To date, most NASA missions have been "textbook" flights and all have ended with the successful recovery of the astronaut crew.1 There is a danger that this success record could obscure the  extraordinary, and sometimes heroic, efforts that were needed to complete some of these missions successfully. Space remains an extremely dangerous environment. A general history of emergencies which have arisen in space may serve to illustrate this point.
Experience in Space
The Mercury flights witnessed three emergencies, all related to reentry and splashdown. The hatch on MR-4 was blown prematurely after splashdown and astronaut Gus Grissom had to exit hurriedly and swim clear of his sinking craft. An emergency on MA-6 made it necessary for John Glenn to reenter the atmosphere uncertain if his heat shield was sufficiently secure to remain in place. On MA-7 Scott Carpenter overshot the designated splashdown area and a full hour intervened before ground control could be sure that he had landed and exited his spacecraft safely.
The Gemini and Apollo programs also had their share of difficulties. On Gemini 8, astronauts Armstrong and Scott experienced the first emergency to occur in space. After docking with an Agena rocket, the vehicle began to spin out of control. The crewmembers were able to escape by firing their retrorockets, returning to Earth 2 days ahead of schedule. On Apollo 11, the first moon landing, Commander Neil Armstrong was forced to take over control of the lunar module to avoid descending into a giant crater; a crater near-miss was also experienced by the crew of Apollo 16. On returning to Earth, the crew of Apollo 15 experienced a rough landing when one of their vehicular parachutes failed to deploy during final descent. The most critical U.S. emergency to date occurred on Apollo 13. With the spacecraft almost a quarter of a million miles from Earth, an oxygen tank exploded. The astronauts moved to the lunar lander for emergency return to Earth. Again, possible damage to the heat shield added to the concern during reentry.
There were also some dangerous situations during the Skylab series. Skylab 1 (unmanned) arrived in orbit with its meteorite/ thermal shield torn away, with a solar wing broken off, and with the second solar wing jammed. The Skylab 2 crew had the unenviable job of trying to correct these problems so that the main habitat could be made operational. Astronaut Paul Weitz engaged in hazardous extravehicular activity in an unsuccessful attempt to release the jammed solar panel, after which Commander Charles Conrad attempted for 4 hr to dock with the damaged Skylab. The docking finally succeeded, but the crew never really knew until the end of the mission  whether or not they would be able to undock for the return trip home. Working in extreme heat, the crew managed to deploy a parasol to shield the vehicle from the Sun, allowing the Skylab missions to proceed. On Skylab 3 an emergency flight home was contemplated for a time when a leak was detected in the command-module thruster.
Our information concerning Russian flights is incomplete, but we know that they have experienced emergencies, occasionally with severe consequences. Although the Russians had experienced difficulties with some of their earlier flights, they suffered their first known loss when Vladimir Komarov, flying Soyuz 1, was killed when the parachute controlling his landing craft became tangled. The second known Russian space tragedy occurred on Soyuz 11. Three Russian cosmonauts lost their lives when a valve failed to operate properly, causing a sudden decompression in their craft. The crews of Soyuz 13, Soyuz 15, Soyuz 21, and Soyuz 23 also apparently experienced some kind of space emergency, since they were returned to Earth hastily, under suboptimal landing conditions.
On Soyuz 26, cosmonaut Giorgi Grechko, returning from a space walk and closing the external hatch, found the monitoring instrument in the transfer compartment signaling a valve malfunction. Such a malfunction would have made repressurization of the compartment impossible. Without repressurization, neither he nor his space partner could enter Salyut, and therefore could not return to Soyuz for the trip back to Earth. The crisis was resolved when repressurization was attempted and it was discovered that the malfunction had been in the signal and not in the valve.
In the only case to date of an abort prior to orbital insertion, Soyuz 18 (renamed Soyuz 00) was returned to Earth after a failure occurred in an upper-stage rocket. During the abort the cosmonauts were said to have "indicated substantial concern," not only about the hazards of their return but also about the possibility of landing in China. After a search lasting 1 day, the cosmonauts were located in the foothills of the Altai mountain range and were returned home.
The Soyuz 33 flight, with its Soviet/Bulgarian crew, also ran into difficulty when failure in the propulsion system prevented docking with Salyut 6 and its resident Soyuz 32 crew. The crew of Soyuz 33 were forced to fly an unusual reentry profile for their emergency return to Earth, ending what Soviet Deputy Flight Director Viktor Blagov termed "the most complicated flight we ever had."
 Nor was the Apollo-Soyuz effort, which rendezvoused an American crew and a Soviet crew in space, without incident. First, there was a rough redocking of Apollo with Soyuz, which caused the Russian ground control to voice concern over the integrity of their vehicle. Then, after separation and while returning to Earth, the American crew apparently made an error in executing the landing sequence. As a result of actions taken to correct the original error, nitrogen tetroxide oxidizer, a highly poisonous gas, was released into the space capsule. After a very hard landing, the crew was able to escape the gas by donning oxygen masks, but not before one of their members lost consciousness.
This description of problems in space does not suggest that manned spaceflight has been more dangerous or less successful than anticipated. On the contrary, the safety record has been remarkable considering the incipient nature of this endeavor. It does point out that space is, and will continue to be, an extremely hostile and unforgiving environment. We can expect that added experience will tend to reduce the risk involved, but we can also anticipate that emergencies will continue to arise with some regularity.
Individual Response to Threat
Although it is unlikely that all future space travelers will possess the technical skills to deal with structural emergencies, all must possess the psychological constitution to withstand such threatening events. In the following section we will consider how individuals respond to fear-inducing situations generally, and the possible implications of these findings to the space environment.
Conceptual view of fear- The psychological literature contains a myriad of meanings of the words "fear" and "anxiety," as well as such related words as "stress" and "arousal." Spielberger (1972) describes the traditional distinction that is drawn between fear and anxiety: fear is the emotional response to a real or objective danger whereas anxiety is the fear- like reaction that occurs in the absence of specific, objectively threatening circumstances. Spielberger further suggests that in this traditional view, fear is proportionate to the magnitude of the danger, whereas anxiety exceeds or is unrelated to the danger.
Such distinctions are not universally applied. Some theorists view fear as following from anxiety (Epstein, 1967). For instance Danesh (1977) holds that a threat leads to anxiety, which in turn  leads to one of a number of secondary, Iearned responses. One such secondary response is fear, which prompts the individual to escape or withdraw; another is anger, which inclines the person to attack. Other theorists use the terms fear and anxiety to describe contrasting reactions (e.g., see Sarnoff and Zimbardo,1961). Still other theorists take the view that anxiety is a derivative of fear (see Izard,1972). To add to the confusion, a distinction is often drawn between fear and fright. In contrast with fright, fear describes an instrumental response which allows the threat to be perceived and problem-solving behavior to be initiated. Fright describes a purely emotional and noninstrumental response to an ambiguous situation, one which thwarts rather than initiates problem-solving activity. Several recent attempts to untangle the lexical confusion of this research area have focused attention on the feeling state of the individual. Epstein (1972) and SpieIberger (1972) suggest that if the feeling state in fear and anxiety is the same, no further distinction is needed.
In spite of the difficulties in describing this research area, there remain some substantive agreements. It is generally acknowledged that an emotional reaction to a threat of danger can have positive adaptive advantages, in that this reaction rallies the individual to act. It is also agreed that beyond a certain point the response becomes so strong that it impairs or even paralyzes action. There also appears to be a consensus that anxiety (and/or fear-fright) follows from stress. Spielberger notes that this latter conclusion "represents an important point of theoretical convergence in an otherwise chaotic literature' (Spielberger, 1972, p. 483).
The measurement of fear- There are two basic approaches to the measurement of fear: autonomic responses and subjective assessments. The former approach has the disadvantage of reflecting responses other than fear; the latter has the disadvantage of being colored by the perceived acceptability of reporting fear. In spite of these limitations, a positive correlation has been found between physiological measures of stress and subjective reports of fear, and between fear (however measured) and performance (Hodges and Spielberger, 1966; Thackray and Pearson, 1968). For certain situations, self-report has been found to be a convenient and a reliable predictor of performance (Pearson and Thackray, 1970).
Experimental findings- In this section we will consider the observational and experimental literature related to fear, with an eye to understanding the conditions under which fear could develop in space and the consequences that could follow.
 Noyes and Kletti (1976) report that, for some individuals involved in serious accidents, a complete depersonalization occurs in the split second or so prior to the accident. These individuals describe a perceived slowing of time just before the accident, a heightening of awareness, and most significantly, a seeming separation of the experiential self from the physical body. They report a feeling of having "left their bodies" and of witnessing the accident from some distance away as disinterested third parties. Depersonalization, out-of-body experiences (Mitchell, 1981), and similar phenomena have been reported over the years by many people and under a variety of circumstances. The significance of the Noyes and Kletti study is that these authors found a high degree of response consistency, suggesting a possible tendency to depersonalize in reaction to a severe threat. The accidents described in this study were all unavoidable. A significant question is whether such a tendency would prevail if individuals, by their actions, could ameliorate the situation. A further question concerns how widespread such a response is, and how individuals differ in this response.
A series of experiments involving less immediate but serious threats to life were conducted at Fort Ord, California, in the early 1960s (Berkun, 1964). One experimental group consisted of Army trainees who were led to believe that (1) an aircraft in which they were passengers was about to make an emergency landing, (2) the outpost at which they were stationed was an artillery impact area or (3) they had caused serious injury to a buddy by the faulty wiring of an explosive charge The soldiers involved in the tests were required to perform certain tasks related to their duties. In reality these tasks were carefully controlled performance measures. Trainees in each experimental condition performed significantly poorer than did subjects who performed similar tasks under nonthreatening control conditions.
The relationship between fear and performance found for Army trainees is confirmed in the work of Helmreich (1967) and Radloff and Helmreich (1969) These authors found that those divers of Sealab I I who reported themselves as most frightened performed worse and spent less time in the water than did their less frightened colleagues.
Although there appears to be a close tie-in between fear and impaired performance, the relationship between the stressful situation itself and the fear response is Iess predictable. In the Fort Ord study series, an experienced group of soldiers reported themselves to  be less distressed in the experimental than in the control condition, and their performance improved in the "life-threatening" situation.
Time of arousal - Not all threatening events occur instantaneously and without warning. Fenz and his colleagues (Fenz and Epstein, 1967, 1969; Fenz and Jones, 1972) have shown that when one has some time to consider the threat, the temporal pattern of fear response can vary significantly. Using the sport of parachuting, these authors have detailed the most complete work to date on time-dependent responses to life-threatening situations.
Employing three basic response measures (physiological responses to the stress situation, physiological reactions to words associated with parachuting, and self-reports of fear) Fenz and his collaborators found that experienced jumpers generally have a lower Ievel of anxiety than inexperienced jumpers. However, the main difference between the two groups is not in the level of response but rather in the pattern of response. For novice jumpers, fear response rises steadily from the morning of the jump until the "ready" signal preceding bailout, and drops to normal after landing. For experienced jumpers, the fear response peaks early on the day of the jump and drops to below normal just before bailout. Measurements taken after landing show that the fear response of experienced jumpers rises to normal shortly after landing.
In summary, the preponderance of data indicates that, unlike novice jumpers, experienced jumpers become fearful well in advance of the event, becoming calm as the event approaches. Similar results are reported by Capel, Youngblood, and Stewart (1970) and Knapp and Capel (1976), who found anxiety levels of aquanauts to be low prior to a hazardous dive, rising to normal immediately after the dive.2
 Mechanism of arousal - From the data on time of arousal, Fenz and his collaborators conclude that fear is inhibited under certain conditions, and that a particular pattern of inhibition is related to optimal performance. The optimal pattern (shown by experienced jumpers) allows fear to occur, but at a time considerably in advance of the fear event. Such a pattern calls attention to the danger, but leaves the person emotionally capable of performing necessary acts. Too much inhibition results in lack of attention to relevant details and consequently in poor performance, whereas too little inhibition leads to emotional chaos. These authors describe the pattern of novice jumpers as frequently reflecting an "all or nothing" inhibition process, with both extremes ill-suited to the task. Good performance results, not from allowing anxiety to be shut off completely, but rather from regulating the pace at which anxiety is experienced.
The question remains: What triggers the fear response to occur at a particular time? Fenz and his colleagues suggest that response corresponds to, and shifts with, decision activity. For instance, for the experienced parachutist, the decision point occurs on the morning of the jump. At that time the potential jumper assesses all conditions and decides whether or not to go. In this way, experienced jumpers have made their mental preparation and done their "worry work" in advance of the actual event (Fenz and Epstein, 1967). For the novice jumper, the final decision to make the jump is probably not made until the moment he or she hears the signal for bailout.
Individual differences- From the preceding discussion we can conclude that, although the mechanisms are complex, fear generally inhibits performance; however, optimal performance is achieved not by totally eliminating fear but by controlling the time and intensity of the fear experience. In this section we will turn to the question of how fear control relates to the personal and psychological characteristics of the individual, that is, to characteristics that might impact astronaut selection.
Danger control and fear control- Levanthal (1967, 1968,1970) underscores the importance of the direction of mental focus in determining how individuals will respond to a threatening situation. If the individuals focus their attention on the external forces creating the threat, they prepare themselves to avoid or reduce the danger. If they concentrate on their subjective reaction to the danger, they prepare themselves to control their fear. Levanthal has termed these alternate coping strategies danger control and fear control. The successful strategist must somehow do both, either by dealing with fear in  advance, as suggested above, or by dealing with both the danger and the fear (possibly by alternating attention between the two). Fenz (1973) found experienced parachutists to have an external orientation; this suggests that, although both danger and fear must be controlled, there is a success bias favoring those whose focus is predominantly external.
Levanthal and Lindsley (1972) imply that knowing certain personal characteristics of an individual (such as the ability to analyze and separate features of the perceptual field, or the inclination toward fantasizing) may help to predict an individual's inherent tendency to focus externally or internally in response to a threatening situation. These and related measures could be useful in studying the significance of external/internal focus on managing crises, including the potential crises of space.
Gender effects - Levanthal, Jones, and Trembly (1966) found that women report higher levels of fear response than men. Since men are generally constrained in expressing fear, it is frequently assumed that men experience a higher level of distress than their fear reports indicate. To test this possibility, Katlin and Hoffman (1976) examined the hypothesis that, for a given level of fear reported, men would experience a higher level of autonomic activity than women. These authors failed to confirm this hypothesis. Accepted at face value, this finding leads to two conclusions: (1) subjective reports are accurate measurements of disturbance and are sensitive enough to overcome even cultural biases, and (2) women actually experience more fear than men. However, the conclusions of Katlin and Hoffman should not be considered definitive. Although these authors report no statistically significant differences in autonomic response between men and women, the autonomic responses of women were less than those of men for each level of self-reported fear. Since the question of whether women actually experience more fear than men is extremely important for future spaceflight, this area needs further examination.
Birth order effects - Radloff and Helmreich (1969) found that first-born and only children did not perform as well in the hazardous environment of Sealab as did latter-borns (a finding which they report to be true also of fighter pilots during the Korean War). Lester (1969) also notes that there were few first-born or only children among the men on the climb of Mt. Everest, suggesting again the greater willingness of latter-borns to expose themselves to danger. Similar findings have been reported for women (Schachter, 1959;  Gerard and Rabbie,1961). However, results in this area have been far from consistent. Gerard and Rabbie (1961) found that, unlike the women in their study, first-born men showed lower fear-induced skin conductance than latter-born men, whereas Miller and Zimbardo (1966) found first-born women to be less fearful than their latter-born counterparts.
A variable that presents such shifting results would not seem to have much to contribute to the understanding of fear responses or spaceflight selection. However, one cannot avoid being impressed by the strength and consistency of effects within individual studies. It has frequently been observed that many of the astronauts that have flown to date are first-born children, and 20 of the original 31 Mercury astronaut candidates were only or eldest children (Ruff and Levy, 1959). Since these statistics continue to command attention, it would be helpful to understand more fully their relationship, if any, to performance under stress.
Training for danger- A practical application of the theoretical and experimental investigation of fear and its control lies in the possibility of training individuals to deal with dangerous situations. The objective or mechanical skills necessary to reduce or eliminate a particular danger obviously can and must be trained. But how does one train an individual to deal with danger itself? There are several approaches to this question.
The first involves training a person in a particular stress situation for the same or a similar stress situation. There is considerable evidence that a person can be trained to deal with the dangers of a particular situation. Babin and Loiko (1973) report success in increasing the efficiency of pilots in their reactions to on-board emergencies. Hammerton and Tickner (1968) have shown that the level of ground training makes a significant difference in the ability to perform a tracking task just before a first parachute jump. And, the "experience" predictive of good performance in the studies of sports parachuting (Fenz and Epstein, 1967, 1969; Fenz and Jones, 1972) might more accurately be called training.
Analyses based on opponents processes provide an explanation of how such danger training might work (see Epstein, 1967). As applied by Solomon and Corbit (1974), a particular stimulus arouses an effect (State A). When the stimulus terminates, a new state (State B) opposite to the first appears. This new state persists for a while and then dies out. States can change in their quality and  intensity with repeated stimulation. When this occurs, State A becomes weaker and State B becomes both stronger and longer lasting. For example, a person in a threatening situation experiences fear (State A) which, when the threat has been eliminated, gives way not to a neutral state, but to a state of elation or relief (State B). With repeated exposure, the aversive state becomes less pronounced and the reinforcing state becomes more pronounced and longer lasting. To the extent that this model accurately describes reactions to certain stimuli, the threatening stimulus becomes less fearful, and the experience of meeting the threat more rewarding.
A second major approach to stress training involves training an individual in one kind of stress situation for another kind of stress situation; that is, training a person to deal with danger generally. How much value this "irrelevant" stress training has for an individual is not totally clear, although many organizations which must deal with emergencies behave as if the value of such training were assured. For instance, a great deal of the training given to members of underwater demolition teams is for the purpose of pushing the individuals to their limits (Rubin and Rahe, 1974). Similarly, in describing stress training at the military academy, P. Patterson (1975) notes that cadets are stressed "for the purpose of experience, for the purpose of converting a young man into a professional soldier, and perhaps to identify cadets who cannot function under stress" (p. 2133). This explanation points out a second rationale for generalized stress training-its use as a screening device.
The Russian space program assumes that benefits follow from generalized stress training. Lieutenant General Georgy Beregovoi, head of crew training, believes that dealing with various life-threatening events prepares cosmonauts psychologically for the rigors of space. Parachute jumping and remote survival tests form the bases of this training program. Each cosmonaut must make at Ieast 100 parachute jumps. As training progresses, the cosmonauts are required to perform tasks of increasing difficulty before releasing their chutes. Survival tests are similarly structured. Crews are deposited in extremely inhospitable environments (including the Siberian forests, Cuban jungles, and the Black Sea) and survive only by their own wit and endurance. There are no rescue teams to help out if trainees get into trouble (Bluth, 1981 b).
The astronaut training program has not explicitly adopted the value of generalized stress training. However, elements of generalized stress training such as scuba diving, survival testing, and flight  training historically have been part of the astronaut indoctrination program. However, their inclusion has frequently been defended on the grounds that such skills could be required for spaceflight. Now, for some astronauts, flight training can no longer be linked to any obvious spaceflight need. As a result, flight training has been dropped as a requirement for certain candidates (e.g., Mission Specialists and Payload Specialists on Shuttle). This move has met with mixed reactions: concurrence on the part of those who believe that flight training has needlessly restricted NASA's pool of qualified applicants (see O'Leary, 1970) and concern on the part of those who believe that an important element in spaceflight preparation has been eliminated.3
Significant questions remain concerning whether generalized training to deal with danger is effective, and if so, what skills need to be developed. Research to help resolve the debate on generalized training for danger and research to further the understanding of preparing for specific danger could have direct relevance to future astronaut training.
Thus far, we have concentrated on those qualities which might affect an individual's ability to respond to a life-threatening situation. In this section we will consider responses which emerge, not at the level of the individual, but from the group as a whole, either during or subsequent to a crisis situation. Before considering responses that we believe are likely to occur in space, let us consider one that, although it has been raised as an issue, we believe is unlikely to occur.
Some have expressed the concern that, in the immediate aftermath of a spaceflight accident, panic4 could develop within the space crew, especially if the group were large and diverse (Hartman and Flinn, 1964). Early work on the panic phenomenon contained assumptions that made a panic response seem likely in a variety of situations. It was assumed, for instance, that emotion (as opposed to  rational thought) influenced one negatively, and that crowd membership had an essentially brutalizing effect on people in terms of facilitating emotion. In 1951 Mintz proposed a theory of panic based, not on emotional demands, but on the reward structure of the situation Looking at panic from a rewards perspective makes such behavior seem Iess likely, since in many situations the needs of individuals are best served if everyone behaves in an orderly and cooperative manner,
Assuming the rewards structure model, Rosengren, Arvidson, and Sturesson (1974) identified three conditions which, taken together, could lead to a panic response: an individual
1. sees a threat toward his or her own existence,
2. sees a possibility of escape, and
3. believes this possibility is soon to disappear (e.g., because it is not sufficient for all who wish to use it).
From this description one can see how panic, or panic flight, might develop in, for instance, a crowded, burning building. However, to the extent that panic does not contribute to an individual's chance for survival, as it is unlikely to in space, the probability of its occurrence is low. Individuals exposed to an accident in space may experience intense fear, but the probability of panic on-board is remote.
Responses to threat- W. Smith (1966) examined the interactions within a group of men facing the hazards of crossing an Antarctic crevice field. He found that reaction to danger was not uniform, but changed over thee, dividing into two major phases. The first or ineffective phase consisted of three parts and began with the group's displaying an inability to comprehend or to accept the seriousness of the situation (perhaps related to the reactions reported by Noyes and Kletti, 1976). Next came a period of energetic but incautious activity which was followed by a period of inactivity. The second or effective phase was marked by cautious action in which the group members recovered from the disorganization that had marked the earlier phase and began to work together toward resolving their problem. It would be of interest to know if the phases observed in this experience represent a general pattern of group response to crisis, and if so, how training might eliminate or shorten the ineffective period.
 An area of particular importance in marshalling a group to deal with a crisis involves the coordination of team members. Insight into this area can be gained from examination of aircraft crews. There is growing evidence that inadequate management of cockpit personnel can result in crew error, and ultimately in accidents or near-accidents (Ruffell Smith, 1979; Lauber, 1980). The inability of aircraft crews to coordinate their activities during critical periods has been shown to relate to communication styles, social-psychological factors, and personality characteristics (Foushee, 1982). Significantly, problems of crew coordination similar to those found among flight crews have been identified by Russian cosmonauts. In noting these issues, Leonov and Lebedev (1972) stress the necessity for understanding, and even anticipating, the responses of other crewmembers, a condition which they call "homeostatic balance.''5 These authors suggest that lack of such balance can result in failure to coordinate activities because of "the inability to understand one another in critical situations, the asynchronous nature of psychomotor reactions, and differences in the degree of attention, thinking, and other inherent or acquired individual characteristics . . ." (p. 6). Although the goal of homeostatic balance among crew members is intuitively appealing, more work is needed to determine how such balance is established and maintained.
One reaction which can be predicted with some confidence is a tendency of crewmembers to come together in response to a threat (Gerard and Rabbie, 1961). From a planning perspective, it is important to understand why this happens. It is usually assumed that persons who find themselves in a fearful situation seek the company of others to determine just how they should feel or behave (Schachter, 1959; Gerard and Rabbie, 1961; Wrightsman, 1960; Zimbardo and Formica, 1963). However, Miller and Zimbardo (1966) and Sarnoff and Zimbardo (1961) have found that fear affiliation goes beyond information seeking, since individuals seek out others with certain compatible personal characteristics. It would be desirable to understand more fully what needs are addressed by affiliation in a crisis to  ensure that contacts which are available are supportive of both the practical and the emotional requirements of the situation
A crisis is accompanied by a general state of instability within the group (Hartman and Flinn, 1964) Effective management of a crisis requires, then, that certain actions be taken prior to and following an incident, as well as during the crisis itself. For instance, planners and managers must identify in advance individuals whose specific skills could be used to achieve particular goals, and then set up the infrastructures to engage these skills. The quality of leadership displayed during a crisis will determine to a large extent how quickly stability can be restored (Llano, 1955).
Post-emergency concerns- The disaster literature provides evidence that threatening situations do not necessarily lead to antisocial behavior. On the contrary, it appears that in most crises altruistic behavior becomes the norm (Quarantelli, 1978). However, when a normal routine is reestablished, a different kind of behavior may emerge.
Scapegoating - Attribution of blame is a common occurrence during post-crisis recovery. Based on the work of Bucher (1957), of Veltfore and Lee (1943), and of Drabeck and Quarantelli (1967), Wenger (1978) concludes that beliefs about blame take a period of time to evolve. Accusations of blame can be expected to emerge, not in the immediate aftermath of a crisis, but at some later time.
Blaming or scapegoating has been found to occur in those circumstances in which human action is involved and in which there is the possibility of recurrence. Bucher opines that blame results from seeking an explanation to something which cannot be explained satisfactorily in conventional terms. In other words, blaming or scapegoating serves a psychological need. It is an attempt to control the future by creating a structure whereby inexplicable events become explicable. Apparently, blaming also has an instrumental effect. A person is more likely to offer help if the injured individual implies that the person is to blame (Schwartz and David, 1976).
People in certain positions are more likely to be blamed than Others. The lowly ranked individual is usually not an acceptable scapegoat, and the tendency is to reject that person in favor of someone higher in authority (Bucher, 1957). However, the highest ranked individual may also be an unacceptable scapegoat, since condemning him or her tends to condemn the values of the organization.
 Focusing blame on the second- or third-in-charge would seem to satisfy the psychological needs associated with scapegoating, while leaving the values of the organization unchallenged. Excluding the "person at the top" from blame would appear to be more functional for group members than for outsiders. In terms of space travel, one might expect that crewmembers will hold a second- or third-in-command responsible, while ground control might look to the commander.
Since blaming or scapegoating gives one some control over the action of others, and at least illusory control over future events, it is reasonable to expect that a crisis in space could result in scapegoating. Specific efforts need to be taken to deal with this post-crisis phenomenon.
Hysterical contagion - In hysteria, or hysterical conversion, a person demonstrates bodily symptoms which are without an ascribable physiological basis. Hysteria occurs when a person experiences extreme anxiety which is relieved by converting the anxiety to physical symptoms (Frazier and Carr, 1974). In hysterical contagion, the symptoms are passed from one individual to another, triggered by a precipitating event which heightens the sense of an immediate and tangible threat (Kerckhoff and Back,1968).
Although the general nature and course of hysterical contagion are reasonably well understood, there is some controversy as to the dynamics of this rather strange phenomenon. Kerckhoff and Back (1968), Smelser (1962), and Klapp (1972) favor the explanation that hysterical contagion is part of a negative cycle of stress illness-greater stress. Gehlen (1978), on the other hand, argues that in hysterical contagion the behavior adopted is positively rather than negatively reinforcing, allowing the individual to relieve stress by participating in the benefits associated with the sick role. The latter explanation seems to describe the data of hysterical contagion better than does the runaway-fear hypothesis. Hysterical contagion, at least under Earth conditions, has a brief and well-defined life span of a few days, after which the symptoms rapidly disappear. An interesting aspect of the hysterical contagion phenomenon is that it is probably related to a conscientious work ethic. Kerkhoff and Back found that hysterical contagion is unlikely to occur among individuals who are willing to take an occasional day off.
An episode of hysterical contagion in the aftermath of a fear-inducing event is unlikely to occur in space in the foreseeable future.
 In fact, our present generation of astronauts has incurred some displeasure by trying to conceal, rather than disseminate, information concerning physical symptoms (Cooper, 1976). However, future spacecrews are expected to be large and to include individuals with backgrounds different from those of today's astronauts. Further, the conditions of space would seem to support the development of this phenomenon These conditions include an elevated level of anxiety associated with living in space, a highly directed work-oriented sub-society, and a closed environment in which information would travel rapidly and in which social pressure could encourage the spread of symptoms. It seems that some cathartic measures will need to be devised for diffusing anxiety in space, both for routine prevention and to avoid the possibility of post-crisis maladaptive responses.
Implications for Space
The research reported here suggests many parallels to space. We can expect that space travelers will continue to be subjected to fearful, and even life-threatening situations. This realization could help shape our training programs and influence our selection criteria. It is imperative that space travelers learn to recognize and acknowledge threat at the earliest moment, so that valuable response time can be saved. And they must deal with both the threat and their emotional response to it in an effective manner. Since many of the potential challenges of space cannot be fully anticipated, it is important to determine if generalized stress training can be of benefit.
A crisis situation can threaten the integrity and stability of a group. Structures and procedures must be in place that allow space travelers to coordinate their activities with minimal loss of time. The role of the leader or leaders and the effectiveness of coordination among crewmembers will be crucial in formulating group responses to a life-threatening event.
Some mechanisms for dealing with stress can result in unacceptable consequences. For instance, in the aftermath of a threatening but poorly understood event, blaming or scapegoating is a common response. Such blaming in the confinement of space could place an intolerable burden on an individual; and if, as is likely, the individual were in a position of leadership, blaming could contribute to further group instability. Space travelers must be helped to control fear without resorting to mechanisms which could themselves threaten the welfare of the community.
1 Three American astronauts were killed in a capsule fire on January 1967, while training for an Apollo mission.
2 Although data showing different patterns of responses for novices and experienced jumpers are quite convincing, it should be noted that not all studies show the same consistent pattern. Hammerton and Tickner (1968), using heartrate measures, found no differences among three groups: very experienced jumpers, inexperienced jumpers with full ground training, and inexperienced jumpers with abbreviated ground training. They did, however, find performance differences favoring training. Basowitz et al. (Basowitz, Harold, Harold Peraky, Sheldon J. Korchin, and Roy R. Grinker. Anxiety and Stress, McGraw-Hill:NY (1955)) found that base Ievels of anxiety in airborne trainees are often low during the initial stages of training, and rise after training is completed. Capel et al. ( 1970), examining anxiety in divers, reported that anxiety control became more difficult with experience. Thus, although experience may be a highly significant factor in the regulation of fear response, it appears that other factors also must he considered.
3 A difference in training among classes of astronauts raises broad issues For instance, a lack of shared experience may lead to the perception that some crew members do not have the appropriate "membership characteristics." This issue is considered in chapter V.
4 Panic (also known as "panic flight response") usually refers to the frantic, nondirected activity sometimes observed in crowds in response to a severely threatening event.
5 The concept of homeostatic balance as used here follows a similar use of the term in the family therapy literature. Since Jackson's original use of "homeostasis" or "homeostatic equilibrium" to describe the emotional state which every family seeks to maintain, the term has been used widely in conjunction with events or behaviors that contribute to group equilibrium (Jackson D. D.; The Question of Family Homeostasis. Psychoanalytic Quarterly Supp. 31, 79-99, 1957).