In this article, we discuss what animal models of depression should be attempting to 'model'. One must first determine if the goal is to model the regulatory mechanisms by which antidepressant treatments alleviate the various symptoms of depression, or to model the dysregulatory mechanisms underlying the etiology of those symptoms. When modeling the mechanisms of antidepressant effects, a key feature that is often overlooked is the time course required for behavioral efficacy. Even in the clinical literature, there is considerable confusion and inconsistency in defining and identifying 'time of onset' of clinical effect. Although the 'therapeutic lag' may not be as long as has been commonly believed, it does occur. Observable improvement in either global symptomatology or specific symptoms becomes evident after 7-14 days of treatment, and more complete recovery takes considerably longer. Thus, any model addressing potential mechanisms of antidepressant action should exhibit a similar time-dependency. Second, whether attempting to address mechanisms underlying behavioral effects of antidepressants, or the neurobiological substrates underlying the development and manifestation of depression, it is essential to recognize that the syndrome of depression is a diagnostic construct that includes a variety of disparate symptoms, some of which may be related mechanistically, and others that may not be specific to depression, but may cut across categorical diagnostic schemes. Further, it is critical to recognize the close relationship of depression and anxiety. Psychological studies have suggested that the myriad symptoms of depression and anxiety may be subsumed within a more limited number of distinct behavioral dimensions, such as negative affect (neuroticism), positive affect, or physiologic hyperarousal. These dimensions may be related to the functioning of specific neurobiological systems. Thus, rather than trying to recreate or mimic the entire spectrum of symptoms comprising the syndrome of depression, it may be more informative to develop animal models for these behavioral dimensions. Such models may then provide access not only to the neural regulatory mechanisms underlying effective antidepressant treatment, but may also provide clues to the processes underlying the development and manifestation of depression.