difficulties in their marriages, and did not show dismissive attachment. They did, however, have relatively high scores on several prebereavement measures suggestive of the ability to adapt well to loss (e.g., acceptance of death, belief in a just world, instrumental support). As in previous studies, no unequivocal evidence for delayed grief was found. Finally, it is important to note that even among these resilient individuals, the majority reported experiencing at least some yearning and emotional pangs, and virtually all participants reported intrusive cognition and rumination at some point early after the loss (Bonanno, Wortman, & Nesse, in press). The difference between the January 2004 ● American Psychologist 23 resilient individuals and the other participants, however, was that these experiences were transient rather than enduring and did not interfere with their ability to continue to function in other areas of their lives, including the capacity for positive affect. Resilience to Violent and Life-Threatening Events Epidemiological studies estimate that the majority of the U.S. population has been exposed to at least one traumatic event, defined using the DSM–III criteria of an event outside the range of normal human experience, during the course of their lives. Although grief and trauma symptoms are qualitatively different, the basic outcome trajectories following trauma tend to form patterns similar to those observed following bereavement (see Figure 1). Summarizing this research, Ozer et al. (2003) recently noted that “roughly 50%–60% of the U.S. population is exposed to traumatic stress but only 5%–10% develop PTSD” (p. 54). However, because there is greater variability in the types and levels of exposure to stressor events, there also tends to be greater variability in PTSD rates over time. Estimates of chronic PTSD have ranged, for example, from 6.6% and 9.9% for individuals experiencing personally threatening and violent events, respectively, during the 1992 Los Angeles riots (Hanson, Kilpatrick, Freedy, & Saunders, 1995), to 12.5% for Gulf War veterans (Sutker, Davis, Uddo, & Ditta, 1995), to 16.5% for hospitalized survivors of motor vehicle accidents (Ehlers, Mayou, & Bryant, 1998), to 17.8% for victims of physical assault (Resnick, Kilpatrick, Dansky, Saunders, & Best, 1993). Although chronic PTSD certainly warrants great concern, the fact that the vast majority of individuals exposed to violent or life-threatening events do not go on to develop the disorder has not received adequate attention. It is well established that many exposed individuals will evidence short-lived PTSD or subclinical stress reactions that abate over the course of several months or longer (i.e., the recovery pattern). For example, a population-based survey conducted one month after the September 11th terrorist attacks in New York City estimated that 7.5% of Manhattan residents would meet criteria for PTSD and that another 17.4% would meet the criteria for subsyndromal PTSD (high symptom levels that do not meet full diagnostic criteria; Galea, Ahern, et al., 2002). As in other studies, a subset eventually developed chronic PTSD, and this was more likely if exposure was high.