Hippocampal volume reductions and functional impairments are reliable findings in posttraumatic stress disorder (PTSD) imaging studies. However, it is not clear if and how hippocampal dysfunction contributes to the etiology and maintenance of PTSD. Individuals with PTSD are often described as showing fear responses to trauma reminders outside of contexts in which these cues would reasonably predict danger. Animal studies suggest that the hippocampus is required to form and recall associations between contextual stimuli and aversive events. For example, the hippocampus is critical for encoding memories in which a complex configuration of multiple cues is associated with the aversive event. Conversely, the hippocampus is not required for associations with discrete cues. In animal studies, if configural memory is disrupted, learning strategies using discrete cue associations predominate. These data suggest poor hippocampal function could bias the organism toward forming multiple simple cue associations during trauma, thus increasing the chances of fear responses in multiple environments (or contexts) in which these cues may be present. Here we will examine clinical and preclinical literature to support a theory of hippocampal dysfunction as a primary contributory factor to the etiology of PTSD, and discuss future research required to test these hypotheses. This article is part of a Special Issue entitled ‘Post-Traumatic Stress Disorder’.