Veridical perception of visual cues is necessary for spatial orientation and controlling our movements as we navigate within our environment. Driving and athletics are two arenas with which everyone can associate in terms of visually guided behavior and successful execution of desired goals. In typical daily life our interface with the environment consists of our feet on or near the ground, movements in the left-right and/or fore-aft direction as well as a one-gravitational force (1-G) acting vertically on our bodies with the horizon straight ahead. The interpretation of visual cues from the environment and the perception of vestibular inputs as we maneuver ourselves are founded on these typical constants, leading to confidence about where our feet are, where the horizon is, and which is “up.”
Aviation, however, allows the human operator to accomplish visually guided actions not experienced anywhere else. With the additional spatial dimension of altitude and the possibility of extreme vertical movement, combined with potential extreme velocities and accelerations in the left-right and fore-aft directions, flying poses challenges to humans that are not faced in other domains. When flying, obtaining and maintaining spatial orientation is predominantly accomplished by the visual system. Thus, while flying 30 m (100 ft) above the ground at high speeds or controlling the aircraft for a night visual approach to landing, pilots must accurately perceive and interpret environmental cues with their eyes. Herein lies the problem; the human body is not physiologically prepared to cope with these extreme and sometimes violent movements and forces that occur in aviation. These visual perception challenges must be recognized and appreciated by pilots and aviation researchers.
Despite physiological limitations for sensing and perceiving their aviation environment, pilots can often make the required visual judgments with a high degree of accuracy and precision. At the same time, however, visual illusions and misjudgments have been cited as the probable cause of numerous aviation accidents, and in spite of technological and instructional efforts to remedy some of the problems associated with visual perception in aviation, mishaps of this type continue to occur. Clearly, understanding the role of visual perception in aviation is key to improving pilot performance and reducing aviation mishaps. Furthermore, with the implementation of enhanced and synthetic visual systems, the next generation of aviation is banking heavily on knowledge of visual perception.
Over the years numerous articles, pamphlets, and books have been written on the topics of spatial disorientation and visual illusions in aviation (e.g., Benson, 1988; Cocquyt, 1953; Kern, 2002; Kraft and Elworth, 1969; Lessard, 2000; Newman, 2005; Ostinga et al., 1999; Pitts, 1967; Previc, 2004; Schiff, 1990 and 1994; Wulfeck, Weisz, and Raben, 1958; as well as the Federal Aviation Administration flying safety pamphlets, military flying manuals). In 2007 the Australian Transport Safety Bureau published An overview of spatial disorientation as a factor in aviation accidents and incidents. Also, magazines have dedicated entire issues to the subject. For instance, IEEE Engineering in Medicine and Biology, in their March/April 2000 edition, addressed aeronautical illusions, and the Naval Aviation magazine, Approach, in May/June 2004, did the same.
This book intends to update and synthesize the previous work to provide the reader with a single resource for comprehensive and detailed explanations of visual disorientation as well as the physiological and perceptual background of the visual system associated with aviation-related perceptual illusions. Vestibular physiology and disorientation is also presented as it is highly integrated with our body’s spatial orientation system. Examples of aircraft accidents are included to illustrate failed visual perception and spatial orientation and to demonstrate that pilots have been and are still today far too confident in their limited visual perceptual capabilities when flying. It is not the intent to disparage any of the pilots involved but rather to ensure that others learn from their experiences. The objective of this book is to help educate pilots and others regarding the seduction of visual misperception, with the intention to provide not only a resource for pilots but also a starting point for further research into aviation visual perception.
Visual perception within aviation is not well understood (Calvert, 1950; Havron, 1962; Warren and Owen, 1982; Mulder, et al. 2000), and there is still much to be learned about visual perception in general. While our subjective experience leads us to believe that our brain has access to a perfect, high-definition image of the outside world, Smallman and St. John (2005) point out that this “naïve realism” is simply not consistent with what is known about the human visual system. In actuality, visual perception is “sparse and sewn together” and assumptions must be made to simplify complex scenes that “distort interpretation and result in imperfect, just-in-time, just-good-enough approximations of reality” (p. 8). Perception can be thought of as a series of educated guesses regarding the outside world rather than the detection of what is there with 100 percent accuracy and certainty. The brain sometimes guesses wrong, resulting in visual illusions. Small and St. John summarized these points nicely:
The illusion of objectivity is that the ubiquity of these errors goes unobserved, thereby fostering and maintaining naïve realism. The brain is a master at concealing its tricks, and only occasionally does one get to glimpse the real Wizard of Oz behind the curtain. (p. 9)
This lack of veridicality makes the challenge of understanding visual perception an incredibly complex endeavor. Because perception does not involve a perfect 1:1 representation of the external environment, we cannot understand how vision works simply by investigating the basic physical characteristics of the world. Knowing that a runway is 3,000 m long × 50 m wide tells us little about how a pilot will perceive this object, because the visual system does not even use these units. Thus, while the physiology of the eye and the study of optics are straightforward and well understood, higher-level visual processes such as understanding how a three-dimensional (3D) world is recreated from a two-dimensional (2D) retinal image are still not completely understood (Smallman and St. John, 2005).
A further challenge to understanding the role of vision in aviation arises from the fact that the pilot is a part of a complex human-machine-environment system (illustrated in Figure 1.1), in which there are numerous, complex interactions between the factors of the environment, the pilot, and the aircraft. Environmental factors such as weather, time of day, geographic location, and G-forces can directly alter a pilot’s perception of the world (e.g., by reducing visibility), and/or change the way in which visual cues are used to control the aircraft (e.g., by altering the amount airspeed changes for a given stick movement). The pilot adds both capabilities and potential liabilities into the system. The pilot’s skill, proficiency, training, and experience are all factors that increase overall system success. However, if a pilot is not prepared for a particular flight (e.g., is lacking in motivation, personal life stressors or fatigued), the environment can quickly expose a pilot’s vulnerabilities. The aircraft in the system has certain characteristics and limitations given its particular design, including properties such as aircraft manuals, procedures, the level of automation, displays, controls, warnings and alarm systems, redundancy in emergency systems, and the overall interface design.
Figure 1.1 Environment, pilot, and aircraft interaction
Chapter 6 presents a detailed summary of visual perception illusions and spatial disorientations. Aviation accident examples involving visual misperception are given in Chapter 7. To help introduce misperception in aviation, below is a fairly recent mishap that highlights several issues within aviation visual perception: the misperception of height and distance, the meteorological limitation of night (lack of visual cues), and the physiological limitation of a color deficiency. Increased awareness of both the visual phenomenon of misperception of height and distance as well as night flying hazards are major themes of this book.
July 26, 2002, prior to sunrise, a Boeing 727, operated by a commercial freight carrier, struck trees during a short final approach and crashed 472 m (1,550 ft) short of runway 09 at Tallahassee Regional Airport (National Transportation Safety Board [NTSB] report, 2002). Three crew members were seriously injured and the airplane was destroyed. Although there were many interesting aspects of this particular mishap, detail will only be presented on aspects of the visual factors contributing to the accident.
The flight had departed Memphis, TN, for Tallahassee and operated on an instrument flight plan. The forecast weather for the arrival destination was night visual meteorological conditions. The crew had debated whether to land on runway 27 with an Instrument Landing System (ILS, precision glide-path approach) or the more conveniently aligned visual approach to runway 09, and had decided to use runway 09. Runway 09 did have Precision Approach Path Indicator (called PAPIs) lights available to assist in glide-path control. As the pilot maneuvered the airplane into alignment with the runway, the descent rate of the aircraft increased beyond the 3-degree desired glide-path. According to the Flight Safety Foundation report (2005), the profile view of the approach had the concave shape characteristic of the black-hole illusion as illustrated in Figure 1.2. The PAPI lights are also depicted in Figure 1.2, showing how they visually inform pilots of their glide-path.
Although a more detailed discussion of the black-hole illusion occurs in the chapter on visual illusions and misperception (Chapter 6), to better understand this mishap it is briefly explained. When a pilot approaches a runway that lacks terrain features and other ambient visual cues during a dark night, the only visual referent is the lighted runway shape. This approach-and-landing environment makes it very difficult for a pilot to estimate height above and distance to the runway. Due to the lack of terrain features the pilot perceives the plane to be higher and farther from the runway than it actually is and, consequently, initiates an unwarranted descent below the normal glide-path. In this scenario a pilot will realize far too late that the plane is on an extremely shallow approach angle to the runway and dangerously low; controlled flight into terrain often results. Other phrases used are “landing short” or “under-shooting” the runway. The profile view of this type of approach glide-path has a concave shape due to the excessive descent rate that then shallows out as the pilot approaches the landing runway (shown in Figure 1.3).
Figure 1.2 Mishap glide-path
Source: From NTSB report, 2002.
In the case of the accident in Florida, upon examining the profile of the aircraft’s descent, one could argue the black-hole illusion caused the pilot to misperceive the glide-slope starting at 10.2 km (5.5 nautical miles) from the runway until impact. The final approach to runway 09 required the aircraft to fly over a national forest area which had no lights or terrain features (FSF report, 2005). Prior to impact the airplane was flying at 270 km/hr (146 knots) airspeed with a descent rate of 161 m/min (528 ft/min), but 20 seconds earlier it had a descent rate of 380 m/min (1,248 ft/min), nearly twice what it should have been.
The PAPI lights for the approach (shown in Figure 1.2) signaled “below glide-path” from a point 8.3 km (4.5 nm) from the runway to “well below glide-path” at the 5.6 km (3 nm) point. Procedurally, any indication of a “too-steep” glide-path should be immediately followed with a positive correction. All crew members stated they were shocked upon hitting the ground (NTSB report, 2002). Despite the PAPI indications, none of the pilots perceived their glide-path to be below normal and had not imagined the accident that was about to occur. Figure 1.4 provides a better description of how the lights inform the pilots of their position relative to the desired 3-degree glide-path.
Figure 1.3 Black hole illusions
Source: From Gibb, 2007.
Figure 1.4 Precision approach path indicator lights
The colored lights were developed to assist pilots regarding their glide-path to the runway. Granted, each pilot may have his own perceptual interpretation of the lights and manner by which he controls the aircraft relative to the lights; regardless however, all white indicate an approach that is too steep and all red signals a flight path that is dangerously low regardless of technique or aircraft. Consequently, immediate and appropriate control inputs are needed by the pilots in these extreme situations.
To better understand the accident sequence, the description of events will begin at approximately 5:13 AM local time, about 24 minutes prior to the aircraft landing short. At that time the flight engineer, after coordinating with the airfield for their parking plans upon landing, briefed both the captain and first officer, as required by the commercial freight carrier’s procedures, that Tallahassee was a “moderate” risk for controlled flight into terrain. What follows are the voice cockpit recordings and summaries of their conversations up to the point of terrain impact taken from the 2002 NTSB report.
At 5:16 AM local time, the first officer (right seater or copilot), who was going to fly the approach and landing, thoroughly briefed the other two crew members on the approach into runway 27 and the captain concurred with the briefing details.
At 5:19 AM, the first officer, although having just briefed an approach and landing to runway 27, suggested to the captain runway 09 instead of 27, being that they were more conveniently aligned to land on runway 09. Runway 27 required a longer flight path to the far side of the airfield to get aligned for landing compared with already being somewhat aligned to land on runway 09. Their flight path was coming from the northwest and heading towards the east, thus landing runway 27 would require flying east for some time to then turn back to the west and then land heading 270 degrees. Because it was prior to sunrise with no traffic conflicts, landing to the east was more convenient given their position.
At 5:24:03 AM, the controller advised the crew to expect a visual approach into Tallahassee and to report when they had the airport in sight.
At 5:24:23 AM, the captain queried, “Runway nine … PAPI on the left side … I don’t know, you wanna try for nine?” The first officer responded, “We’re pointed in the right direction, I don’t know, like you said. Kinda long … taxiback.” The runway debate continued with the captain saying, “… the only advantage you have, landing to the west you have the … glide-slope … which you don’t have to the east.” (The decision to not land on runway 27 with the glide-slope, needless to say, was a key link in the chain of events leading to the mishap.)
At 5:26:41 AM, the captain asked, “You familiar with the airport here at Tallahassee?” to which to first officer replied, “No, I’m not.”
At 5:28 AM, the crew finally decided upon runway 09 for their landing.
At 5:30 AM, the first officer reported to the crew that he had the airport’s beacon in sight. The light he saw, however, was that of a power plant. According to the NTSB report, pilots flying in from the northwest direction often misperceive the power plant’s light to be that of the airport. The captain corrected the first officer. The crew then configured the aircraft for landing, accomplished the before-landing check, and the remaining transcript contains the pilots discussing their approach.
At approximately 5:35 AM, the landing gear was extended and the “before-landing check” was initiated.
At 5:36:20 AM, the first officer apologized for his final approach and said, “Sorry, ‘bout that … I was lining up on the paper mill or something.” At about the same time the aircraft’s ground proximity warning system sounded an alert, announcing passing through 305 m (1,000 ft) above the ground. With that warning and then in response to the first officer’s apology, the captain responded, “That’s all right, no problem.”
At 5:36:37 AM, the aircraft was 4.6 km (2.5 nautical miles) from the runway and correcting to final. The investigative analysis determined at this time the aircraft’s spatial position relative the PAPIs was three red lights and one white light. This is slightly low in regard to glide-path, or in other words, given their distance from the runway, their vertical position (altitude) was just below of the desired 3-degree glide-path.
At 5:36:40 AM, the PAPI would have showed all four lights as red. This indication is too low or well-below the desired 3-degree glide-path.
AT 5:36:43 AM, the ground proximity warning system sounded an altitude alert; the captain replied, “Stable.” The safety board determined at this time the aircraft was 152 m (500 ft) above the ground and 3.3 km (1.8 nautical miles) from the runway. Their vertical descent speed was 380 m/minute (1,248 feet per minute). The aircraft’s glide-path at this point in the visual approach was acceptable (just slight low), approximately 2.6 degrees to the runway. The problem was that their descent rate was twice what it ought to have been and they had already gone so far below the desired 3-degree glide-path so that, although their current 2.6 degree glide-path was no longer excessive, an immediate level-off or climb was needed to correct and avoid a hazardous situation. Due to the low altitude of their approach, their problem was exacerbated by the elevated terrain prior to the runway.
At 5:36:49 AM, the first officer said, “I’m going to have to stay just a little bit higher, or I’m going to lose the end of the runway.” Power was slightly increased and the rate of descent decreased from 427 m/minute to 293 m/minute (1,400 ft/min to 960 ft/min); however, they remained below glide-path.
At 5:36:56 AM, the flight engineer, finishing the before-landing check, asked about landing clearance; the captain replied, “Clear to land runway … nine.”
The captain stated at 5:37:09 AM, “It’s startin’ to disappear a little bit in there, isn’t it?” It is noteworthy that during the visual approach and while the discussion centered on losing sight of the runway, no discussion was made regarding the need to accomplish a go-around.
At 5:37:13 AM, the flight engineer announced the completion of the before-landing checklist. This was the last recorded communication on the voice cockpit recorder.
At 5:37:14 AM, the ground proximity warning system sounded a 30 m (100 ft) above the ground altitude alert. According to the flight-data recorder, the aircraft was 1.3 km (0.7 nautical miles) from the runway. This equated to approximately 1.3-degree glide-slope to the runway, which was dangerously shallow, not accounting for the higher than runway terrain between the aircraft and the runway. Thus, the 1.3-degree glide-path combined with the high terrain made it certain that terrain impact prior to the runway would occur. It is unclear why the pilots did not initiate a go-around at this point. Perhaps they could not see the terrain due to the dark night as well as confusion in terms of a mismatch of where they thought they were on the visual approach contradicting with the audible warning sound of their low altitude.
At 5:37:19 AM, the ground proximity warning altitude alert announced descending through 15 m (50 ft) above the ground. Then at 5:37:20 AM, the cockpit voice recorder had the sound of a “crunch” (most likely due to hitting trees) as the aircraft passed through 12 m (40 ft) above the ground.
At 5:37:26 AM, the data recording ended due to the crash; approximately 13 seconds of silence marked the final stages of the approach to landing. Roughly 74 minutes prior to sunrise and resting some 472 m (1,556 ft) from runway 09, the aircraft came to a burning stop. It is noteworthy that in the final seconds of the flight no conversation took place and no other comments were recorded since the “cleared to land … runway nine” by the captain. This is another indication of confusion by the flight crew because when things get busy or pilots get task-saturated, they often stop talking. In this case the silence may actually have said a lot.
The captain was 55 years old with between 13,000 and 14,000 flying hours. Records indicated that he had not received the company’s required “black-hole” and “controlled flight into terrain” training that took place in 1995 and 1999. The captain also had not gotten much sleep in the time prior to the flight. The captain monitoring the approach never challenged the pilot flying regarding his descent rate and/or his spatial position relative the PAPIs. Also of note was that the first officer readily admitted he was not familiar with the airport and then followed up with the visual confusion of perceiving the power plant’s lights to be that of the airports. Further, according to the NTSB report, the captain did not recall seeing any form of a “low” PAPI warning. Passing through 244 m (800 ft) above the ground he recalled perceiving a “white red” PAPI … which is equivalent to being “on” glide-slope (about 3 degrees). About five weeks prior to this flight, the captain had his flying medical exam and the only limitation was that he was required to wear corrective contact lenses when he flew.
The first officer was 44 years old and had between 7,500 and 8,500 total flying hours. He admitted to having trouble adjusting to the night-flying cycle that his flight schedule required, but he had received company’s “black-hole” training in 1999. In the post-crash interview, the first officer did not recall seeing any “low” indications on the PAPIs.
The flight engineer was 33 years old and totaled approximately 2,600 flying hours. Like the captain, he had never received the “black hole” training. Similar to the captain and first officer, the flight engineer in his NTSB post-accident interview stated when he first saw the runway, the PAPI lights were a white, a pink, and two red lights; and the runway was in clear sight, “plain as day.” The flight engineer’s interpretation of the “crunch” of the trees was that they had encountered some turbulence.
As alluded to previously, the final 13 seconds of the cockpit voice recorder was silent, implying that the pilot flying, the first officer, was not the only one that perceived the visually controlled night approach as perfectly safe. Given some of the discussion centered on momentarily losing sight of the runway it is hard to imagine that they were so confident to fly a night visual approach without being cognitively primed to accomplish a go-around. The silence by the more senior pilot, the captain, approved the manner in which the first officer was flying the approach. The NTSB report stated (p. 55):
The Safety Board concludes that the approach to runway 9 at TLS [Tallahassee] (which was flown over unlighted terrain and in night visual conditions) resulted in black hole conditions, which likely contributed to the flight crew’s failure to properly perform the approach. However, the Safety Board also concludes that PAPI lights, such as those installed at runway 9 at TLH, are a recognized countermeasure for use in black hole conditions and should have been, but were not effectively used to maintain an appropriate glidepath by the first officer (who was the flying pilot) or by the captain and flight engineer (who, under the principles of basic crew coordination, were in a position to receive this information and initiate a corrective response).
The confusing aspect of this portion of the mishap description/assessment is how the pilot monitoring the approach, the captain, medically cleared to fly by all standards, also did not perceive the need to amend their approach angle to the runway. In terms of cockpit dynamics, the captain is ultimately “in charge” and responsible for the aircraft and should not hesitate to correct the first officer’s flying. In fact, research has shown that when the captain is flying, the lesser ranking, non-flying pilot at times fails to correct the more senior pilot but usually not the other way around (NTSB report, 1994). The recommendation in this 1994 report suggested having the captain monitor challenging approaches—as was the case in the present accident. The captain, however, was fully aware that the first officer was unfamiliar with the Tallahassee runway environment. Despite these facts, the captain never challenged the first officer’s flying of the approach to landing, implying that from the captain’s perspective the approach was being flown safely in terms of visually guided aircraft control inputs as well as the interpretation of the PAPIs.
The NTSB report (2002) further presented counter-arguments regarding the handling of the aircraft by the first officer. In attempting to de-conflict interpretations of pilot perceptions and aircraft inputs, the following was presented:
It is possible that the first officer interpreted the uniform PAPI light indications as “white” because that was consistent with available visual indications (for example, the black hole illusion and the slight runway upgrade) that would lead him to perceive that the airplane was higher on the approach than it was. Such interpretations would be consistent with the first officer’s conduct of earlier portions of the approach, with occasionally excessive rates of descent and lower-than-normal engine power settings. However, just after the airplane descended through 500 feet agl [above ground level], the first officer stated, “I’m gonna have to stay just a little bit higher, (or) I’m gonna lose the end of the runway.” About this time, the FDR [Flight Data Recorder] data indicated that the airplane’s descent rate began to decrease from about 1,400 to 900, then to 500 fpm [foot per minute]. It was not clear exactly why the first officer moderated the descent rate at this time; however, it is possible that he was trying to reconcile a conflict between the 500-foot GPWS [Ground Proximity Warning System] callout and a mistaken illusion of the airplane’s elevation above the field. (pp. 61–62)
As had been mentioned previously, had the first officer flying, or the captain monitoring the approach, recognized four red PAPI lights, an aggressive go-around would/should have been initiated. Also, as presented in later chapters, at this point in a black-hole illusion, the runway begins to appear flat. The concave approach shape brings the pilot in rather shallow, consistent with the decreased descent rate. In retrospect, it was odd that the captain (the pilot not flying with any color vision deficiency) failed to notice the four red PAPI lights or that neither pilot noticed the flattening of the runway. Either visual cue should have been salient enough to prompt an immediate go-around; however, the pilots failed to perceive that their current position relative to the runway and terrain was dangerous, warranting a go-around. This inaction may be the result of cognitive overload in attempting to understand their false perception and the unfolding reality of their situation.
Mishaps in general rarely are the result of a single cause. Aviation accidents especially come about due to a complex temporal sequence of organizational, supervisory, and individual contributing factors. The accident discussed here in Chapter 1, as well as those presented later in Chapter 7, is no exception. Multiple factors contributed to the eventual landing short mishap. In this particular situation, fatigue and poor crew resource management (lack of monitoring flying performance) also contributed to the accident, per the NTSB report. More relevant to the topic of this book, the investigators unveiled a visual perception factor in addition to the black-hole illusion and deemed it also as contributing to the accident. The formal report found that the first officer had a color vision deficiency which prevented him from properly interpreting the color changes of the PAPI lights from white to red. This finding alone spurred a new national and international interest in color vision assessment in both the civil aviation and military aviation communities.
The NTSB report stated that the first officer’s vision throughout his 16-year naval aviation career was reported as 20/20 in terms of Snellen acuity. Also, his color vision was assessed annually and that he passed the Farnsworth Lantern test, which is the US Navy’s standard color vision screening test. The first officer passed the test 13 times, 10 with a perfect 9/9 score, twice with no documentation, and once with “passed-by history” (p. 30). According to the NTSB report, the first officer reported he had no identified color-vision deficiencies while flying for the Navy. In 1995, however, the first officer had failed a color-vision assessment, one using Pseudo-Isochromatic Plates (referred to as PIP). These plates consist of colored dots and embedded within the dots are more colored dots forming a number. The individual taking the assessment must identify the colored number embedded within the colored dots. If a color vision deficiency exists, the individual fails to sense/perceive (detect/recognize) the number. (See Chapter 3 for more detailed information on color vision testing.)
The first officer had indeed failed the PIP test, but despite the identified “mild red-green defect,” he was issued a flying certificate and a Statement of Demonstrated Ability based on his past operational experience/capability (NTSB report, 2002). After the 2002 mishap, a color vision examination of the first officer was conducted by the US Air Force Aerospace Medicine specialists. Although he again passed the Farnsworth Lantern test, he proceeded to fail seven other red/green color vision assessments. It was concluded that the first officer had a “severe congenital deuteranomaly” that could result in difficulties differentiating reds, greens, and whites (NTSB report, p. 61). Deuteranomaly is a common form of color deficiency, affecting 4.6 percent of males and stands for “green weak”, meaning that a person requires more green stimulation than normal to acquire a color match (Tredici and Ivan, 2008). The Young-Helmhotlz theory of color vision describes color as a perceptual experience defined by the individual based on a combination of the three different cone types in the retina. (The topic of color vision is addressed in more detail in Chapters 2 and 3.) The NTSB report cited a letter from the US Air Force Aerospace Medicine specialists (p. 61):
We believe that he [first officer] would definitely have had problems discriminating PAPIs … because the red light would appear not to be red at all, but more indistinguishable from white than red … it would be extremely unlikely that he would be capable of seeing even the color pink on the PAPI … more likely a combination of whites and yellows and perhaps, not even that difference.
The NTSB report continued in reference to the USAF Aerospace Medicine specialists in describing that the first officer had successfully perceived his aviation world based on other visual strategies that, up until this mishap, had worked. For instance, brightness, location, color shades all may have helped differentiate color perception. The NTSB report continued examination of the role of the color vision and its role in this mishap (p. 61).
However, during the approach to runway 9 at TLH [Tallahassee], the first officer had to rely more heavily on his color vision because the PAPI lights were the only reliable source of glide-path information in the black-hole approach environment leading to runway 9. The first officer’s interpretation of the PAPI lights would have been even more challenging because all four lights were red during most of the final approach. As a result, there would have been no differing levels of brightness for the first officer to perceive across the lights (as might have been apparent if both white and red PAPI lights were visible), nor would there have been a change in brightness to observe (as there might have been when a PAPI light transitioned from white to red during the descent). Either of these would likely have assisted the first officer’s color interpretation.
Although one could argue the major or minor role that color vision deficiency played as a contributor in this specific mishap, there is no argument that this accident highlighted the problems of color vision assessment. This accident spurred research regarding color vision and color vision tests for pilots as well as attempts to standardize assessments between the Federal Aviation Administration, the US Army, the US Navy, the US Air Force, and international flying organizations. The NTSB issued a 2004 Safety Recommendation to follow up this mishap and further emphasized the role of color vision assessment problems in aviation. In that report the NTSB cited an Australian study that examined color vision in pilots with a similar deficiency to the first officer in this mishap. This study found that approximately 29 percent such pilots mis-identified a red light signal with a white light signal. The NTSB Safety Recommendation concluded with a call for research into color vision assessment to include color differentiation tests in time-critical situations and mild hypoxic conditions.
The accident investigation (NTSB report, 2002) concluded that the probable cause of the accident was “failure to establish and maintain proper glide-path during the night visual approach to landing” (p. 68). Thus, key aspects of this mishap were night conditions, visual misperception, a black-hole approach environment, and color vision—all applicable to this book.
The human visual system is a phenomenal combination of physiological and psychological processes that allow us to interact with our environment. Because of our abilities to sense and perceive, detect and discriminate, and recognize and react, we more often than not are able to successfully negotiate dynamic and challenging activities. Acts of human expertise in aviation are difficult to quantify and thus are rarely reported and researched. Every pilot has had situations that almost resulted in tragic or destructive consequences; however, due to quick reactions a disaster or an incident was avoided. These “almost accidents” are rarely documented; no headline news reports were made of them. As far as the organizations (e.g., military, commercial airline company) are concerned, they operate safely based upon the lack of any known problems (e.g., mishaps)—but are they truly safe or just lucky? Reason (2008, p. 265) said it best: “safety is a term defined more by its absence than its presence.” Quantifying safety is difficult. Of course just because accidents are not occurring does not necessarily equate to safe operations; articulated best by Dekker (2005, p. 26), “past success does not guarantee future safety.”
We use examples of aviation spatial-disorientation mishaps to demonstrate that what killed pilots and destroyed airplanes decades ago is still occurring today. Respect and awareness of human limitations in terms of spatial orientation has not significantly improved over the years. Consequently, our intent is to inform the pilot or the aviation researcher of the need to be significantly more aware of the risk in trusting visual perceptual systems. Our depiction of research data and accidents spanning decades is to show clearly that the aviation community has not made sufficient progress, and consequently lives and resources are still lost due to the contributing factor of visual misperception. Occasionally, however, an incident occurs in which it is obvious that all the aviation training, education, skill, and proficiency came to fruition, allowing human capability to outshine any human limitations. An incident occurs that the aviation community can take pride in regarding their proactive safety efforts. An incident occurs that despite all odds are against success safety clearly was present.
While nearing the completion of writing this book a heroic event occurred in aviation that few will forget. On 15 January 2009, the crash-landing and successful ditching of US Airways Flight 1549 in the Hudson River amazed the world. The pilots and crew of Flight 1549 did a superb job in handling the emergency and controlling the evacuation of the aircraft as water filled the cabin (Figure 1.5). Even though the final NTSB report has not been released, enough information has been reported through NTSB preliminary reports that allow for a visual perception discussion.
Flight 1549 departed LaGuardia International Airport, New York, bound for North Carolina. The airplane was an Airbus A320 carrying 150 passengers and 5 crew members (NTSB Preliminary report, 2009). It was estimated that as the aircraft climbed through 900 m (3,200 ft) both engines ingested Canadian Geese, rendering them no longer able to produce the required thrust for flight. Bird strikes are common in aviation and migratory birds are one of the most dangerous hazards that airports try to mitigate. Large flocks of birds are often observed on radar screens by traffic controllers; however, in terms of visual perception, it is very difficult to perceive a bird and react prior to it striking the aircraft. Even if the aircrew of Flight 1549 had detected the birds in time, an aircraft the size of the Airbus would have had difficulty maneuvering to avoid the birds. Given the number of birds, regardless of Flight 1549’s maneuvering, there was a good chance they would have had a bird strike somewhere on the aircraft. Thus, in terms of visual perception, the detection of a hazard (birds) while flying is often left to chance, and if birds are sensed and perceived they are very difficult to evade. One pilot technique is to initiate some type of pull-up maneuver as birds normally dive below and away from an aircraft.
Figure 1.5 Hudson River ditching
Source: Permission granted, photographer Greg Lam Pak Ng, Flickr.com.
Captain Sullenberger, the pilot in command of Flight 1549, is a 1973 graduate of the Air Force Academy. In Checkpoints, a magazine published by the US Air Force Academy Association of Graduates, an article summarized an interview that he gave for a television news program. When asked about the actual bird strike, Captain Sullenberger responded, “… about 90 seconds after takeoff I noticed there were birds filling the entire windscreen—from top to bottom, left to right—large birds, too close to avoid. It felt like the airplane was being pelted by heavy rain or hail” (2009, p. 18). Thus, in terms of detection, sensing, and perceiving the birds, the pilot flying Flight 1549 had no visual indication of a threat until the birds were within very close range because of the small size of the birds relative to the backdrop of the sky and the closure rate between the Airbus and the birds. Visually detecting such a small retinal image is a physiological limitation at those speeds.
The second aspect of visual perception involved in Flight 1549 was the accomplishment of the ditching maneuver, an emergency water landing. Every pilot in emergency training has read about ditching an aircraft with the realization that the odds of that happening being extremely remote. Aircraft emergency procedures discuss at length how to safely ditch, but pilots realize that the engineers designing the aircraft really don’t know what is going to happen to the aircraft upon water impact.
Configuration and airspeed are the two primary aspects for a ditching maneuver. Normally for a given weight, the airspeed of an aircraft prior to “landing” on water is 10 knots below normal landing speed. For an aircraft right after initial takeoff, landing speed would be very high. So, although the target speed is 10 knots slower than a normal approach, such a landing would still be quite fast. At high speeds, landing on water is still similar to concrete, hence the need for a “soft” water landing.
On that day the winds were calm and the water was fairly smooth (as shown in Figure 1.5), the pilot of Flight 1549 was able to accomplish a perfect ditching maneuver and gently settle the aircraft onto the water during the flare. The pilot had to smoothly put the aircraft into the Hudson River in order to avoid excessive structural damage that could have torn the fuselage of the aircraft. A real danger in a ditching maneuver is catching a wing-tip and throwing the aircraft into a tumble in the water. Any of these events would have most certainly resulted in significant loss of life. But not a single life was lost. Captain Sullenberger explained his landing:
I needed to touch down with the wings exactly level. I needed to touch down with the nose slightly up. I needed to touch down at a descent-rate that was survivable. And I needed to touch down just above our minimum flying speed but not below it. And I needed to make all these things happen simultaneously. (Checkpoints, 2009, p. 19)
In many cases, a smooth water surface for “landing” causes perceptual errors due to a lack of features for height and distance perception. Specifically, an open body of water provides few referents for global and local visual cues regarding the aircraft’s glide-path toward impact. Fortunately, the Hudson River and its shoreline provided plenty of ambient cues regarding height above the water’s surface to the pilot of Flight 1549. Additionally, the calm winds and smooth water surface benefited the ditching maneuver (water impact). Had the ditching maneuver occurred on smooth water in a large open area, like an ocean or large lake, height and distance perception may not have been as accurate as they were on the Hudson River.
Based on Captain Sullenberger’s explanation of the ditching, his thoughts during the last few seconds of flight were on a smooth descent for the slowest possible water impact while maintaining a wings-level attitude. To accomplish a safe water landing the pilot’s ambient vision system was unconsciously computing horizon perspective (to control wings level and bank angle) and sink rate (to provide input to the pilot’s control surface movements). He may have been using focal vision (conscious attention) to perceive airspeed readings and radar altimeter information or to determine where on the water ahead the airplane would splash down. The ambient visual system, however, provided the unconscious but critical environmental inputs for global perception as the aircraft glided into the water, leading to a success story for aviation and for the human visual perception system’s capabilities.
To understand the raw visual perception mechanisms at work on this January day, it is worth emphasizing that the pilots accomplished this maneuver with no instrument approach procedures to help guide them. There was no runway outline or runway shape to assist in visual glide-path guidance. There were no approach lighting systems or PAPIs; all that was available to the pilots were the environmental cues of the water surface (which were not much) and the shoreline.
Previously it was mentioned that documenting expert performance is difficult and not often discussed in research. One author has tackled the topic and his work outlines many parallels to the Hudson River ditching incident. In Reason’s (2008) book, The Human Contribution, the author articulated factors for heroic human action in high risk activities; that is, aspects of the situation and of the individuals involved and how they successfully overcame adversity. He examined how many total disasters have been avoided due to phenomenal actions by operators within complex systems. All of the attributes of such incidents described by Reason apply to Flight 1549, and those were training, discipline, and leadership; sheer unadulterated professionalism; luck and skill; and inspired improvisation. Reason addressed how humans have to learn and train to cope with both expected and unexpected scenarios. For instance, pilots routinely practice and rehearse takeoff emergency actions. Every pilot prior to releasing brakes for takeoff should have an emergency return scenario in mind if an engine fails, an unlikely but possible scenario.
The pilots of Flight 1549 had minimal time to assess the situation and take appropriate action. The water landing choice, however, came about due to limited options and a quick risk assessment of any possible alternatives other than the Hudson River. On another day with the same crew or a different crew, the Hudson River option might not have been as successful. If it had happened at night, the pilots would have had significantly increased difficulty in perceiving their height above the water and the ditching maneuver could have been tragic. So yes, luck played a role; however, the pilot, given the situation and the environment, had the appropriate environmental cues for veridical perception to accomplish the ditching. Often “luck” only comes to those that are prepared. And without a doubt, the entire crew, and especially the pilot at the controls, had Reason’s “sheer unadulterated professionalism.”
We strongly believe that spatial disorientation’s contribution to aviation mishaps can be significantly decreased through awareness and through education utilizing research-based operational/realistic training scenarios. Spatial disorientation is attributed to nearly one quarter of all mishaps in military aviation; thus it is greatly beneficial to chip away at this source of accident factors. The US Naval Aviation Safety Center (2009) reported that between the years 1990 and 2008 the number one human factor involved with nearly 80 Class A mishaps (those accidents involving a fatality, destroyed aircraft, or >$1M in damage) was spatial disorientation. While people often associate spatial disorientation with vestibular processing misperceptions, there are actually substantial visual perception influences in spatial disorientation. The role of visual spatial disorientation happens more often than expected and is greatly underreported as a contributor to mishaps. Veronneau and Evans (2004) summarized the main objective of this text while addressing their own thoughts on spatial disorientation:
Successful SD (spatial disorientation) countermeasures will impact all types of aviation operations, including those in the civil community, as there will always be an essentially universal susceptibility of human pilots to SD illusions. To achieve measureable success in improving the overall aviation safety climate, such as the FAA’s goal of an 80 percent reduction in the commercial accident rate, then many of the so-called minor contributors to aviation risk must be addressed. Spatial disorientation is one such factor that can be clearly identified and specifically addressed. (p. 220.)
Indeed the problems of aviation visual perception are not going to go away. In Wingman, the US Air Force’s safety magazine, an article by Sabric (2009) shared a very recent F-16 near-mishap story. The instructor pilot concluded that, “the student had convinced himself that the access road was Runway 03L, and instead of trusting his instruments, he trusted his visual perception” (p. 29). Interesting choice of words by the F-16 pilot, but fitting and accurate.
Throughout the chapters of this book it will become apparent to the reader that all pilots are susceptible to spatial disorientation and that mishap reports regarding the incidence of spatial disorientation are underreported. Even less reported and less respected in accident investigation reports is the role of visual misperception as a contributor to spatial disorientation and the mishap sequence. This book intends to highlight visual spatial disorientation to academics, researchers, accident investigators and to the pilot community. As Veronneau and Evans (2004) advocated, the potential of spatial disorientation awareness/education for reducing mishap rates is significant; therefore we also promote this area ripe for aviation research and safety enhancement. By bringing forth both research and historical mishap examples, visual misperception’s presence is undoubtedly demonstrated. It is not simply “pilot error” but a human perceptual limitation that needs to be better accounted for in aviation education, training, and technological advances as well as in accident investigation.
It is our belief that understanding the role of vision in aviation requires an interdisciplinary approach. Our main goal in writing this book is to provide a comprehensive, single-source document encompassing all the aspects of aviation visual perception. Thus this book includes the foundations of visual and vestibular sensation and perception, and how visual perceptual abilities are assessed in pilots (Chapters 2 and 3), the pilot’s perspective of visual flying (Chapter 4), a summary of human factors research on the visual guidance of flying (Chapter 5), examples of specific visual and vestibular illusions and misperceptions (Chapter 6), mishap analyses from military, commercial, and general aviation (Chapter 7), and, finally, how knowledge from these other disciplines is being used to create the next generation of aviation visual perception (Chapter 8).
Benson, A. J. (1988). Spatial disorientation—general aspects. In J. Ernsting and P. King (Eds.), Aviation Medicine (pp. 277–96). London: Butterworths.
Calvert, E. S. (1950). Visual aids for landing in bad visibility with particular reference to the transition from instrument to visual flight. Transactions of the Illumination Engineering Society of London, 6, 183–219.
Checkpoints Editorial Staff (2009, March). Service Before Self: Sullenberger, ’73, saves the lives of 155 people in a watery crash landing. Checkpoints, 18–19.
Cocquyt, P. (1953). Sensory illusions. Shell Aviation News, 178, 19–24.
Dekker, S. W. A. (2005). Ten Questions About Human Error. London: Lawrence Erlbaum Associates.
Gibb, R. W. (2007). Visual spatial disorientation: re-visiting the black hole illusion. Aviation, Space, and Environmental Medicine, 8, 801–808.
Havron, M. D. (1962). Information available from natural cues during final approach and landing. Federal Aviation Agency, National Aviation Facilities Experimental Center, HSR-RR-62/3-MK-X.
Kern, T. (2002). Controlling Pilot Error: Approach and Landing. New York: McGraw-Hill.
Kraft, C. L. and Elworth, C. L. (1969, March–April). Night visual approaches. Boeing Airliner, 2–4.
Lessard, C. S. (2000). Spatial disorientation: dealing with aeronautical illusions. IEEE Engineering Medicine and Biology, 19, 25–27.
Mulder, M., Pleijsant, J., van der Vaart, H., and van Wieringen, P. (2000). The effects of pictorial detail on the timing of the landing flare: Results of a visual simulation experiment. The International Journal of Aviation Psychology, 10, 291–315.
National Transportation Safety Board. (1994). Safety Study: A review of flightcrew-involved, major accidents of U.S. air carriers, 1978 through 1990, Department of Commerce, National Technical Information Service, NTSB/SS-94/01.
National Transportation Safety Board. (2002). Collision with trees on final approach: Federal Express Flight 1478, Tallahassee, FL.
National Transportation Safety Board. (2008, June 10). Safety Recommendation.
National Transportation Safety Board. (2009). Preliminary Report, NTSB ID DCA09MA026.
Newman, D. G. (2005, Sep-Oct). Eyeball error. Flight Safety Australia, 27–32.
Newman, D. G. (2007). An overview of spatial disorientation as a factor in aviation accidents and incidents. Australian Government, Australian Transport Safety Bureau, Aviation research and analysis report—B2007/0063.
Ostinga, J., Wolff, M., Newman, D., and White, S. (Nov-Dec, 1999). What killed JFK Jr? Lessons for VFR pilots. Flight Safety Australia, 22–29.
Pitts, D. G. (1967). Visual Illusions and Aircraft Accidents. USAF School of Aerospace Medicine Division, Brooks AFB, TX, Report No. SAM-TR-67–28.
Previc, F. H. (2004). Visual illusions in flight. In F. H. Previc and W. R. Ercoline (Eds.), Spatial Disorientation in Aviation, Progress in Astronautics and Aeronautics, Vol. 203 (pp. 283–322). Reston, VA: AIAA.
Reason, J. (2008). The Human Contribution. Burlington, VT: Ashgate.
Sabric, G. (2009). It’s not over… Wingman: United States Air Force Journal of Aviation, Ground, Space, and Weapons Safety, 2(Fall), 28–29.
Schiff, B. (1990). Visual illusions can spoil your whole day. Flight Safety Foundation Accident Prevention, 47, 1–4.
Schiff, B. (1994, January-March). Black hole approach. Boeing Airliner, 16–20.
Smallman, H. S. and St. John, M. (2005). Naïve realism: Misplaced faith in realistic displays. Ergonomics in Design, Summer, 6–13.
Tredici, T. J. and Ivan, D. J. (2008). Ophthalmology in aerospace medicine. In J. R. Davis, R. Johnson, J. Stepanek, and J.A. Fogarty (Eds.), Fundamentals of Aerospace Medicine, (pp. 349–379). Philadelphia: Lippincott, Williams, and Wilkins.
United States Naval Aviation Safety Center (2009). Raw mishap data provided from the Navy’s safety center.
Veronneau and Evan (2004). Spatial disorientation mishap classification, data, and investigation. In F. H. Previc and W. R. Ercoline (Eds.), Spatial Disorientation in Aviation, Vol. 203, Progress in Astronautics and Aeronautics (pp. 197–241). Reston, VA: American Institute of Aeronautics and Astronautics, Inc.
Warren, R. and Owen, D. H. (1982). Functional optical invariants: A new methodology for aviation research. Aviation, Space, and Environmental Medicine, 53, 977–983.
Wulfeck, J. W., Weisz, A., and Raben, M. W. (1958, November). Vision in Military Aviation. Wright Air Development Center, United States Air Force, Wright-Patterson Air Force Base, OH, TR 58–399.
