It would take a clinical psychologist to explain, but there are some people who are not meant to fly airplanes.
It certainly doesnt take a superman or superwoman to become a competent pilot, of course. Like riding a bicycle, most people can learn to do it competently, with a few becoming exceptionally skillful. The flip side of that is that there are also those who cant get the hang of it.
So it is with flying: There are some people who, for whatever the reason, simply can not put it all together and fly competently. The difference is that in aviation, this apparent learning disability costs lives and money, rather than scrapes and bruises.
My first exposure to this phenomenon was at age 19, while working one summer as line-boy for a WWII GI Bill flight school. A nice looking young man, who was a junior in college, signed up for the private pilot course.
Despite flying the schools J-3 Cubs daily, he had not yet soloed after 20 hours of dual. The assigned flight instructor was exasperated with his students performance, and I overheard him discussing with his boss how to handle the situation.
Simply put, the student did not seem to absorb information. He kept making the same mistakes over and over. Finally, with 23 or so hours of dual, the instructor reluctantly allowed him to solo. But the next day, on his second flight alone, he spun-in on go-around from a touch-and-go landing. Because the Cub was a float plane and he crashed into the lake, he was only bruised and scratched, but his flying career ended abruptly.
The accident files are full of similar stories. Even the military services, with the ultimate in close supervision, have their share of these problems. Its not bad karma that causes one individual to have a rash of incidents and ultimately an accident. During my time in the service, a flying school classmate had one major mishap during pilot training, then three more (all in jet fighters) once he became rated. Mercifully he was finally grounded.
Military Accidents
Several years ago an Air Force KC 135A tanker pilot killed his six-man crew when, upon landing, he rounded out high along the left side of the runway, pulled the power to idle, then dropped it in and bounced down the runway, finally dragging a wing tip.
The airplane eventually burst into flames and exploded. The 10 passengers escaped through a rear hatch, but the crew was asphyxiated by the smoke.
Investigators found the relatively high time pilot (1,300 hours total and 926 hours in the KC 135) had a long record of landing problems. While flying the T-38 trainer for proficiency he was graded unqualified for landings. His instructor noted, His landings are characterized by a sinking final, high flare, and early power reduction. … Characteristically lands left of centerline and many landings were firm (hard) with incomplete flares.
At another base he again had landing problems which his instructors tried to help remedy. The dilemma instructors had was that his bad landings were not an every-time occurrence.
Airline Errors
Despite careful screening of pilot applicants, the airlines also experience this problem. In one recent case a major airline discharged a probationary first officer after 11 months of employment, reportedly because he simply could not fly on instruments. The surprising thing was that he was an ex-Air Force pilot and Academy graduate.
In another case, a relatively new first officer had a history of proficiency problems, particularly with instrument flying, and the aircraft she was flying crashed. While the actual cause of this mishap has been challenged, the NTSB report shows that one rainy night, while on an instrument departure from Raleigh-Durham International Air-ports runway 23R, she was told to report established on 290 degree heading and make that turn as soon as feasible. During the turn she apparently became disoriented and crashed 100 feet from the shoreline of a reservoir. All 12 people aboard the American Eagle flight were killed.
Her background was in light twins and singles. During transition training to the Metro II she required additional dual and three check rides before qualifying as second in command. The first check airman had written, Needs more work on landing, having trouble maintaining glide path and speed control and keeping torque matched on landing. The second check airman wrote, refuses to fly the aircraft … performance unsatisfactory … recommend termination.
He later told the Director Of Operations that to bring her up to standards would take a long time. A third check pilot qualified her as second in command on the Metro II. It took a tragedy such as this to prove which check pilot was right.
Yet another instance involved a Continental Airlines DC-9. It too crashed on take off with the first officer flying. While wing ice was involved, the first officers employment and training record, combined with his take off rotation procedure on the fatal flight, was paramount.
On a previous job with a Part 135 operator he required 30 hours of flight training and still failed the check ride. NTSB investigators found that his initial multi-engine training consisted of only 4.8 hours. His multi-engine check ride lasted only 0.5 hours, with no instrument time logged and only one landing. (The FAA later revoked the examiners authority upon discovering that he had issued a license to another pilot without requiring all the required flight test events.)
The charter companys chief flight instructor noted that the pilot routinely had difficulties with single engine procedures and directional control. He also had the chronic problem during engine-out procedures of stepping on the wrong rudder and becoming disoriented. He was described as tense and unable to cope with deviations from the routine. Once again, the pilot continued to make the same mistakes repeatedly. In essence he couldnt seem to retain information. The lack of training progress and three unsatisfactory check rides ultimately led to his dismissal from the charter operation.
After he was hired by Continental, his learning problems resurfaced. His training records included write-ups such as Scan! Needs to review procedures and profiles. Pitch control jerky, altitude control when pressure is on is somewhat sloppy. On his sixth simulator period his instructor wrote Scan a real problem, completely lost control of airplane with engine out and at 2,000 feet. Went into 45-60 degree bank and lost 1,500 feet. Had to be arrested by (instructor). His sixth simulator session was unsatisfactory and had to be repeated. The instructor still failed to note normal progress. One item mentioned was Scan was a problem during first half of three hour period but improved toward the end.
On his seventh simulator period – with a different instructor – he completed his proficiency check with comments from this instructor of Nice Job! No problems. He then completed a proficiency check in the airplane and his initial operating experience.
After 24 days on reserve with no flying, the new first officer was given a trip to help him maintain proficiency. Departure weather included a 500 foot overcast and two miles visibility in light snow; almost ideal conditions to let a first officer exercise his newly honed skill in the DC-9. Thus, with the first officer flying the first leg of the trip, after a long delay for departure, the flight was cleared for takeoff . But by now the aircraft had accumulated a thin layer of wing ice. During take off roll the captain called V1 and rotate, then positive rate. Suddenly there was an exclamation from one of the pilots followed by sound of engine compressor stall and then impact.
The NTSB concluded, The first officer rotated the airplane at a rate about twice the normal rate, and the captain failed to arrest this rapid rotation. While wing ice contamination was known to produce roll and unexpected pitch-up tendencies, the Board felt the greater than normal pitch rate … was present during initial rotation, indicating it was pilot-induced. The Flight Data Recorder showed a (rotation) trace … about twice what it was on the six previous flights.
It appears that the first officer had reverted to the jerky pitch control noted during his initial simulator training.
The GA Dilemma
An even greater dilemma exists in general aviation. A pilot with a learning deficit-induced proficiency problem cannot easily be thwarted. First, FAR 91 is quite lax for non-commercial pilots. And even if a flight instructor or professional learning center identifies a problem, its ability to prevent the pilot from further flying is limited for several reasons, both economic and human.
First flight instructors realize that pilots are their customers. Their income depends on each pilots happiness with their training. If a customer fails a check ride, the ego injury can cost the instructor a bundle in lost business as the bad-mouth filters through the pilot community. A Cessna 421 accident provides a classic case in point.
The 66-year-old president of a company in a mid-sized southern town purchased the Cessna 421C in October 1995. According to the accident report, he attended an initial simulator-based pilot training program from Nov. 6 to Nov. 10. The course consisted of 20 hours of ground school and 10 hours simulator instruction. His training records showed Unsatisfactory performance.
The simulator instructor noted that the pilots instrument scan was extremely slow and insufficient, to the extent that under IMC conditions he could not maintain altitude within 1,300 feet or heading within 40 degrees when given the simplest tasks. On Nov. 8 the student requested that all remaining time in the simulator be given under VFR conditions with systems orientation. On the last day of training the instructor noted: continued poor aircraft control coupled with limited cockpit scan, awareness, and system retention prevented satisfactory completion of requirements even under VFR conditions.
Further investigation showed that this wasnt just a matter of the pilot having a bad couple of days. The pilot had owned a Cessna 340A previously, and had attended both the initial checkout and a recurrent course in Jan. 1992 and Feb. 1993, respectively.
During the initial course he showed serious weaknesses. On his final grade sheet the instructor wrote, Wonder as a sim instructor how he got an instrument rating or M.E. Flight training completed to VFR standards.
During the recurrent course his learning problem was again noted. On the first simulator session his instructor wrote, He can do only one thing at a time. It seems very difficult for him to fly and tune the radios at the same time. After completing the course his instructor wrote, He met the very minimum standards for VFR operations. Not within IFR standards at all.
Just over a month after failing the pilot initial training in the Cessna 421, the pilot organized a Christmas skiing vacation to Aspen, Colo., for five friends. At a refueling stop in Tulsa, Okla., he checked weather but did not file a flight plan. He was advised of an AIRMET in effect for his route of flight for instrument meteorological conditions and that VFR flight was not recommended from Ponca City through western Kansas. Further, a radar map showed icing conditions northwest of Tulsa.
An employee at the FBO helped the pilot use the weather machine in the flight planning room. As they discussed the weather he seemed nervous, this witness reported. While talking to him I noticed his hands would shake a little, like a much older person. This is an interesting aspect as it shows the pilot was apprehensive about weather conditions.
At 14:24 CST the VFR flight was cleared for takeoff on Tulsas runway 36L. Thirteen minutes later, Tulsa Radar West controller advised the pilot that radar service was terminated. The pilot acknowledged the transmission and was not heard from again. FAA radar data showed the airplane in a climb from 2,500 feet to 9,800 feet. During the next 1 minute 28 seconds the airplane was seen to descend to 5,400 feet then climb to 8,300 feet. It then entered another descent with the last radar hit showing it descending through 3,200 feet and 0.1 nautical miles south of the impact area.
Ground witnesses saw the airplane descend out of the clouds in a flat spin. Verifying the flat spin report, investigators found that the left engine was at low power with the right engine indicating higher power and energy at impact. In addition investigators found that the aircraft, with full fuel tanks and six occupants, was overloaded by 150 pounds.
Weather in the impact area was reported by a witness as 500 feet overcast and good visibility with light snow flurries.
Regarding the pilots nervousness before the flight, the autopsy results are interesting. Toxicological tests were positive for Valium, the antidepressant Trofranil, Benadryl, and Tylenol. As if this pilot did not have enough problems already, the FAAs Civil Aeromedical Institute found the levels of these drugs could have caused impairment of judgment in the cockpit.
Once again a seriously substandard pilot, with a recognized learning and performance deficiency, continued flying. But then realistically how was he to be stopped?
High Performance Failure
One final example involved a Mitsubishi MU-300 Diamond 1A. The commercial pilot was very experienced, with over 17,600 hours. His total time in the MU-300 at the time of the accident was estimated at about 80 hours. Like the other pilots, this pilot had great difficulty absorbing information. Whether it was age-related – the pilot was just shy of his 66th birthday – or just a basic skills problem was never addressed, but aging in some people is known to cause cognition and learning disabilities.
While the accident pilot held a Citation type rating, he had no significant experience in the airplane. However he was flying the King Air 200 prior to attending the MU-300 initial course.
His simulator records show great frustration on the part of his instructor, who made notations such as Cant retain information. Can be told procedures over and over and no help. Client cant remember most basic operations; put N1 on airspeed bug … landing out of control … stalls marginal … busted altitude by 1,000 feet … ILS crashed on landing … Five V1 cuts and five crashes … cant do trim check even reading the procedures. Despite the simplicity of the Diamonds systems he barely passed the academic final exam, making a minimum acceptable score of 70 percent.
Because his instrument flying was unsatisfactory, he asked to be trained and checked in the airplane rather than the simulator; since he felt he could fly it with no problem. It should be noted that simulator flying is 100 percent instruments, while flying the airplane allows the student to see out, especially peripheral vision. Also most training in Transport category airplanes is without use of a view-limiting device, since it is felt that you can not fly accurately by looking out anyway.
After finishing the prescribed syllabus and failing to progress in simulator training, he was given another instructor who was thought to be more pragmatic. After an additional simulator session he too found the pilot unqualified for his Simulator 85% check ride for the type rating.
By mutual agreement the subject pilot left for a period of flying the airplane for the owner, with a type-rated pilot in the right seat. When he returned several days later there was a noticeable improvement in his knowledge base. In his first simulator session since returning he was found qualified in the limited items accomplished. However he remained unqualified in many of the required training events.
After a second training session he was recommended for the 85% simulator check; still without qualifying in all required events. The record shows he passed the simulator check, although the events covered were not graded individually. After a 3.3 hour flight check, he was issued the type rating.
Seven months later he was flying a 69 nm trip from Scott City, Kan., to Goodland, Kan. The weather at Goodland was IFR with a 400 foot broken ceiling and 10 miles visibility. The flight was preparing for an ILS to runway 30. The last recorded radar plot showed the airplane at 9,400 feet, and 16 nm southeast of the runway, tracking two miles east of the ILS centerline. What happens next can only be classified as bizarre, unless you had read this pilots simulator training records.
The flight was cleared for the ILS to Goodlands runway 30. As they headed for the compass locator the cockpit voice recorder recorded sounds of the over-speed warning warbler. The copilot then asked speed brakes? after which the over-speed warning ended. During the next three minutes the flaps were extended 10 degrees and the landing gear lowered. In the cockpit there was conversation about intercepting the ILS localizer, then a call to Goodland Unicom concerning runway conditions, and then more conversation about intercepting the ILS.
Suddenly the copilot exclaimed Thats a shaker. Two seconds later he repeated, Shaker, power, power … Four seconds later the CVR recorded sound of a horn similar to an altitude alert. This was followed by a frantic call from the copilot of Full Power. Then Get your nose down, get your nose down: Lets get it, get it to flying! The recording ended five seconds later.
Several things are worth mentioning regarding the copilot. To save money for the owner he had been trained generally in accordance with FAR 61.55 by the captain. But there were two problems with this. First the captain held a commercial license without a flight instructors rating. This made his flight and ground instruction invalid under the FARs. Second, the training was not documented in any form. Thus his qualifications could not be verified. (Another contract copilot had been trained also by the Captain and told investigators of their effort.) Consequently the flight did not have a legal crew. As a result, the insurance company reportedly refused to pay for the loss.
Despite his lack of proper training, the copilot was the only one functioning correctly, as the Captain reverted to old habits noted in the simulator and proceeded to lose control of the airplane. The copilots apparent failure to take control was no doubt the result of his comparatively low flying time and lack of total self-confidence due to inadequate training and experience. After all, this was his first opportunity to fly a jet airplane.
Whos Teaching Whom?
All of us are human and most of the time try to help our fellow man. This is especially true in the aviation community. But as a flight instructor or check airman, being a good guy can mean untold pain and suffering to others.
My only personal example involved a 32-year-old who had been hired by a truculent old man to fly his Sabreliner. He was an excellent instrument pilot but an obviously flighty or immature personality. In addition, to save on expenses, he flew with an untrained copilot.
After lift off one dark night they flew into Lake Pontchartrain while casually discussing a faulty altimeter, killing all aboard except the newly rated captain. He had told me of three other major accidents in his background, but this one finished his aviation career.
I had been reluctant to address these factors with his employer, which may have cost him his job. Think of the potential liability of costing a person his/her livelihood based on speculation. This is where pre-employment neuropsychological evaluation and information processing tests prove invaluable in both screening potential employees and in making retirement decisions.
While the military and airline crews undoubtedly received pre-employment psychological screening, the GA pilots certainly did not require this for their licenses.
A recent paper by Drs. Diane Damos and Elizabeth Parker, published in NBAA Digest, concluded that both neuropsychological assessments and information processing tests can detect subtle deterioration in cognitive capacity that can be caused by any number of different factors, such as stroke, use of medications, aging, (and presumably heavy alcohol consumption).
The authors found that Information processing tests have good characteristics for repeated testing and can be administered frequently, even on a daily basis without biasing the results. Indeed they observed that the more data available the easier it is to detect subtle changes in cognition.
Some years back the FAA made EKGs mandatory on Class One physicals for applicants over a certain age. This was aimed at preventing sudden in-flight incapacitation. While there are now much more accurate means of predicting sudden heart attack, such as ultrafast computed tomographic scanning (EBCT), the same old obsolete heart test is used.
With this thought in mind, there has been a consistent pattern of subtle mental incapacitation accidents. The mishaps described are but a sampling of case histories. Yet despite the pattern of accidents associated with mental-related dysfunction, there is no FAA requirement for testing in this area.
Both of the two GA pilots were 66 years old and their performance in training clearly identified a serious cognitive problem, even to their laymen instructors. Yet there was no guidance from the FAA. And who in his/her right mind is going to tell a client hes mentally deficient? Meanwhile, in the real world of modern medical science, there is the capability to detect and prevent these kinds of accidents.
As things stand, flight instructors and designated examiners are the last bastion of aviation safety. In the case of the military and airline pilots, their deficiencies had been identified long before, so supervision was lacking. Their records were replete with evidence of a problem. Yet their instructors and supervisors played good guy and kept trying to get them qualified.
The GA pilots were more difficult in that they answered only to themselves (with the exception of the contract MU-300 pilot, who was beholden to the aircraft owner). Their simulator instructors clearly identified the problem. Then the warmth of human kindness was added to the equation (the good-guy approach) and the opportunity to prevent two fatal accidents was lost.
In the Mu-300 case the first simulator instructor clearly and positively identified a serious learning problem. Yet a different instructor, who was known to be more customer-friendly, was brought in to nurse the pilot through to his type rating. The check airman was also taken in by the desire to help this affable old guy get his rating so he could continue to make a living.
Common sense should have led to a discussion of the problem with the MU-300 owner, who after all was paying the training bill. The pity of this accident is that, had the pilot been trained in use of the auto pilot, which his records show was omitted, he could have coupled up at cruise altitude and let the auto pilot fly the entire ILS approach. Then his only function would have been modulating the power levers to maintain airspeed.
In the case of the Cessna 421C pilot, his instructor in the Cessna 340A two years earlier had highlighted the problem in a final summary statement found in the training records Wonder how he got an instrument rating and M.E.? This is where action should have been taken to see that he never flew another airplane unless the problems were solved.
This means reporting the pilot to the insurance company or the FAA, yet consider the business implications of such a move.
Watching the Watchers
Designated pilot examiners and flight instructors hold an important key to flight safety, and their ethical standards will often be tested. What does it say about the ability of the industry to police itself when an examiner is allowed to type-rate a new captain in a Sabreliner in two or three days of training or issue a Citation rating during a weekend of touch and go landings?
You cannot adequately review the Airplane Flight Manual of a Transport category airplane in three days. But some examiners continue to do it, smiling to themselves as they walk away with the $3,000 fee. Not bad for a weekends work
The MU-300 pilot had gotten just such a quick and easy type rating in the Citation from a DPE in Texas some months before his Diamond training. Considering his simulator record in the MU-300, how thorough could his Citation evaluation have been? Yet he was certified as pilot in command.
In a night VMC into IMC accident involving a light twin that killed three, investigators found the accident pilots designated multi-engine examiner had not confirmed the applicants night training. And on a 0.7 hr. check flight he failed to check the pilots instrument flying ability. Considering that the applicant scored only 72 percent on his third take of the written exam, the examiner should have had reason to be thorough. The DPE was found to have given 253 commercial checks and 187 private checks in a two-year period. Only two commercial check rides resulted in pink slips, with 11 private pilot failures.
In the crash of a Cessna 172 during a full flap go-around, investigators found that the private pilot had never been taught the procedure. Nor had he been taught spot landings, short field landings or when to initiate a balked landing approach. The accident pilot told investigators that his private pilot check ride was about 30 minutes. His oral lasted about 25 minutes. On the date of his flight check, the DPE had flown a five-hour Citation trip and then certificated four private pilots. True he made good money that day, but he also set the stage for disaster.
Anyone who observes others flying, whether a CFI, examiner, training pro or flying buddy, needs to be alert for the signs of trouble. You do not need a degree in psychology to spot a serious learning problem. Ethics must predominate to keep people alive.
Its counter-productive to rely on post-crash lawsuits to enhance the safety equation, because the damage is already done. And if your loved ones were aboard, nothing is going to ease the pain.
As my German friends used to say Dont be too late smart.
-by John Lowery
John Lowery, a former Air Force and corporate pilot, is an aviation safety and training consultant.
