the duration of these symptoms. People who engage in ruminative responses to depression, focusing on their symptoms and the possible causes and consequences of their symptoms, will show longer depressions than people who take action to distract themselves from their symptoms. Ruminative responses prolong depression because they allow the depressed mood to negatively bias thinking and interfere with instrumental behavior and problem-solving. Laboratory and field studies directly testing this theory have supported its predictions. I discuss how response styles can explain the greater likelihood of depression in women than men. Then I intergrate this response styles theory with studies of coping with discrete events. The response styles theory is compared to other theories of the duration of depression. Finally, I suggest what may help a depressed person to stop engaging in ruminative responses and how response styles for depression may develop.