The incidence of depressive illness is high worldwide, and the inadequacy of currently available drug treatments contributes to the significant health burden associated with depression. A basic understanding of the underlying disease processes in depression is lacking; therefore, recreating the disease in animal models is not possible. Popular current models of depression creatively merge ethologically valid behavioral assays with the latest technological advances in molecular biology. Within this context, this study aims to evaluate animal models of depression and determine which has the best face, construct, and predictive validity. These models differ in the degree to which they produce features that resemble a depressive-like state, and models that include stress exposure are widely used. Paradigms that employ acute or sub-chronic stress exposure include learned helplessness, the forced swimming test, the tail suspension test, maternal deprivation, chronic mild stress, and sleep deprivation, to name but a few, all of which employ relatively short-term exposure to inescapable or uncontrollable stress and can reliably detect antidepressant drug response.