For decades, the world has benefited from continual improvement in commercial air transport safety. However, in some high-profile accidents over the last few years, airline pilots failed to react as expected when suddenly surprised by critical situations. In some cases, these pilots reacted ineffectively or inappropriately — or even failed to react at all in a timely manner.
Recent interviews with Australian airline pilots, a flight simulator experiment, analysis of selected reports on airline accidents and serious incidents, and a literature review including insights from neuroscience and psychology suggest that pathological reactions induced by severe startle — or by consequent acute stress reactions called freezing and denial — may occur more often than the aviation community realizes. Some subject specialists label extreme startle as strong or serious.
Any such inaction-type behavior potentially has significant effects on aviation safety, possibly resulting in undesired aircraft states, serious incidents or accidents. A doctoral thesis by this article’s primary author (Wayne Martin) concludes that further research — especially development of interventions suitable for future training — is warranted. Such additional work could validate the assumption that by exposing pilots in the simulator to unusual critical events, they are likely to develop both specific and generic strategies for dealing with them. Ideally, such exposure also would engender a greater sense of self-efficacy, in turn making pilots significantly less likely to experience acute stress reactions after real-world severe startle.
Humans are particularly susceptible to acute stress reactions, and unexpected, threatening and critical events often present circumstances in which some individuals fail to cope well. The physiological stress reaction has been shown to have severe effects on working memory and other cognitive functions, with constructive thoughts being replaced by task-irrelevant, anxious thoughts. Acute stress reaction, which is associated with an appraisal of threat, can create situations in some people in which they are overwhelmed and freeze, or institute a coping mechanism such as freezing or denial.
Paralyzed With Fear
Acute stress reactions are common during life’s emergencies. One only has to see human behavior shown on the six o’clock evening news as a disaster unfolds somewhere around the world. Earthquakes, floods, fires, ship sinkings, oil rig disasters or train wrecks often turn up a mixture of behaviors. Studies by several researchers,1,2,3 through eyewitness accounts and interviews with survivors, have looked at why some people survived a disaster and others did not. The survivors often have reported seeing people who were apparently paralyzed with fear and incapable of movement, even when such movement would have helped them survive.
Inaction in the face of imminent threat especially raises concern from the aviation safety perspective. Airline passenger behavior during aircraft accidents, for example, has followed similar patterns, with some researchers4,5 finding that even in simulated evacuation trials, behavioral inaction was displayed by a number of the passengers. One study suggests that 10 to 15 percent of people typically display such pathological behavior when faced with life-threatening situations,3 and real-life examples exist such as a fatal aircraft fire6 after a rejected takeoff known to have involved passive inaction among passengers.
This inaction, which is most likely an acute stress reaction to an overwhelmingly threatening stimulus, may be due to an elementary freezing mechanism within the brain7,8 or indeed to a coping/defense mechanism that seeks to deny the existence or severity of the threat.9,10 Inaction also may be the result of the severe startle, and experiments by other researchers have shown that cognitive and dexterous (that is, hand/foot dexterity) impairments could last for up to 30 seconds following this degree of startle.11,12,13,14
While these reactions could be considered typical among a small fraction of the “innocent” participants in an unfolding non-aviation disaster, people expect that professional pilots — who are generally well trained, very experienced and presumably endowed with the “right stuff”15 — will nonchalantly, competently and flawlessly deal with critical emergencies to avoid disaster. Unfortunately, this may not be the case.
Extremely high aircraft reliability has become the norm, so official findings after an aircraft accident or serious incident more often are peppered with human failings involving the pilots. In some relatively recent fatal accidents, the findings showed that flight crews mishandled critical events and failed to recover the aircraft [ASW, 8/12, p. 14; ASW, 6/10, p. 32; ASW, 3/10, p. 20; ASW, 4/11, p. 46]. Typically, there was some delay in acting, or incorrect action taken, which exacerbated the problem.
Although airline pilots routinely practice engine failures, engine fires, depressurizations and major system malfunctions, the types of critical events that prevail in recent accident data are commonly regarded as “black swan” events, that is, highly unusual.16 They involved unexpected situations in which pilots became very surprised and/or overwhelmed. Moreover, the response to severe startle — or subsequent acute stress reactions of freeze or denial — is sometimes exacerbated by the flight crew’s conditioned expectation for things never going wrong. This unintentional sense of complacency is born of the ubiquitous normality in line operations, week-in and week-out for long periods of time.
Central to the acute stress reaction is the person’s appraisal that some particular stimulus is threatening. Some researchers17 have described appraisal as “an evaluative process that determines why and to what extent a particular transaction or series of transactions between the person and the environment is stressful.” They further have suggested that appraisal involves two distinct processes: primary appraisal, which determines level of threat, and secondary appraisal, which determines an appropriate method of coping. This process is very rapid and appears to precede conscious-thought processing in the cortex of the brain.18,19 This fact is clearly advantageous in situations when immediate action is required but may induce a pathological acute stress reaction that is unwarranted.
If they appraise their situation as threatening, humans involuntarily will apply some form of unconscious emotional homeostatic (stress-relieving) coping or conscious defense mechanism. This may take the form of trying to fix the problem. That is an entirely appropriate method, which is employed by most pilots in most situations. If no immediate fix is at hand, however, or the situation is appraised as being overwhelming, then the possibility exists that some form of emotionally focused coping mechanism will be employed.
Emotionally focused coping, however, is largely pathological and may include elements such as avoidance, denial, self-deception or reality distortion.17,20,21 These coping mechanisms can have severe effects on the constructive processing of information, problem solving and decision making. In the aviation context, this is very problematic in critical situations. One useful conceptual model (Figure 1) illustrates this flow of threat, appraisal and information processing.22
The startle reflex is a normal and universal human response to some unexpected/surprising stimulus. When a person appraises the stimulus as threatening, activation of the sympathetic nervous system triggers widespread and rapid changes in the body. This arousal, which is also associated with the acute stress reaction, is generally known as the fight or flight response and has been shown to have significant effects on cognitive and psychomotor processes.
The startle reflex invokes a pattern of aversive movement (that is, away from the stimulus) and aligns attentional resources to the source of the stimulus. This process is remarkably fast, with first signs of reaction occurring in as little as 14 milliseconds (ms) in some tests on humans.26,27 Given that cognitive processing of new stimuli takes more than 500 ms,28 that “quick and dirty” reflexive reaction is clearly an innate process for avoiding harm.
The brain’s amygdala region, which is strongly associated with emotional memory of fear, appears to be where initial appraisal of threat is made. Projections from the amygdala then initiate the startle reflex and, if the threat persists, the full startle or surprise reaction.7,8 Essentially, this is still the fight or flight response, the same process invoked during acute stress reactions.
The rapid changes in the body’s systems result from arousal of the sympathetic nervous system. Changes include increasing heart and respiration rates, routing more blood flow to vital organs, and introducing hormones such as adrenaline (epinephrine) into the bloodstream.29,30 Other characteristic changes are blinking and contraction of arm and leg muscles.
As noted, other research11,12,13,14 has shown that cognitive and dexterous impairment can last for up to 30 seconds during reaction to severe startle. This has significant implications for aviation because sudden, unexpected and critical events are typical of aircraft emergencies. A significant number of accidents and serious incidents reviewed for the thesis — in which pilots performed less than optimally during critical events but startle/acute stress reaction was not an official cause — were cases in which startle may have contributed to a poor outcome.
The simulator experiment involved an airline’s large commercial jet simulator flown by a sample of 18 type-rated pilots. This part of the study found that seven of the 18 performed very poorly during a critical event when startled (Figure 2, p. 32). In the experiment, a startling stimulus was introduced 40 ft above decision altitude on an approach where the cloud base was 100 ft below the minima. While five pilots performed nominally and six displayed some slight reactionary delay, the seven showed either impulsive behavior (immediate go-around) or significant delays in reactions.
Three pilots continued descent to below 100 ft above ground level (AGL), and two pilots continued to land despite severely unstable approaches. Enhanced ground proximity warning system “PULL UP” warnings resulted on two of the three approaches that had become unstable and had continued below 100 ft AGL. While this was a relatively small sample size, the results were both statistically and qualitatively significant, with the majority of pilots admitting to having experienced physiological and cognitive effects following the startle.
Freezing is an acute stress reaction that may be a conscious or subconscious method of dealing with an overwhelming stressor. One study3 defined it as “a stress coping mechanism entailing a subconscious mechanism for stress relief, or a conscious disbelief that the phenomenon is actually occurring.” Freezing is generally a major breakdown in the normally integrated cognitive processes within the brain. Sometimes the people affected remain aware of what is going on around them, but are incapable of taking any participative behavioral and/or cognitive action.
This reaction has been relatively common during non-aviation disasters and also has been noted in real-life and simulated aircraft accidents and evacuations. It appears that in freezing, the cognitive processes required to initiate action are overcome by an acute sense of dread, with the working memory being consumed by irrelevant thoughts of fearful outcome. “Paralyzed” or “petrified with fear” have been common recollections from some people who have survived such critical situations.
Unlike denial, which is quite difficult to quantify, most people are familiar with the concept of freezing under conditions of acute stress. Like “a deer caught in the headlights,” a popular analogous phrase for this phenomenon, freezing is not uncommon in aviation accident and incident data.
Freezing also has been described as a response to overwhelming threat in which, at the rudimentary subconscious level, the brain is unable to cope with the complexity and danger presented by sudden circumstances. Reports from survivors3 of an oil rig fire and a ship sinking, for example, described people who were simply frozen or paralyzed and unable to save themselves, despite encouragement or abuse from other passengers.
Similarly, the known aviation cases include pilots who have simply frozen during critical events. In one accident, the captain apparently experienced freezing after commencing a rejected takeoff.24,25 He closed the thrust levers, but failed to brake or select reverse thrust, simply staring straight ahead. The aircraft ran off the end of the runway at 70 kt, killing two people.
During part of the data collection for the thesis — interviews with a sample of pilots who had experienced critical events/emergencies during airline operations — one pilot recalled an actual event in which the captain froze during an approach, having set up a high rate of descent. The aircraft continued to descend well below glide path until becoming visual at very low level on a collision course with an apartment building. Fortunately, once visual, the captain recovered, and a last-minute evasive maneuver narrowly avoided the building. The first officer, who repeatedly had tried to alert the captain to the glide path deviation, also tried several times to take over control and even resorted unsuccessfully to hitting the captain to gain control.
Another of these interviewees described a situation in which a military pilot under instruction, while practicing spins with high rotation rate, simply froze during recovery from a spin. The student became unresponsive, and the instructor pilot had to physically hit the student to get him to release his iron grip on the controls. A successful recovery was finally made close to bailout altitude.
Denial also is an emotionally focused coping mechanism, and, like freezing, a very rudimentary human process. If a person appraises the stimulus as being particularly threatening, and this mechanism is implicitly invoked, then the stressful stimulus may simply be ignored.
Denial also appears to have a subconscious strategic purpose. This is remarkably common, particularly in people with life-threatening illnesses. Many would ignore the symptoms for some time rather than confront the stressful issue of their mortality.
Denying the threat’s existence can be very effective in relieving stress; however, continual reappraisal and dynamic denial are required for this coping mechanism to persist. Dynamic denial occurs when the flow of critical information is not continually processed as part of this pathological, acute-stress coping mechanism.
While this could be problematic when situations such as deteriorating weather or aircraft status compound the threat, the more immediate stressors — those conducive to dynamic denial — generally are of greater concern in critical events. So dynamic denial could have severe implications in airborne critical events because of the careful analysis and logical problem solving required.
The airline pilot interviews for the thesis revealed that short-term denial was relatively prevalent during these events, with participants reporting that some level of denial had been experienced in 15 of the 45 events they recalled. This was generally short-term denial, and it did not turn out to be of catastrophic consequence. However, it raises the question of how many fatal accidents have involved denial, with the pilots never achieving recovery or with recovery being delayed too long.
The pilots’ interview responses also indicated that such brief periods of denial were not unusual, although in all of the situations that interviewees discussed, denial was quickly overcome as rational processing kicked in. Dynamic denial, if it had persisted, could have been particularly detrimental to the outcome of the situation, although it is impossible to tell from accident data whether denial was involved. However, there are several examples of instances in which pilots took no action at a time when intervention was required, indicating that dynamic denial is at least a possibility. Further research in this area is required.
One subject specialist23 even has described a pathological taxonomy of seven different stages: denial of personal relevance, denial of urgency, denial of vulnerability, denial of affect, denial of affect relevance, denial of threatening information and denial of information. While the early stages are mildly concerning in the aviation context, the latter stages — when threatening information or all information is denied — are particularly worrisome in the aviation safety context.
In summary, if a critical situation arises during flight operations but an individual pilot’s brain unconsciously and involuntarily ignores the cues for threats presented, then the chances of recovering are substantially reduced.
Wayne L. Martin, a Boeing 777 first officer for Virgin Australia Airlines, has submitted a doctoral thesis on this subject to Griffith University, Brisbane, Australia. His career includes work in human factors and airline pilot training, and he is a member of the International Pilot Training Consortium Working Group on Training Practices. Patrick S. Murray, an associate professor and director of the Griffith University Aerospace Strategic Study Centre, is a member of the Line Operations Safety Audit Collaborative currently conducting research on regional airline safety. His career includes experience as a military pilot and airline pilot, and in a senior position at the Australian Civil Aviation Safety Authority. Paul R. Bates, an associate professor and head of aviation at Griffith University, chairs the Outreach Committee of the International Civil Aviation Organization Next Generation of Aviation Professionals Task Force.
- Fritz, C.E., and E.S. Marks. 1974. “The NORC studies of human behaviour in disaster.” Journal of Social Issues 10 (3): 26–41.
- Glass, A.J. 1959. “Psychological aspects of disaster.” JAMA 171: 222–225.
- Leach, J. 2004. ‘“Why People ‘Freeze’ in an Emergency: Temporal and Cognitive Constraints on Survival Responses.” Aviation, Space, and Environmental Medicine 75 (6): 539–542.
- Muir, H.C., C. Marrison, and A. Evans. 1989. “Aircraft evacuations: The effect of passenger motivation and cabin configuration adjacent to the exit.” London, U.K.: Civil Aviation Authority, CAA Paper 89019.
- Muir, H.C., D.M. Bottomley, and C. Marrison. 1996. “Effects of cabin configuration and motivation on evacuation behaviour and rates of egress.” International Journal of Aviation Psychology 6 (1) 57–77.
- U.K. Air Accidents Investigation Branch. 1988. “Report on the accident to Boeing 737-236 Series 1, G-BGJL at Manchester International Airport on 22 August 1985 (Aircraft Accident Report 8/88).” London: Her Majesty’s Stationery Office.
- Le Doux, J.E. 1996. The Emotional Brain. New York: Simon and Schuster.
- Le Doux, J.E. 2000. “Emotion circuits in the brain.” Annual Review of Neuroscience 23: 155–184.
- Carver, C.S., M.F. Scheier, and J.K. Weintraub. 1989. “Assessing coping strategies: A theoretically based approach.” Journal of Personality and Social Psychology 56 (2): 267-283.
- Folkman, S., R.S. Lazarus, C. Dunkel-Schetter, A. DeLongis, and R.J. Gruen. 1986. “Dynamics of a stressful encounter: Cognitive appraisal, coping, and encounter outcomes.” Journal of Personality and Social Psychology 50 (5): 992–1003.
- Thackray, R.I., and R.M. Touchstone. 1983. “Rate of recovery and subsequent radar monitoring performance following a simulated emergency involving startle.” U.S. Federal Aviation Administration Office of Aviation Medicine, Report no. FAA-AM-83-13.
- Vlasek, M. 1969. “Effect of startle stimuli on performance.” Aerospace Medicine 40: 124-128.
- Woodhead, M.M. 1959. “Effect of brief noise on decision making.” Journal of The Acoustic Society of America 31: 1329–1331.
- Woodhead. 1969. “Performing a visual task in the vicinity of reproduced sonic bangs.” Journal of Sound Vibration 9: 121–125.
- Wolfe, T. 2001. The Right Stuff. New York: Bantam Books.
- Taleb, N.N. 2010. The Black Swan. New York: Random House Publishing.
- Lazarus, R.S., and S. Folkman. 1984. Stress, Appraisal, and Coping. New York: Springer.
- Duckworth, K.L., J.A. Bargh, M. Garcia, and S. Chaiken. 2002. “The automatic evaluation of novel stimuli.” Psychological Science 13: 513–519.
- Zajonc, R.B. 1980. “Feeling and thinking: Preferences need no inferences.” American Psychologist 35 (2): 151–175.
- Lazarus, R.S. 1966. Psychological Stress and the Coping Process. New York: McGraw-Hill.
- Monat, A., and R.S. Lazarus. 1991. Stress and Coping. New York: Columbia University Press.
- Martin, W.L., P.S. Murray, and P.R. Bates. 2010. “The effects of stress on pilot reactions to unexpected, novel, and emergency events.” In Proceedings of the 9th Symposium of the Australian Aviation Psychology Association, Sydney, 19–22 April, 2010.
- Breznitz, S. 1983. “The seven kinds of denial.” In S. Breznitz (Ed.). The Denial of Stress. New York: International Universities Press.
- Heaslip, T.W., N. Hull, R.K. McLeod, and R.K. Vermil. 1991. “The frozen pilot syndrome.” In Proceedings of the 6th International Symposium on Aviation Psychology, Columbus, Ohio: 782–787.
- Stokes, A.F., and K. Kite. 1994. Flight Stress: Stress, Fatigue, and Performance in Aviation. Aldershot, UK: Ashgate.
- Simons, R.C. 1996. Boo!: Culture, experience, and the startle reflex. U.S.: Oxford University Press.
- Yeomans, J.S., and P.W. Frankland. 1996. “The acoustic startle reflex: Neurons and connections.” Brain Research Reviews 21: 301-314.
- Asli, O., and M.A. Flaten. 2012. “In the blink of an eye: Investigating the role of awareness in fear responding by measuring the latency of startle potentiation.” Brain Science 2: 61–84.
- Jansen, A.S.P., X. Van Nguyen, V. Karpitsky, T.C. Mettenleiter, and A.D. Loewy. 1995. “Central command neurons of the sympathetic nervous system: Basis of the fight or flight response.” Science 270 (5236): 644–646.
- Stratakis, C.A., and G.P. Chrousos. 1995. “Neuroendocrinology and Pathophysiology of the Stress System.” Annals of the New York Academy of Sciences 771: 1–18.