hope you find this helpful: http://www.nature.com/npp/journal/v38/n1/fig_tab/npp201279b2.html

I suggest Hagit Cohen's model which has the great face validity. 

Please see below:

Behavioural Brain Research

Volume 205, Issue 2, 28 December 2009, Pages 544–549

Consequences of extinction training on associative and non-associative fear in a mouse model of Posttraumatic Stress Disorder (PTSD)

Yulia Goluba, b, et al. 

Abstract

A common approach to the clinical treatment of Posttraumatic Stress Disorder (PTSD) has focused on the facilitation of fear extinction through cognitive behavioural therapy that involves both safe exposure to the trauma-related cues and subsequent changes in conditioned stimulus–unconditioned stimulus (CS–US) contingency expectations. PTSD symptoms can be tracked back to pathologically modified associative fear, hyperarousal and a time-dependent fear generalization. We have used a mouse model of PTSD that is based on a brief exposure to an inescapable foot shock in order to investigate the influence of early (starting 1 day after the shock) and late (starting 1 month after the shock) extinction training. Both early and late extinction training led to a long-lasting reduction of contextual and generalized fear, but only early extinction caused an amelioration of hyperarousal. Consequently, our results suggest early post-shock intervention as a successful strategy for reducing hyperarousal in the aftermath of a trauma.

Journal of Psychiatric Research 41 (2007) 848–860

A mouse model of posttraumatic stress disorder that distinguishes between conditioned and sensitised fear

Anja Siegmund, Carsten T. Wotjak,

Abstract

The pathomechanisms of posttraumatic stress disorder (PTSD) are still unknown, but both fear conditioning and stress sensitisation are supposed to play a crucial role. Hence, valid animal models that model both associative and non-associative components of fear will facilitate elucidation of the biological substrates of the illness, and to develop novel and specific approaches for its prevention and therapy. Here we applied a single electric footshock to C57BL/6N (B6N) and C57BL/6JOla (B6JOla) mice and recorded the conditioned response to contextual trauma reminders (associative fear), the sensitised reaction to a neutral tone in a novel environment (non-associative fear, hyperarousal), social interaction and various emotional behaviours using Modified Holeboard, Test for Novelty-Induced Suppression of Feeding and Forced Swimming Test, after different incubation times (1, 14, 28 days). Freezing generally increased as a function of shock intensity. In B6N mice, sensitised fear was maximal 28 days after trauma and was accompanied by signs of emotional blunting and social withdrawal. B6JOla mice, in contrast, were less susceptible to develop PTSD-like symptoms. The phenotype of B6N exhibited high behavioural variance, allowing distinction between vulnerable and resilient individuals. Only in vulnerable B6N mice, chronic fluoxetine treatment – initiated after an incubation period of 28 days – ameliorated sensitised fear. This new mouse model fulfils common criteria for face and predictive validity and can be used to investigate the biological correlates of individual fear susceptibility, as well as the impact and interrelationship of associative and non-associative fear components in the development and maintenance of PTSD.