What might happen is generally difficult to control and can range from innocent mistakes made by employee to natural disaster?

Imagine you are standing beside some tram tracks. In the distance, you spot a runaway trolley hurtling down the tracks towards five workers who cannot hear it coming. Even if they do spot it, they won’t be able to move out of the way in time.

As this disaster looms, you glance down and see a lever connected to the tracks. You realise that if you pull the lever, the tram will be diverted down a second set of tracks away from the five unsuspecting workers.

However, down this side track is one lone worker, just as oblivious as his colleagues.

So, would you pull the lever, leading to one death but saving five?

This is the crux of the classic thought experiment known as the trolley dilemma, developed by philosopher Philippa Foot in 1967 and adapted by Judith Jarvis Thomson in 1985.

The trolley dilemma allows us to think through the consequences of an action and consider whether its moral value is determined solely by its outcome.

The trolley dilemma has since proven itself to be a remarkably flexible tool for probing our moral intuitions, and has been adapted to apply to various other scenarios, such as war, torture, drones, abortion and euthanasia.

Variations

Now consider now the second variation of this dilemma.

Imagine you are standing on a footbridge above the tram tracks. You can see the runaway trolley hurtling towards the five unsuspecting workers, but there’s no lever to divert it.

However, there is large man standing next to you on the footbridge. You’re confident that his bulk would stop the tram in its tracks.

So, would you push the man on to the tracks, sacrificing him in order to stop the tram and thereby saving five others?

The outcome of this scenario is identical to the one with the lever diverting the trolley onto another track: one person dies; five people live. The interesting thing is that, while most people would throw the lever, very few would approve of pushing the fat man off the footbridge.

Thompson and other philosophers have given us other variations on the trolley dilemma that are also scarily entertaining. Some don’t even include trolleys.

Imagine you are a doctor and you have five patients who all need transplants in order to live. Two each require one lung, another two each require a kidney and the fifth needs a heart.

In the next ward is another individual recovering from a broken leg. But other than their knitting bones, they’re perfectly healthy. So, would you kill the healthy patient and harvest their organs to save five others?

Again, the consequences are the same as the first dilemma, but most people would utterly reject the notion of killing the healthy patient.

Actions, intentions and consequences

If all the dilemmas above have the same consequence, yet most people would only be willing to throw the lever, but not push the fat man or kill the healthy patient, does that mean our moral intuitions are not always reliable, logical or consistent?

Perhaps there’s another factor beyond the consequences that influences our moral intuitions?

Foot argued that there’s a distinction between killing and letting die. The former is active while the latter is passive.

In the first trolley dilemma, the person who pulls the lever is saving the life of the five workers and letting the one person die. After all, pulling the lever does not inflict direct harm on the person on the side track.

But in the footbridge scenario, pushing the fat man over the side is in intentional act of killing.

This is sometimes described as the principle of double effect, which states that it’s permissible to indirectly cause harm (as a side or “double” effect) if the action promotes an even greater good. However, it’s not permissible to directly cause harm, even in the pursuit of a greater good.

Thompson offered a different perspective. She argued that moral theories that judge the permissibility of an action based on its consequences alone, such as consequentialism or utilitarianism, cannot explain why some actions that cause killings are permissible while others are not.

If we consider that everyone has equal rights, then we would be doing something wrong in sacrificing one even if our intention was to save five.

Research done by neuroscientists has investigated which parts of the brain were activated when people considered the first two variations of the trolley dilemma.

They noted that the first version activates our logical, rational mind and thus if we decided to pull the lever it was because we intended to save a larger number of lives.

However, when we consider pushing the bystander, our emotional reasoning becomes involved and we therefore feel differently about killing one in order to save five.

Are our emotions in this instance leading us to the correct action? Should we avoid sacrificing one, even if it is to save five?

Real world dilemmas

The trolley dilemma and its variations demonstrate that most people approve of some actions that cause harm, yet other actions with the same outcome are not considered permissible.

Not everyone answers the dilemmas in the same way, and even when people agree, they may vary in their justification of the action they defend.

These thought experiments have been used to stimulate discussion about the difference between killing versus letting die, and have even appeared, in one form or another, in popular culture, such as the film Eye In The Sky.

If so, you’ll be interested in our free daily newsletter. It’s filled with the insights of academic experts, written so that everyone can understand what’s going on in the world. With the latest scientific discoveries, thoughtful analysis on political issues and research-based life tips, each email is filled with articles that will inform you and often intrigue you.

Editor and General Manager

Find peace of mind, and the facts, with experts. Add evidence-based articles to your news digest. No uninformed commentariat. Just experts. 90,000 of them have written for us. They trust us. Give it a go.

If you found the article you just read to be insightful, you’ll be interested in our free daily newsletter. It’s filled with the insights of academic experts, written so that everyone can understand what’s going on in the world. Each newsletter has articles that will inform and intrigue you.

Terrorism, a subset of human-caused disasters (Figure 2-2), can have a particularly devastating impact on psychological functioning. Terrorism carries with it a potentially greater impact than other disasters on distress responses, behavioral change, and psychiatric illness by virtue of the unique characteristics of terrorism events (see Table 2-1).

Terrorist attacks, and the threat of a terrorism event, may also result in more severe psychological consequences than other types of traumatic events due to a perceived lack of control. Perceptions of risk are influenced by the degree to which individuals feel they have knowledge of and control over an outside event and how familiar and catastrophic the event will be (for review, see Slovic, 1987). People are more likely to feel that an activity or event is not dangerous if they can control it. Under these circumstances, it becomes less effective to cope by distancing oneself from the population at risk if the risk is seemingly random. For example, the degree of public anxiety resulting from the 2001 Washington, D.C., area sniper attacks was much greater than the anxiety levels related to the violence that is endemic to many Washington, D.C., areas. The event affected many people in the region for weeks. It was easier for people to distance themselves from urban violence (which is controllable by staying away from urban centers) than from the sniper attacks that were perceived as more threatening and random than everyday shootings.

In addition to its distinctive characteristic of intent, terrorism can uniquely disrupt societal functioning. Terrorism has the capacity to erode the sense of community or national security; damage morale and cohesion; and open the racial or ethnic, economic, and religious cracks that exist in our society, as evidenced by an increase in hate crimes following the September 11, 2001, attacks (Human Rights Watch, 2002; FBI, 2002).

Following a terrorism event, most people will experience stress-related symptoms across the spectrum of psychological responses as illustrated earlier in Figure 2-1. Many of the psychological consequences of terrorism are similar to those seen in the aftermath of other disasters. However, the literature specific to the psychological sequelae of terrorist attacks is much more limited in both prevalence and detail than that related to other types of disasters. Similar to research in the broader trauma field, most of these studies have focused on PTSD or symptoms of PTSD as outcomes. Less is known about other, nonpathological outcomes. It is often difficult to compare studies because of the use of varying and previously unvalidated measurement instruments. Furthermore, the significance of selected PTSD symptoms for determining longer-term functioning is unclear. Methodological issues regarding this line of research are discussed further at the conclusion of this chapter. A review of the psychological consequences of terrorist attacks by Gidron (2002) found six studies that met his criteria for inclusion.5 In this review, Gidron calculated the rate of PTSD for those who were directly exposed to a terrorist attack to be 28.2 percent. Given that the terrorist attacks on the World Trade Center and the Pentagon on September 11, 2001, in addition to directly impacting thousands of people, may have had more than 100,000 direct witnesses (Schuster et al., 2001) and that millions across the country experienced the events through repeated media depictions, it can be presumed that the impact of these events was quite significant. Table 2-2 presents significant findings from a selected set of studies examining a range of terrorism events.

Studies taking place outside of the United States have frequently used similar designs to those within the United States, often focusing on PTSD or symptoms of PTSD. Although limited in number, these international research efforts add useful information to the knowledge-base on psychological consequences of different types of terrorism events. In general, findings regarding the psychological sequelae of terrorist attacks are similar to those seen in United States–based studies; commonly reported effects include PTSD and symptoms of PTSD, major depression, and general psychological distress as determined by various measures. Some of these studies also provide unique perspectives because they have been carried out on populations that have been exposed to varying forms of terrorism events such as smaller-scale bombings and shooting attacks (e.g., Abenhaim et al., 1992; Wilson et al., 1997) and a chemical attack (Kawana et al., 2001; Ohbu et al., 1997; Tochigi et al., 2002), while much of the body of research from the United States has focused on large explosive attacks.

Two of the most significant acts of terrorism in the United States, the 1995 Oklahoma City bombing and the attacks of September 11, 2001, prompted a small, but growing, literature on psychological consequences of terrorism in this country. Research on the Oklahoma City bombing revealed PTSD in approximately one-third of survivors of the direct bomb blast six months after the bombing, and nearly three-fourths of these were individuals with no prior history of PTSD (North et al., 1999). North and colleagues (1999) identified a specific constellation of symptoms that was highly predictive of PTSD. Avoidance and numbing symptoms were much more common among Oklahoma City bombing survivors with PTSD. In contrast, the symptoms of intrusive reexperience and hyperarousal were “nearly universal” among survivors and were not predictive of PTSD when occurring by themselves.

Several studies conducted after the Oklahoma City bombing focused on an adolescent population from the Oklahoma City Public School District. More than 40 percent of the middle school students who participated in one survey reported that they knew someone who was injured in the bombing, while more than 30 percent knew someone who was killed. Seven weeks after the bombing, 14.6 percent of the youth reported not feeling safe and 34.1 percent reported worrying about themselves or their families (Pfefferbaum et al., 1999). This survey of middle school students also found that television and emotional exposure to the terrorism event was associated with posttraumatic stress symptoms (Pfefferbaum et al., 2001b). School officials in Oklahoma City reported a 25 percent decrease in attendance in the first weeks following the bombing. Students' initial apprehension about returning to school was shared by parents who sought evidence of better protective measures (Wong, 2001). Teachers and school administrators became concerned about their ability to identify future perpetrators and to ensure the safety of students and staff.

A number of authors have investigated the impact of the terrorist attacks of September 11, 2001, on the United States population in general and on New York City residents specifically. Galea and colleagues (2002) examined PTSD symptoms in New York City residents one to two months after the attack. Results indicated that 7.5 percent of Manhattanites reported criterion symptoms of PTSD that were then used to estimate the prevalence of the disorder, while 20 percent of those near the World Trade Center at the time of the attacks reported such symptoms. Schlenger and colleagues (2002) studied a nationally representative cross-sectional sample one to two months after the attacks using self-reported symptoms of PTSD and general psychological distress to measure what they termed “probable PTSD.” They found that residents of the New York City metropolitan area had the highest rate of probable PTSD in the country at 11.2 percent; the rate in the Washington, D.C., metropolitan area was 2.7 percent, in other major metropolitan areas 3.6 percent, and across the rest of the country 4.0 percent. A similarly designed nationally representative study by Silver and colleagues (2002) found that two months after the attacks, 17 percent of the country (not including those residing in New York City) had symptoms of September 11th–related posttraumatic stress, while at six months, this number decreased to 5.8 percent. The discrepancy between the rates found by these two studies (4 percent versus 17 percent) likely reflects the different methodologies and populations used to estimate posttraumatic stress. In a national telephone survey conducted immediately after September 11, 2001, Schuster et al. (2001) measured the presence of various symptoms of distress responses. The symptoms reported by adults included feeling very upset when reminded of the events (30 percent), having trouble falling or staying asleep (11 percent), and feeling irritable (9 percent).

A large study commissioned by the New York City Board of Education examined the psychological consequences of the September 11, 2001, terrorist attacks on 8,266 public school students in grades 4–12 throughout the five boroughs of New York City six months after the attacks (Hoven et al., 2002). Results indicate widespread distress responses and symptoms of psychiatric illness that were not limited to students in proximity to the World Trade Center. Prevalence rates of symptoms such as those related to PTSD, generalized anxiety disorder, and separation anxiety were significantly higher than would be expected in children not exposed to a traumatic event. However, because pre-event baseline data are not available for the children surveyed, it is difficult to ascertain whether these findings reflect exposure to the terrorism event or other features of the population.

Similar to findings in the disaster mental health literature, some evidence indicates that terrorism events may lead to increases in substance use. One survey of residents of New York, Connecticut, and New Jersey found that 21 percent of cigarette smokers reported an increase in smoking after the attacks (Melnik et al., 2002). Similarly, Vlahov and colleagues (2002) reported increases in substance use, including alcohol, in New York City and the surrounding areas in the months after September 11, 2001. In comparison, a study of survivors of the Oklahoma City bombing found no new cases of diagnosable substance use disorder subsequent to the attack (North et al., 1999). It is important to make a distinction between increases in substance use and substance abuse. Data indicating a simple increase in alcohol or tobacco use do not necessarily indicate problematic or long-standing behavior changes.

Other behaviors and outcomes reflecting functional impairment after terrorism events are in need of further study. School dropout rates, divorce, and domestic or interpersonal violence and conflict are potential future research topics in this area. Increases in school or work absenteeism, which may indicate functional impairment, have been noted following terrorist attacks. A survey by Melnik and colleagues (2002) found that 27 percent of respondents who were working in New York City at the time of the September 11, 2001, attacks missed work in the following days. This was due primarily to transportation problems caused by increased security measures such as surveillance of bridges and tunnels leading into Manhattan. Increased absenteeism from work or school has also been reported after other violent events. For example, during the serial sniper attacks in the Washington, D.C., metropolitan area in October 2002, a significant increase in school absences occurred, with attendance rates as low as 10 percent at several elementary schools near one of the shooting sites (Schulte, 2002). However, this behavior may be considered an appropriate response rather than a distress response because one of the victims of the sniper was a child who was shot while walking from a car into a school. A similar distinction can be made when looking at behavioral responses to the anthrax attacks of 2001. An average citizen using gloves to open mail may have been considered to manifest an adverse behavioral change related to psychological distress. However, if the person was a staff member in one of the offices specifically targeted in the anthrax mailings, the use of gloves might be considered an appropriate response.

Health care seeking by individuals who are not actually at risk or injured, but seek health care due to fear and anxiety, has been observed in response to terrorism events. This phenomenon was noted following the sarin poisoning in the Tokyo subway and during the anthrax attacks in the fall of 2001 when tens of thousands of people who were not at risk for exposure obtained prescriptions for the antibiotics ciprofloxacin and doxycycline (Shaffer et al., 2003). Accurate and timely risk communication becomes particularly important in limiting the potential stress on the health care system because unaffected individuals flood services. This type of behavior is most likely to occur in the event of chemical, biological, radiological, or nuclear attack and is discussed further below in the section detailing the consequences of these types of terrorism.

Research from the disaster mental health field has developed models that stratify groups based on exposure level. These levels include those indirectly or remotely affected—individuals who are not in close geographic proximity to the incident, but who witness the event through the media; those who are negatively exposed through secondary effects such as an economic downturn; and those who experience the death of or immediate risk to a loved one from the terrorism event (i.e., relatives, friends, coworkers, rescue workers, witnesses). The populations that will be directly affected may vary according to the type of event (e.g., bombing; hijacking; chemical, biological, radiological, or nuclear attack). For example, a biological attack on the U.S. food supply may have a direct impact on agricultural workers through both physical and economic effects, and the resulting disruption may have an indirect impact on society as a whole. Given the large number of individuals, from those remotely to those directly exposed, who may be affected by a terrorism event, it is important to recognize variations among these exposed subpopulations in order to identify those who are most vulnerable to the psychological consequences of the event. This will allow for the focus of limited resources on prevention and intervention for those most in need.

Virtually all members of communities affected by terrorism are vulnerable to negative psychological outcomes. The type of vulnerability may vary substantially and may not always be obvious. Diverse variables that may enhance the prediction of adverse outcomes following a terrorism event are presented below in pre-event, event, and post-event temporal categories.

Pre-Event. Shalev (2001) reviewed a previously conducted meta-analysis examining predictors of adverse outcomes for traumatic events in general and concluded that preexisting factors have less influence on an individual than the disaster itself and subsequent factors such as community support. Some models of response propose that the impact of pre-existing factors is confounded with the dose of exposure; when the dose is less, the impact of pre-existing factors is more evident, and as the magnitude of the event increases, pre-event characteristics become less important. Regardless, these preexisting factors are useful to consider when planning service delivery because they allow for a better understanding of those who may be at increased risk and require particular attention.

Gender, age, experience, and personality have all been implicated in moderating adverse outcomes. Female gender has been associated with worse short-term outcomes in a number of studies of the general population after September 11, 2001 (e.g., North et al., 1999; Schlenger et al., 2002; Silver et al., 2002). Prior marital separation and preexisting physical illness have also been implicated in predicting greater psychological distress after these events (Silver et al., 2002).

As in studies of disasters, the pre-event experience of traumatic events may be related to psychological consequences following terrorism events. For example, the investigation of New York City public school students after September 11, 2001, found that nearly two-thirds of the students surveyed reported one or more prior traumatic events such as seeing someone killed or seriously injured and experiencing the violent or accidental death of a family member. In this sample, a history of prior traumatic events was associated with significantly increased rates of symptoms consistent with PTSD (Hoven et al., 2002). It is difficult, however, to discern the relative contributions of the prior traumatic events and the actual terrorism event to the reported symptoms given the lack of pre-event baseline data in this population.

Age has been identified as possibly moderating psychological responses to terrorism. While several studies examining adult populations have found no significant influence of age on the severity of psychological responses to terrorist attacks (e.g., Abenhaim et al., 1992), the psychological impact of terrorist attacks on children and adolescents is frequently noted as an area of concern as described above. One study reported that students in the fourth and fifth grades were significantly more likely than those in grades six through twelve to endorse symptoms consistent with PTSD after the September 11, 2001, terrorist attacks (Hoven et al., 2002). Further research is needed to determine if children and adolescents are at greater risk for psychological consequences than adults.

The disaster literature has also identified ethnic and racial minority status as a potential moderating factor on adverse outcomes. Norris and colleagues (2002b) reviewed studies that included ethnicity as a variable and found that among adults, ethnic majority groups had better outcomes after disasters than minorities in all of the samples. Among youth, however, the results were more variable. The research base examining racial and ethnic minority status as a factor predicting outcomes to terrorism events is extremely limited, although a few studies provide some indication. For example, Galea et al. (2002) found that Hispanic ethnicity predicted symptoms consistent with both PTSD and depression among Manhattan residents after the September 11, 2001, terrorist attacks. Similar results were found among New York City public school students after those attacks; Hispanic students were more likely than either African-American, white, or Asian students to have symptoms of PTSD (Hoven et al., 2002).

Findings from the disaster mental health literature have indicated that first responders and rescue workers are a population at risk for adverse psychological outcomes after responding to disasters (e.g., Duckworth, 1986; Jones, 1985; Weiss et al., 1995), likely due to their direct and often ongoing exposure to traumatic experiences. Findings after terrorism events reveal similar results. One study of New York City Fire Department rescue workers found a seventeenfold increase in stress-related incidents (e.g., depression, anxiety disorders, bereavement issues) during the 11-month period following the September 11, 2001, attacks as compared to the 11-month period preceding the attacks (Banauch et al., 2002). These data, however, may not represent the typical experiences of first responders and rescue workers because of the deaths of so many fellow firefighters in the immediate aftermath of the attacks. North and colleagues (2002b) found a PTSD rate of 13 percent among rescue workers in Oklahoma City. PTSD was associated with more days spent working at the site and more time spent in the central bombing pit. However, this study compared rescue workers to primary victims of the bombings and found that PTSD was significantly lower among rescue workers. The authors speculated that this may be related to characteristics of rescue workers such as preparedness, experience with job-related traumatic events, and self-selection for the type of work, as well as lower injury rates among rescue workers and exposure to education and debriefing aimed at mitigating psychological consequences (North et al., 2002b).

Event. While it is clear that certain populations may be particularly vulnerable to adverse outcomes following a terrorism event, there are factors related to the event itself that may affect the degree of impact. Findings from the disaster and other trauma literature have suggested that the duration and intensity of exposure to the traumatic event, including indirect exposures such as traumatic grief and loss, are some of the most important predictors of an adverse impact on subsequent functioning. Evidence suggests that terrorism events are similar to other traumatic events in this regard. As described earlier, psychological consequences will vary across the population in relation to the quality and extent of exposure: some people will experience direct physical trauma or threat of trauma; others, such as family members and friends, will experience grief and loss; and a wider population will be affected by secondary adversities and a general climate of fear. Silver and colleagues (2002) found that the degree of exposure to the September 11, 2001, attacks (as measured by a composite of proximity to the various attack sites, presence at a site, contact with a victim whether visually or by phone during the attacks, and degree of watching the events live on TV) was significantly predictive of psychological distress, more so than the degree of loss,6 although both exposure and degree of loss were associated with distress. Similarly, Schlenger et al. (2002) suggested that the amount of time spent watching television coverage predicted both PTSD symptomatology and general distress, although these authors were careful to note that this association did not necessarily imply causation (e.g., more symptomatic people could have been drawn to watching the television news coverage). See Box 2-4 for additional information on the role of the media during terrorism events.

Role of the Media During Terrorism Events. Speculation about the communicability of terror through media accounts has increased recently. Some evidence has revealed an association between exposure to media accounts of terrorist acts and psychological (more...)

Other important event-related characteristics include the duration and type of attack. Unlike other disasters, terrorism events may manifest as a single massive attack (e.g., Oklahoma City bombing), multisite event (e.g., events of September 11, 2001), multisite continuous or repeated events (e.g., anthrax attacks of 2001), or continuous or repeated events (e.g., terrorist attacks in Northern Ireland) (Ursano, 2002). The mechanism or type of attack also may moderate outcomes. Biological and radiological attacks may involve considerable on-going exposure to the threat and delayed emergence of physical symptoms, while an attack with conventional explosives will likely be a discrete event with obvious and more immediate injuries. The effects of cyberterrorism events, which have not been adequately studied, are largely unknown. These characteristics of terrorism events can determine the degree of population exposure, and the severity and magnitude of psychological consequences.

Hoaxes and copycat events may initially result in psychological consequences similar to those of actual terrorism events. Although the research base is extremely limited, the psychological impact of a hoax may be as great as that of a true threat. For example, Dougherty, et al. (2001) examined the psychological impact on victims of two incidents of anthrax threats that were later determined to be hoaxes and found evidence of distress symptoms. Results revealed that victims frequently reported a number of posttraumatic stress symptoms even after the hoax was announced. A similar relationship with adverse psychological consequences may exist with false alarms for terrorism events, although research in this area is also limited. False alarms and warnings that are given to people not at risk have implications for future preparedness and response since a “cry-wolf” syndrome may result in which people become less responsive to future warnings (NRC, 2002b).

Post-Event. A number of post-event factors may also help identify those at increased risk for negative psychological outcomes. The investigation by Galea and colleagues (2002) examining residents of Manhattan after the September 11, 2001, terrorist attacks found that post-event factors predicting PTSD symptoms included panic attack during or shortly after the attacks, and loss of possessions due to the attacks. Similarly, post-event factors predicting depression included panic attack during or shortly after the attacks, death of a friend or relative during the attacks, and job loss due to attacks.

Although many people will exhibit some manifestation of distress in the aftermath of a terrorism event, several specific symptoms have been identified as being more predictive of later psychiatric illness. These symptoms include feeling numb, withdrawn, or disconnected; isolation from others; and avoiding activities, places, or people that bring back memories of the event (North et al., 1999, see also Box 2-3). It may be important to screen for these specific symptoms during the post-event period in order to identify individuals who may require mental health care. The ways in which people cope with the stress of a terrorism event is also predictive of later outcomes. Silver and colleagues (2002) found that those who used active coping7 had less distress than those who demonstrated denial, defeatism, and self-distraction—indicating disengagement with coping—had greater distress.

Because terrorism, unlike natural disasters or human-caused technological failures, is a purposeful act by an individual or a group of individuals, terrorist acts are often perceived to be perpetrated by a specific ethnic, racial, or religious group. Recently, debate has increased about the controversial practice of profiling based on these characteristics for law enforcement purposes in the identification of potential terrorists. In addition, discrimination or stigmatization of the identified racial, ethnic, or religious group are potential outcomes of such perceptions, and may constitute threats to community cohesion and to the psychological well-being of those who are the targets of discrimination. Community cohesion can decrease as neighbors become suspicious of strangers and of one another. A multiethnic and multicultural population might exacerbate these fears. After the terrorist attacks on September 11, 2001, the number of hate crimes against Arabs, Muslims, and those perceived to be Arab or Muslim rose sharply (Human Rights Watch, 2002). Violent acts included murder, physical assaults, arson, vandalism of places of worship and other property damage, death threats, and public harassment. Most of these incidents occurred between September 11, 2001, and December 2001. According to Federal Bureau of Investigation (FBI, 2002) statistics, the number of anti-Muslim hate crimes rose from 28 in 2000 to 481 in 2001. Similar increases in the numbers of anti-Muslim hate crimes have been reported in relation to the Oklahoma City bombing, the crash of TWA Flight #800, and the Persian Gulf War (Human Rights Watch, 2002).

In contrast, terrorism events, like other disaster events, can also produce unique positive outcomes for the community. Because terrorism is generally directed at a population or subpopulation, there is often a significant growth of patriotism and pride for the population following the event. For example, after the terrorism events on September 11, 2001, many people reported an increased appreciation for the freedom afforded by living in the United States (Silver et al., 2002). People also reported closer relationships with their family members subsequent to those attacks (Silver et al., 2002).

Chemical, biological, radiological, and nuclear terrorism (CBRN) deserves special mention, given the unique characteristics. Such threats are unfamiliar, usually undetectable while they are dangerous, and often perceived as particularly reprehensible and unfair. These qualities present additional psychological challenges. The presence of an “incubation period” in which an individual may have been exposed to an agent but may not know the outcome is another unique and potentially stressful aspect of CBRN terrorism. In the case of a bombing or other physical terrorist attack, the individual will know immediately whether or not he or she has been physically harmed.

A particularly difficult challenge that may present in cases of CBRN terrorism is the differentiation of apparent anxiety in people due to the possibility of exposure to a chemical or biological agent from direct neuropsychological or behavioral changes due to exposure to the agent. The initial presentation of a chemical and biological weapon attack may be neuropsychological symptoms. For example, acute poisoning with a sub-lethal dose of an organic phosphorus compound (e.g., sarin) produces cognitive impairments characterized by confusion, difficulty in concentration, and drowsiness (Jones, 1995); individuals exposed to cyanide may initially present with anxiety and agitation, reflecting tissue hypoxia (Baskin and Rockwood, 2002); and exposure to fungal toxins can result in psychosis, somatic complaints, anxiety, agitation, and involuntary movements (Benedek et al., in press). Furthermore, physical manifestations of panic such as shortness of breath might be mistaken as symptoms of infection or contamination, which then becomes a self-reinforcing cycle as the individual's panic is increased by the shortness of breath, resulting in an exacerbation of this symptom.

Individuals with nonspecific somatic complaints such as nausea or weakness will be a great concern in the event of biological or chemical attacks when the presenting symptoms of exposure may be nonspecific and similar to other common conditions. For example, during the anthrax attacks in fall of 2001, the initial symptoms of infection mimicked viral syndrome and influenza-like symptoms. Many emergency physicians and primary care physicians were overwhelmed with individuals concerned about their exposure and requesting testing and/or treatment for anthrax exposure, which may or may not have occurred and for which tests were not always available. The extensive publicity about the anthrax threat likely increased self-monitoring for symptoms. This scenario was also seen among the Israeli civilian population during the Gulf War when people went to hospitals concerned that they had been exposed to nerve gas from Iraqi Scud missiles (Golan et al., 1992). The 1995 terrorist attack involving the nerve agent sarin in the Tokyo subway system also illustrated this phenomenon. Almost 75 percent of those who went to the hospital and were reported as “injured” showed no effects of exposure to sarin (Lillibridge et al., 1995). An investigation conducted by Ohbu and colleagues (1997) examined various psychological distress responses in survivors of the sarin gas attack. The individuals reported symptoms such as fear of subways (32 percent), sleep disturbances (29 percent), flashbacks (16 percent), and irritability (10 percent).