Abstract: Two decades of conditioned fear studies reveal impaired extinction in traumatized military and civilian populations with posttraumatic stress disorder (PTSD). We aimed to test fear extinction in women with military sexual trauma (MST), a highly traumatized and largely understudied population. Our well-established, acoustic startle-based fear acquisition and extinction paradigm was administered to 51 age-matched female veterans with prior exposure to MST receiving healthcare at one of three VA Medical Centers. Thirty women had a PTSD diagnosis and 21 did not, per clinician-administered and self-reported measures. Conditioned stimuli were geometric shapes presented on a computer screen and the unconditioned stimulus was an airblast to the larynx. Fear acquisition and fear extinction training occurred on a single day separated by 10 min. Extinction recall occurred 24 h after extinction training. All women showed fear-potentiated startle to the CS+ and CS+/CSdiscrimination. During extinction training, the PTSD- and PTSD + groups significantly differed in terms of extinction success, with PTSD + women showing higher fear to the CS + at the end of extinction training compared to PTSD-women (F(1,35) = 5.52, p = 0.025). Furthermore, 86 % of the PTSD-women exhibited a >= 50 % decrease in fear-potentiated startle during extinction training yet only 53 % of the PTSD + women showed this magnitude of fear reduction [X2(1,50) = 5.83, p = 0.016)]. These data support the mechanistic position that impaired fear extinction can be a central feature of PTSD, however, observed extinction impairments are nuanced and may be related to factors including age, trauma type, symptom profile, hormonal levels, and biological sex.