Animal research on brain mechanisms involved in psychiatric disorders presents an enormous challenge because it is impossible to precisely model symptoms of a human disorder in a rat or mouse. Nevertheless, there are uses for animal models as long as the limitations are recognized. Animal research related to posttraumatic stress disorder (PTSD) points to acute and chronic stressors, such as restraint or immobilization as being the most relevant stimuli to study how neural and endocrine systems are affected, both immediately and long term. Of particular relevance are the onset and duration of effects of stressors on brain areas subserving emotional memories, such as the amygdala, prefrontal cortex, and hippocampus. The hippocampus plays a role in memory and in vegetative functions of the body. The hippocampus receives input from the amygdala and its function in spatial memory is altered by amygdala activity. Repeated stress in the rat suppresses dentate gyrus neurogenesis and causes dendrites of hippocampal and medial prefrontal cortical neurons to shrink. Conversely, it causes basolateral amygdala neurons to increase in dendritic complexity and sprout new synapses. Repeated stress also increases fear and aggression, reduces spatial memory, and alters contextual fear conditioning. Antidepressants and mood stabilizers have diverse effects on these processes. New data indicate that a single stress episode can cause a delayed alteration in synapse formation in the basolateral amygdala without changing dendritic length and branching. Further studies are examining the structural changes in prefrontal cortex and hippocampus as a result of single traumatic stressors, which may reflect the functional interactions with the amygdala. Together with mechanistic studies of the role of adrenal glucocorticoids and catecholamines, these results may tell us how the brain is shaped by acute and repeated uncontrollable stress in ways that then can be investigated in human anxiety disorders.